return to research page    

Kigo and Seasonal Reference in Haiku:

Observations, Anecdotes and a Translation

Richard Gilbert


Publication: Simply Haiku 3.3 (Autumn 2005)



In this essay I would like to offer some informal and anecdotal observations, and have included a translation of a short discourse on kigo (“season words”) by Tsubouchi Nenten, taken from his book, An Introduction to Haiku, which was written for haiku beginners. Some years previously, I visited my local Kumamoto bookstore to see what haiku instruction books for the general reader might be available, and auspiciously found this slim and accessible volume by one of the great poets and scholars of modern haiku. I am indebted to my wife Keiko and to Shinjuku Rollingstone for their patient hours of co‑translation.


When we look for seasonal reference in English haiku, a non‑season‑specific nature image, such as migrating birds, would likely not meet the definition, as we cannot determine a single season for migration—however, “migrating birds” is an autumn kigo, in Japan. This simple fact offers a first clue that “seasonal reference” in English, and “kigo” in Japan may not rest on the same conceptual basis. A seasonal reference in English, we find, should align with one of the traditional five seasons, including the new year as a season, after the three-days of shōgatsu in Japan. Though our own new year is culturally something of a party night, the spirit of the new year resonates, and has been accepted as a haiku season, following Japanese custom, though this extra season is not included in all seasonal‑haiku collections. Beyond this shared concept of five seasons, there are a number of distinct contrasts between the term “kigo” as currently used in English (aka: “season word,” “seasonal reference”), and kigo as found and practiced in the contemporary and classical Japanese traditions.


Parsing kigo and seasonal reference

The two terms “kigo” and “seasonal reference” have been treated as synonyms in English (see “Confusions, confabulations” below). In this essay, I propose that we consider these terms as disparate entities. The use of synonym is quite understandable, as “kigo” translates literally as “season word(s),” from which the idea of seasonal reference springs—though there is a leap, from “word” to “reference.” The trouble begins when we confuse the idea of “season words” as we have it in English, imagining the context of (Japanese) kigo as virtually identical to ours: that the main and indeed only function of kigo in Japan is likewise to present and delimit seasons—just as in our haiku tradition. But as Tsubouchi mentions (see “Tsubouchi Nenten: Kigo” below), “kigo is a culture.” The conceptual base of kigo is its culture, not its season, I would argue. The culture of kigo is the context in which Japanese kigo arise. This culture is precisely what has not been translated along with the kigo words themselves, though there are clues. We do not have a kigo culture: the context in which kigo find meaning and literary relationship in Japanese literature. In English we have season words—words and phrases delimiting seasons in haiku; however, there is no official collection of seasonal references in English, and if there were such a book it would not in itself signify a culture. I hope to sort out some differences between kigo (Japanese) and seasonal reference (English/Western), and will henceforth use “kigo” to indicate the Japanese tradition, as it functions in Japan (in Japanese), and “seasonal reference” to indicate the English‑language tradition.


Two haiku in English lacking seasonal reference, part 1

One of the remarkable aspects of contemporary haiku circles in Japan is the prevalence of the Saijiki. The Saijiki (a kigo glossary[1]) represents the established source of kigo. That we have nothing definitive of this sort in English may or may not be a good thing. There has been a long‑term debate about (what) kigo (would be) in English; their importance and relationship with haiku. As kigo are integral to Japanese haiku, the question of how kigo might be treated in our own haiku literature remains a relevant and unresolved issue.


How might we treat these two haiku below, in terms of a proposed kigo culture in English:


between silent moonlit hills

something waiting

to be named

                        --Leslie Giddens[2]


the river

the river makes

of the moon

                        --Jim Kacian[3]


In both, as a reader, I receive a powerful though secondary sense of season; my impression is subjective, as the season is not given. In Leslie Giddens’ haiku, reading the last phrase, “something waiting to be named” I reflect on origins, on seeds waiting to be born, on the origins of names; I envision these moonlit hills as hills of deep winter or winter’s end. The first part of the haiku, “between silent moonlit hills” grounds the poem’s primary impression in the natural world (with “silence” implying a witness). Yet “moonlit hills” itself is not specific enough to yield a seasonal reference. In Jim Kacian’s haiku there are two rivers, and a moon in the text—though one river is actually a metaphorical river of moonlight and/or idiom of moonlight (i.e. a ‘river of the moon’).


We do not find these natural, primordial elements of “river,” “moon” or “moonlit hills” to be seasonal references, as they encompass our planet in time and space, extending beyond seasonal division. We might say, the power inherent in both these haiku lies in their indication of a non‑human‑centered imagination—a native wildness, wilderness. In this sense they resist humanistic inclinations of seasonal division; how would these haiku be treated in the Japanese tradition? I will next move this question into the Japanese context, imagining these poems in Japanese, following a brief digression on the notion of kigo as it relates with senryu and haiku in Japan.


