The Kokinshū PDF Print E-mail

the lyric and the conventional
by Laurence Mann


Of the extent corpus of Early Heian Japanese waka (for our purposes