Giraffes in Yokohama

When I ask students associated with the Faculty of Letters at Kumamoto University to explain the difference between haiku and senryu, none being haiku practitioners, I receive a ready answer: “haiku have kigo and senryu do not.” This is a basic historical distinction, and seems to be an elegantly simple solution to genre separation. This same distinction has been made in English, yet what has been missed in this simple observation is that senryu in Japan, lacking kigo, can and often do have seasonal reference. Senryu may also contain kigo—that is, words which are kigo in the haiku genre—thus, found in a Saijiki—but these words not treated as such, in senryu. As in this takeoff of Tsubouchi Nenten (with apologies), there is seasonal indication, without kigo:


    in yokohama –

after the us election



The mention of the US election posits November (autumn) for the haiku; giraffes are not found in a Saijiki. Since I’ve parodied a haiku, the poem is senryu by default; even if there were a kigo to be found, it would not be treated as such. If I were to end the poem with “winter giraffes,” the season is now quite overtly indicated, and the word winter (fuyu) is kigo—but here it is merely winter—as seasonal indication. Ergo, we can see that “kigo” and “seasonal reference” are differentiated concepts—and this is true not only with senryu, but also with modern haiku. Early‑modern and modern haiku often combine haiku and senryu elements, with approaches originating from the wider field of modern poetry. Some varieties of modern haiku do not contain kigo, or may contain more than one kigo. As an aside, it’s only fair to include the original haiku written by Tsubouchi:[4]



yokohama no jyuuichi gatsu no kirin kana



  in yokohama –



Those familiar with Tsubouchi’s work will know that he has made pilgrimages to zoos around Japan, holding an intention perhaps similar to that of Bashō in his journeys to far‑distant places, investigating limits of the known and cultural boundaries.[5]


My students seem generally unaware of the departures of modern haiku, and also consider haiku as difficult if not well nigh impossible to compose. When I’ve asked a class, has anyone composed a haiku, I receive shocked, open-eyed looks as though I were joking. I’ve asked this question for some years, with the same response. Mentioning this fact to several local haiku poets, I mused upon what the perceived difficulty in haiku might be—the presence of archaic kanji, grammatical issues, or the creative challenge involved in approaching a 400 year old, ‘high’ artform? While all these issues are significant, the consensus answer I’ve received is that it is likely kigo, in its manner of use and approach, which presents the greatest difficulty.[6]


Two haiku in English lacking seasonal reference, part 2

Returning to Giddens’ haiku, as a Japanese haiku, we are not sure—is “moonlit hills” kigo or not? In a Japanese haiku circle we would proceed to look it up, and a majority of members do seem to bring Saijiki to these events. Saijiki are increasingly incorporated into electronic dictionaries as well; these same dictionaries can also be used to handily sort out archaic kanji, kanji homonym compounds, synonyms and the like. Searching for moonlit hills, a kigo can’t be found, though “moon” by itself indicates autumn.[7] However, we wouldn’t necessarily know for sure whether the kigo “moonlit hills” has existence as kigo or not, without first checking the Saijiki; so, a given haiku may remain unresolved prior to the lookup process; that is, the poem cannot be fully understood, taken in. This mode of reading presents a sharp contrast with the reader’s experience of haiku in English, where accessibility is generally a given. In that there is “moon(lit)” in the haiku, and “moon” itself is a kigo, autumn would be the season by default.[8] The kigo “moon” envisions the moon of autumn moon‑viewing (tsukimi). So, “moon” is not just any moon: in haiku, it is a kigo moon. The bilingual Saijiki published by the University of Virginia offers this explanation:


Since ancient times, the natural phenomena favored above all by Japanese poets have been the triplet "snow, moon, blossoms" (that is, cherry blossoms). The moon appears in all four seasons, of course, but in both classical poetry and haikai it has been firmly associated with autumn, so that unless otherwise specified, "the moon" means the autumn moon. One reason for this is that as blossoms is the pre-eminent image of spring and snow is that of winter, the moon came to connote autumn. No less important a reason, surely, is that the moon seems to shine with a special clarity in the months of autumn.[9]


What we find here is a kind of symbolic, poetic culture implicit in natural phenomena, with certain phenomena assigned to certain seasons, partly for reasons of aesthetic balance, or due to historic antecedents, etc. Such does not imply that kigo do not have depth, quite the contrary—yet at the same time, kigo is a culture, one which a naturalist might take exception to. In any case, we find that Giddens’ haiku has no seasonal reference in English, but acquires the kigo “moon” in Japanese. We are lucky that ‘hills’ is not also a kigo, as only one kigo is allowed in a (traditional) haiku; we keep this rule in mind, when composing. As well, by looking up the correct kigo, we have learned the season of this haiku.


Just to mention, concerning “migratory birds” (wataridori) mentioned above, we find it is an autumn kigo: migrating birds arrive in Japan from Siberia to winter.[10] They also depart in the spring, but in the culture of kigo, migrating birds migrate only one way, in one season, as far as the kigo wataridori is concerned.


In Jim Kacian’s haiku, imbibing the fullness of the river and brightness of the moon, I sense a brilliant, warm summer night—the enfolded metaphoric image of the moon unwraps as if were at its fullest, brightest apotheosis. Once again, the moon figures prominently, and as with Giddens’ haiku, there is no adjectival modifier for “moon,” so moon becomes the kigo in Japanese, and we have a poem of autumn. Luckily “river” (without a modifier) is not a kigo. What I mean by a modifier is, for example, in the kigo risshun no tsuki “beginning‑of-spring moon,” one finds “moon” is adjectivally modified to connote a different seasonal kigo. Since, for kigo, every named phenomena pertains to a specific season, and often a timeframe within a season (early, middle, late), modifiers are often used to locate phenomena (e.g. river, moon, rain) within a particular season—so, we can’t use “moon” if we mean to indicate a moon of spring, as we can with “moon” for autumn. Due to this fact, an autumn moon is a very brief word of 2‑on, (tsuki), while the early‑spring moon above (risshun no tsuki) is a phrase of 7‑on. In these and other ways, seasonal reference becomes a conceptual sub‑set or attribute of kigo culture.


Looking at our two haiku, we might take pause and consider what might be lost by moving these haiku into a formal kigo system. It seems unlikely that their authors wished or needed to posit, overtly or suggestively, a specific season—though season may be hinted, at a distance: the precise distance of the reader’s imagination in meeting the poem. As a reader, I sense the power and purity of nature, image, natural force, life in these haiku. A sense of the purity of not‑me, of nature and Earth beyond seasonal division. It’s tempting to say that a seasonal reference would reduce these poems. The question of kigo or seasonal reference becomes, in such cases, entirely secondary—in either culture or language—as this same argument has also been made in Japan for some decades—since the early-modern period (early 20th century). The name for this type of haiku, a haiku lacking kigo, in Japan, is “muki-haiku.” But we can’t easily use this term in English for those haiku lacking seasonal reference. All haiku in English are muki‑haiku from the Japanese point of view, as we do not have kigo culture. Rather, in English we have haiku with or without seasonal reference.  


We run into another problem in Japanese: in the case of muki‑haiku, the author must either tell us they are muki-haiku, or be known to write muki-haiku. Otherwise, according to the general environment of kigo in the haiku genre, we will grab our Saijiki and find autumn in both these haiku (in Japanese now, imagine). At issue is the treatment (in a Japanese context) of a haiku which appears to have kigo—which the author does not wish to be “read” as having kigo—while still considering it as haiku, and not a senryu variant! We do not confront these issues in English, but we immediately would, if a kigo culture were implemented. Various modern Japanese poets have solutions to the problematics of kigo and haiku, to be witnessed in modern haiku compositions. Natsuishi Ban’ya has for instance offered a system of keywords, a revolutionizing of kigo culture into a suggested keyword culture. In English, we may not appreciate the gravity of the problem of kigo and the consequent desire to reform, resist, transmute, reject, or otherwise alter that culture.


So, every English‑language haiku is muki-haiku (haiku lacking kigo). Do you agree? Might English‑language haiku seem more similar to senryu, when translated into Japanese? It’s quite possible. We do not (yet) have a kigo tradition—does this fact “damage” English‑language haiku when viewed from a Japanese perspective, with its ancient kigo tradition, so fundamental to classical haiku? For some it would seem so.


Delimiting kigo

It cannot be said that kigo exist outside of the Saijiki in any real sense—if this is stated as such, in a Japanese literary context, the topic may become politically contentious. Below, Tsubouchi Nenten broaches the issue delicately when he says, “the Saijiki is only one standard of kigo; kigo are always being born and have died within the nexus of haiku poets.” Yes, quite true, but one notes that until the new term is officially documented, selected, referenced and published in an acceptable Saijiki it has not yet come into definitive existence as kigo. There is a difference between being born and arriving. The “death” of a kigo may occur these days as a function of disuse, but it’s hard to shake kigo out of electronic dictionaries, especially with so much cheap memory available. I think it fair to say that in Japan kigo don’t simply exist, they must also be published—a kigo without a Saijiki is like one hand clapping.


This is part of the existential dilemma of kigo—their necessity for editorial approval, publication, and hence exclusivity, their bureaucratization; factors which have in part caused a number of modern Japanese haiku poets to reject, subvert or revolutionize kigo use. These revolutions are solutions, evolutions and new evocations of the modern and postmodern spirit of our time. I’d like to refer you to recent relevant comments made in this regard by Hoshinaga Fumio (excerpted in the endnote).[11] In the various sorts of haiku circles I’ve attended, with the exception of the most generically known kigo, I’ve  observed poets looking up kigo quite often when reading or working on haiku compositions.


We may, in adopting kigo and building our English‑language Saijiki look forward to keeping our Saijiki ever‑ready, and this would represent a revolution in our haiku practice. It could be quite an inspiration, depending on our degree of sophistication in relating with our own sourcepoints for kigo images, in relating to the culture of our own literature and literary history. Would medieval flower language be a good place to start? Should we focus on seasonal references stemming from our own haiku genre? We might select haiku from acclaimed haiku writers or contest winners. There are excellent haiku and haiku writers to be sure, but as yet there is no wider Anglo‑American literary consensus regarding the excellence of a body of work in the haiku genre, outside the circle of haiku practitioners. We have no Manyōushū, no Kokin Wakashū, and all that these root literary texts represent in the Japanese context, concerning the centrality of short-form season-related poetics to a prevailing national literary culture. In Japan, the first major Saijiki collections were published in the Edo period, centuries after these first poetic anthologies, in which the earliest layer of kigo had arrived from China. Are we attempting to put the cart before the horse, in willing a glossary of official terms into being, in desiring a kigo tradition of our own? Are kigo really a good fit with haiku in English—both in terms of language and our haiku culture? Might not having just “seasonal reference” and “non‑season” haiku serve us well enough? In the first magazine devoted expressly to haiku, John Bull wrote, “If there is to be a real ‘American Haiku’ we must—by trial and error—work out its own standards.”[12] Some experimental kigo trials have occurred (e.g. William Higginson’s Haiku World[13]), but so far such forays haven’t caught on.


A kigo project in English

Recently the World Haiku Club (WHC) began a “worldwide kigo project,” which will collect “viable kigo.” The prospectus of the project states that:


The real issue is whether or not finding local season words pertaining to specific climatic and cultural zones or countries in the rest of the world would be possible, plausible, desirable, useful or necessary in terms of making what is written as haiku more like haiku or better haiku. The fact that many poets have thus discarded or dismissed kigo (some have even condemned it as being no more than a weather forecast and not poetry) as inapplicable or irrelevant has damaged haiku outside Japan and denied it cultural and historical depth.[14]


Certainly, this view posits the need for kigo in English, as it implies that we have up till now been writing faux haiku, and that we could be writing something “more like haiku or better haiku,” with approved kigo. And the result of not having a kigo tradition is damage and denial. What is the damage implied? That of the reputation of haiku in English, as viewed from Japan? Or, from those who feel, in whatever language, that haiku are a joke without kigo? As for the denial of cultural depth, this seems a thorny problem. I agree that in many mediocre haiku, seasonal‑reference-as-weather‑forecast is rife. But then, to look fairly at any literature we ought to examine the best it has to offer not the worst, and there are quite a few excellent haiku not only without kigo but without seasonal reference—in English and in Japan—in any country. So we enter into the zone of kigo politics, with the implicit theme here that without kigo, i.e. a definitive, accepted official published glossary of kigo, we cannot have cultural or historical depth.


after the bombing

ruins of a bridge

linked by the fog

--Nebojsa Simin[15]


In the above haiku, which arguably possesses historical as well as cultural depth, “fog” may or may not connote season; in any case, the felt season here is war. It is any season, the season of hell. In Japanese, “fog” (kiri) is kigo. Its use as kigo in this haiku would subvert the traditional sense of kigo, at the very least. Why? Tell me what “spring” (the kigo season of fog) has to do with this poem. At most, the kigo would imply an additional level of irony. But the point I think you’ll agree of this natural element lies precisely in its insubstantial “as‑if” character, its contrast with the violent machinations of humankind, rather than its possession of a presumed seasonal quality.


As a reader of haiku and composer, if I am to purchase a future kigo‑publication in English, I would hope to learn how modern haiku are to be treated, how the modern and postmodern imagination, vision of haiku is to be expounded. I would hope to be inspired with new approaches to kigo, modern techniques to revolutionize and subvert, which reflect the forefront of modern traditions, worldwide—and especially in Japan, which has had a large headstart. Looking through the various kigo projects, what I see is factory work, specimens, taxonomy. Where is the genius? For surely we will need a work of genius to inspire us to carry that kigo glossary around in our pockets.


Confusions, confabulations: kigo equals seasonal reference equals Nature?

Writing in 1986, Cor van den Heuvel published an influential preface to the second edition of The Haiku Anthology, reprinted in the current third edition (1999).[16] These sentences may have caused some confusion:


It seems useful to me to keep the two genres distinct in somewhat the same way the Japanese do—haiku relates to Nature and the seasons, senryu relates to human nature. Traditionally, the Japanese have ensured this by insisting that to be a haiku the poem must have a season word (kigo), while a senryu does not (pp. xlv-xlvi).


Indeed, one reason for the amazing popularity of senryu, from the Edo period on, was that one didn’t need a Saijiki (or to deal with kigo), and senryu found fertile soil in the haikai tradition. And yet, although haiku is considered a “serious” literature, its roots are likewise sunk deeply into haikai (humor). Recently, a fascinating book appeared addressing this topic, Haiku Humor by Tsubouchi Nenten.[17] The above preface‑remarks were written in 1986, at a time when a focused awareness on modern Japanese haiku was just beginning to be cultivated in English; thus, the fact that over the last century in Japan senryu and haiku elements have been intermingling in numerous ways was seemingly missed. These days, the categorization of haiku as relating to nature, and senryu with human nature, would constitute reductive overstatement. Yes, there is a locus to each form, with senryu being generically comedic, often utilizing ironic, acerbic or witty social comment; and haiku possessing a more objective style, focused on the natural world—but there are many points of crossover. The actual situation is not so simple—as with most things modern in whichever culture, we find interpenetration, synthesis and fusion, rather than exclusivist classicist purity.


From the conservative or traditionalist point of view, there may be an “insistence” that haiku have kigo, but it is not the case that “the Japanese [insist that] to be a haiku the poem must have a season word.” The haiku tradition does not find unanimity regarding muki‑haiku. And we have the term “muki‑haiku” itself, which would be an oxymoron according to the above dictum.


As well, we see the idea of “kigo” being conflated with “Nature” (and opposed to human nature?), and in turn conflated with “seasons,” and that in turn conflated with “season word.” It seems important to parse these ideas, discriminate the nuances which give each term its distinct theoretical and applied meaning.


Two factors make haiku and senryu genre‑separation a bit easy to manage in Japan. The first is social—there are senryu circles, senryu websites, journals. If, as a composer, you call what you do senryu, there’s a place for you, a literary community; and similarly, for haiku. The second factor is kigo—senryu do not “read” with kigo (as previously discussed). When you put these two factors together, and add the stated intention of the poet, you have clear distinctions which are not based on the level or type of humor or topical content of whatever (approximately) 5‑7‑5‑on poem you are looking at. Senryu, whether having or lacking kigo, may have seasonal reference, after all. Likewise, haiku may have humor and present a social subject, and may also lack kigo (muki‑haiku). Hoshinaga Fumio’s poem:[18]


遊園地 にナチスいっぱいです  秋

yūenchi ni nachisu ga ippai desu     aki


the amusement park

full of Nazis  –

this autumn


provides social critique. Its humor is dark, acerbic, biting. It is not senryu. Why not? The author does not state the genre of the poem, one way or another. Nonetheless, it appears in the Chapter “Lament” in a book containing a number of notable haiku. Hoshinaga is a national figure, acclaimed for his haiku, is a noted haiku teacher and founder of a haiku journal. As well, there is something quite serious indicated in this poem—a belly laugh it does not provoke. The author has also mentioned that over 90% of his poems use kigo, and we do find the kigo term “autumn” in this poem (a seasonal reference, in this English translation). Nonetheless, this particular kigo representation seems one which refers more to cultural decline than environmental weather. In fact, Hoshinaga has commented that he never uses kigo merely to convey an environment of naïve realism. His haiku often utilize senryu (haikai) elements; as in Bashō, one finds a radical blend of the vernacular and the serious, “high” and “popular” literary cultures (“amusement park” and “Nazis” relate to modern, international cultural history and themes). In essence, the poet places this poem within the haiku sphere: it has kigo and 17‑on. That is quite enough. Even if this were muki‑haiku, its placement and the poet’s intentionality are enough. If only things were as straightforward in English.


A confusion in English is the idea that kigo equals nature. This is a misreading of kigo, I believe. As Hoshinaga mentions, “kigo [may be] more of a symbolic element.” The writer may experience kigo “through your heart (inner sense), not through seeing, touching, and so on.”[19] This stylism of kigo provides an environment which may be symbolic, historic, literary, surreal, or otherwise impressionistic, interpretive, or subjective, as Tsubouchi also points out, below. What is the true intention of kigo?


This area of the symbolic and subjective brings up a philosophic question. What do we mean by Nature? In his ecocritical essay, “Unnatural Writing,” Gary Snyder offers some insight in critiquing the  assumptions of earlier nature writers:


There is an older sort of nature writing that might be seen as largely essays and writing from a human perspective, middle-class, middlebrow Euro-American. It has a rhetoric of beauty, harmony, and sublimity. . . . Natural history writing [is] semi-scientific, objective, in the descriptive mode. Both these sorts are "naively realistic" in that they unquestioningly accept the front-mounted bifocal human eye, the poor human sense of smell, and other characteristics of our species, plus the assumption that the mind can, without much self-examination, directly and objectively "know" whatever it looks at.[20]


These comments may also serve as a relevant critique of haiku. Snyder, like a number of modern thinkers, asks us in these introductory remarks to carefully examine the nature of human awareness, to question habitually unquestioned characteristics of reality (as embodied in romanticism, realism, naturalism, humanism). It seems that Snyder and Hoshinaga have in common a modern or postmodern spirit of exploration in terms of both poetry and philosophy. Perhaps it is not kigo that will link us as international practitioners of haiku, but a deeper understanding of the modern and postmodern ethos of our respective literatures, and how this understanding is expressed—as we increasingly share a globalized, communal zeitgeist. I wonder why we can’t locate this spirit, value this same ethos, more centrally in contemporary haiku thought.


Simply put, kigo exist in Japanese and do not exist in English. In English we have a season word/phrase tradition that began with translations from Japanese into English, which “interpreted” kigo into non‑kigo literary culture—our literary culture, which does not have the conceptual or historical frame of kigo. And so, non‑kigo “season words,” indicating a seasonal reference, were born. Kigo (in Japanese) and season words in English are apples and oranges. What we've done with kigo is to map and then graft our received idea "seasonal reference" back onto something entirely more dense, troublesome, esoteric and perhaps fantastic and unique in literature: Japanese kigo—reverse‑projecting “seasonal reference” (our received concept) back into the Japanese tradition. In English a wild duck is a wild duck. In Japanese haiku a wild duck is kigo, and kamo (wild duck) is a summer duck. Kigo are not “natural,” but rather, nature reified. Tsubouchi refers to several modes of kigo reification, in locating the great treasure of kigo to haiku: its true intention.

Tsubouchi Nenten: Kigo

Hopefully, An Introduction to Haiku (Haiku Nyūmon) may one day be presented in its entirety, in translation. Tsubouchi’s style is highly informal, intimate, witty, and dialogic. The book includes the cover description, “ofuro de:” it’s part of a published series printed on paper able to withstand the rigors of the daily bath! As such, one imagines its major reader‑audience as likely an older set, enjoying a refreshing and enlightening read during a good soak. I’m grateful for the opportunity to share this small offering of Tsubouchi’s insight and soul.


Please note that text within parenthesis represents my added comment; this method seemed preferable to taxing the reader with footnotes. The linear text was also separated into paragraphs.



Concerning the “Glossary of Seasonal Terms for Haiku Composers (Saijiki)”[21]


So, at this point, perhaps you would order me to explain kigo, and the Saijiki? What particularly do you want to know?


For example, why was kigo first created and used and why is kigo only seen in a Saijiki— are these some of the questions you have?


It is said that kigo were first created in China to convey the concept of a yearly cycle composed of the four seasons of spring, summer, autumn and winter. Before the concept of four seasons came to Japan, we had the concept of two-season pairings. We can consider that the present special seasonal times of New Years (shōgatsu—three days) and the Bon Festival (three days in summer—variable dates) are holdovers of this ancient concept.


In the poetry and verse of the Manyōshū (759CE, the earliest collection of Japanese poetry), the yearly cycle was classified into the four seasons, with differing aesthetic feelings related to each season. This aesthetic sensibility became fully realized some ten centuries ago, in the Kokin Wakashū (also known as Kokinshū, 905CE).


Spring is “flower” (Cherry blossom); summer is “hototogisu” (a little cuckoo—smaller than a pigeon); autumn is “moon” and “fall‑leaf colors;” winter is “snow;” these are the typical images for each season as seen in the poetry collections. Then, in medieval times, kigo came to be used in renga, in which the first stanza (hokku) used a seasonal reference as a kind of salutation, which related to the poetic expression or evocation. So, kigo became very important.


Concerning kigo, at that time (from waka to renga—see “History of Kigo” chart below),


Center: Kigo of Waka [waka no kigo 和歌の季語]

First Ring: Kigo of Renga  [renga no kigo 連歌の季語]

Second Ring: Kigo of Haikai [haikai no kigo 俳諧の季語]

Third Ring: Kigo of Haiku [haiku no kigo 俳句の季語]



kigo was limited (restricted) to a sensibility of highly refined elegance (for example, haigon—the Chinese pronunciation of kanji/word compounds, and/or slang idiom—was not allowed); but in the (later) haikai era, the sense of kigo has expanded to include a commonplace, mundane and “folksy,” or “everyday” sensibility. This evolution in the sensibility of kigo seems quite natural, as the haikai was a poetic form in which slang was used, a kind of “street poetry.” Elegant poetry is waka or renga. This is obvious, isn't it? Thus, the Saijiki of the Edo era is roughly equivalent to our contemporary Saijiki, and (the Saijiki as such) was brought into definitive existence in the Edo era.


There is a measure of covenant in the season word. This covenant can be described as one’s true intention or true sensibility. For example, considering “spring wind” (haru kaze): there is a word, shunpūtaitō (from the Chinese: “wind blowing mild and genial”) which can be applied to human character. It is made of four kanji characters: haru (spring) and kaze (wind) plus the compound (taitō), meaning calm, quiet, peaceful wind.


It is a true intention of the spring wind.


The true intention is a tradition of the spring wind used by the waka, the Chinese poem, and the haiku, etc.


So, the single (kigo) word is a distillation wrought by tradition representing the true intention of kigo. The Saijiki (kigo glossary) elucidates (glosses) the true intentions of such words. In a nutshell, the expression such as “lonely spring breeze” (sabishii haru kaze) does not exist as kigo.




So, when the spring breeze is felt as lonely, what am I gonna’ do?


In this case, the spring breeze: it’s calm and warm; however, I do feel that it’s lonely—nonetheless, there is nothing concretely expressing it.


Here is my haiku,



harukaze ni haha shinu ryuukakusan ga chiri

to the spring wind
mother dead, herbal medicine


Concerning this haiku, in this case the spring wind blows calmly and peacefully. However, the person (figure) who exists in the wind is looking at the spring breeze feeling sad, because their mother has died. Because the spring breeze is calm and peaceful, the person's mind (heart, feeling) is also (sensed as) fleeting, as unreliant as the herbal powder that scatters to the wind.


Recently, there are people who make muki-haiku (haiku lacking kigo); concerning kigo, the external, objective world is divided into four seasons as in a mechanism or system; that is to say, the external, objective world of four seasons (for kigo) is something like wearing spectacles (blinkers). For example, the tomato and the cucumber appear in the market all the year round, though the kigo (for those vegetables) is summer. When the external world is delimited in this way at the four seasons, the delimitation marks the rhythm of life.


You ask me are season words man-made? 


Yes, exactly.


There are originally no four seasons in the natural world, but humankind delimits the natural world at the four seasons, and so it happens that kigo arise, as one result.


In a word, kigo is a culture. Because there is a culture, there are generally trends, but sometimes the change is drastic. In one example of drastic change, during the Meji era the solar calendar was adopted and the kigo seasons changed greatly. Therefore, the Saijiki is a collection of kigo, however the entries in the Saijiki do not cover all kigo. The Saijiki is only one standard of kigo; kigo are always being born and have died within the nexus of haiku poets.  


A measure of covenant—true intentions

Tsoubouchi points out that “the single (kigo) word is a distillation wrought by tradition representing the true intention of kigo.” In this sense, any single kigo term is not only an image or reference to season and nature—this would be the superficial reading. Kigo are distillations possessing a complex alchemy: each term is a multidimensional surface measured within a cosmos. Modern haiku writers may subvert or otherwise alter the means or methods of kigo presentation in their compositions, but many continue to utilize the transformative poetic power inhering in kigo culture; the power of what Hoshinaga refers to as the “environment” spawned by kigo—an environment which includes nature and culture, objective and subjective, fact and fancy, the morphic landscapes of psyche—that is, “reality.” As seen above, Tsubouchi isn’t talking about the true intentions of seasonal reference, but rather the true intentions of a wellspring of literary, philosophic and spiritual culture, with ancient roots indeed. What are these true intentions? And, what are our own intentions, regarding kigo? Perhaps we need to discover its intentions, before willing our own. It seems that rather little of the underlying culture of kigo is known, in English, to the present.


In imagining a kigo culture in English, can we find a historical bullseye to our kigo target, a most ancient layer of kigo—come to observe the growth rings of succeeding centuries, “a natural . . . evolution of sensibility”? It would seem a highly paradoxical approach to “force” naturalness—and yet, kigo are also “man‑made.” Given this inherent paradoxicality in kigo (an iconic or sur‑ real naturalness, in which the cooked serves to indicate the raw), is there potential in the idea of kigo, which might appeal to those interested in the modern and postmodern—the future of haiku?


Shall we look more deeply into the history of our own literary relationship with the natural world, its flora and fauna—or turn to Japanese Saijiki originals? The University of Virginia Library already has an excellent, readily available online bilingual work‑in‑progress in its Japanese Haiku, A Topical Dictionary,[22] impressively informative and scholarly. Or would we do best to avoid collecting terms altogether and seek first the heart of kigo, its true intention. Perhaps only at such a juncture will we have acquired a needed measure of insight to move us further towards new cultural and indeed psychic sensibilities, regarding the actual words of a proposed kigo world. Whatever words they would be, these upstart kigo, they would be marked but not delimited by haiku, as kigo represent a more extensive culture than that inscribed by any single poetic genre; perhaps a collaboration between the arts is in order. In any case, kigo are not a subset of haiku; rather, haiku utilize the historical culture and tradition of kigo, in which the haiku genre participates.


It may be that, as with all unique cultural treasures, we may witness, study and admire an achievement not of our own making rather than possessing it; or alternatively, proceed along some new and entirely different line. In fact it is unclear to me how to proceed, regarding the birthing of a kigo culture in English. As ever, it is likely the poets themselves who will open us new haiku vistas—yet there also exists a need for further understanding.


[1] By including a selected historical sampling of haiku to illustrate usage and application, kigo are glossed, rather than defined. In this sense, “kigo glossary” seems preferable to “kigo dictionary.”

[2] Blithe Spirit, 13:2 (Journal of the British Haiku Society, June 2003), p 5.

[3] First Mainichi Anthology of Winning Selected Haiku [daiichikai mainichi ikutai shyousakuhinshi] (Tokyo: Mainichi Shimbun, 1997).

[4] Tsubouchi Nenten, An Introduction to Haiku [Haiku Nyūmon] (Tokyo: Sekai Shisōsha 1995, 1998), p. 12.

[5] Cf. In The Narrow Road to Oku, Bashō’s famed haibun travelogue, the village of Oku is both actual place and metaphor: it was generically considered the edge and end of civilization, beyond which was wilderness. This edge was in fact a metaphorical cultural boundary, as the existence of the Ainu attest. Tsubouchi, in a contemporary context, observes African animals trapped in Japanese zoos; one senses a continuity with and eulogizing of Bashō’s “Oku,” both as a travel goal, and in positing a desire for specific modes of poetic journey.

[6] Recently, each student in a class of 20 Sophomores studying in the Faculty of Literature, Department of  English, was asked to find a work of contemporary literature in 5‑7‑5‑on.  The student selections: eight senryu; five kotowaza (proverbs, idioms); two traffic mottos and one motto (kokusai hyôgo & hyôgo); two 'other' 5-7-5 works; and, two sayings of Chinese origin. No haiku were selected. Online video presentations of these selections are available:

[7] As can be found in Japanese Haiku, a Topical Dictionary, University of Virginia Library; “a work-in-progress based on the Nyūmon Saijiki by the Museum of Haiku Literature in Tokyo.” Available online: To find the reference, click the link “Full Entries,” scroll down to “aki: Autumn,” find the subsection “The Heavens,” click the link “tsuki.”

[8] For those interested in a Japanese translation of “moonlit hills,” some possibilities might be tsuki oka ni, oka ni tsuki, or okatsuki. In each case, the kigo is “tsuki,” moon.

[9] See endnote 7 (ibid); the entry “tsuki” is found in “Full Entries:”

[10] See the entry “wataridori” (ibid).

[11] “The Miraculous Power of Language: A Conversation with the Poet Hoshinaga Fumio,” Richard Gilbert, (Modern Haiku, 35:3, Autumn 2004), pp. 27-45. Available online: The following is a pastiche of Hoshinaga’s comments relating to kigo: “I have repellence, revulsion exactly against the formal rules and approach, kigo, and various formal necessities. . . . Haiku is a centralized art. For instance, looking at the saijiki (haiku kigo or season-word dictionary), the kigo focus only on the Kyoto or Tokyo (Edo) locales. There are no "local" saijiki: you cannot find local characteristics. Given such a situation, local people have a sense of inferiority, when regarding the "center" of the tradition. This type of inferiority-complex provides a kind of energy for my creation. . . . . [Notwithstanding,] Kigo is very useful and convenient for creating a sense of place (where) and time (when). We can say that a kigo is just one word — but this one word can speak volumes. . . . Finally, how a person lives in the time and the place; makes a relationship with the time and place — you can describe or express a cross section of life just by identifying "person." I can express a cross section of life with kigo — so kigo make it much easier to compose haiku. From this point of view, kigo is very useful and symbolic language. This is why ninety percent of my haiku contain kigo. . . . this use of kigo is more of a symbolic element. . . . I have real experience, real experience of kigo. This is why I can write haiku. It seems that I make haiku with my brain, but I can say I make kigo with my real experience, my sense of reality. . . . Using a seasonal reference may be a good hint or suggestion for an English-language haiku writer, but sometimes you have to write naked.”

[12] American Haiku (1963), quoted in The Haiku Anthology  (3rd edition), “Preface to the First Edition,” Cor van den Heuvel (Norton, 1999), p. lxi.

[13] Of particular note is Haiku World: An International Poetry Almanac, William Higginson, (Tokyo: Kodansha, 1996).

[15] KNOTS: The Anthology of Southeastern European Haiku Poetry, (Dimitar Anakiev and Jim Kacian, trans., eds., (Tolmin Slovenia: Prijatelj Press, 1999). Nebojsa Simin lives in Novi Sad and is editor‑in‑chief of the influential Serbian publication Haiku Letter Magazine.

[16] The Haiku Anthology (3rd edition), Cor van den Heuvel (Norton, 1999).

[17]  Haiku Humor [haiku no yu-moa], Tsubouchi Nenten (Tokyo: Kodansha, 1994).

[18] “Hoshinaga Fumio: Selected Haiku from Kumaso-Ha,” Richard Gilbert, (Modern Haiku, 35:3, Autumn 2004), pp. 46-55. Available online:

[19] Available online, see endnote 11 (ibid).

[20] A Place in Space, Gary Snyder, (Counterpoint Press, 1995), pp. 163-172.

[21]  “Composing Haiku, Part 1: Kigo” in An Introduction to Haiku [Haiku Nyūmon], Tsubouchi Nenten, (Tokyo: Sekai Shisōsha 1995, 1998), pp. 50-54.

[22] See endnote 7.

return to research page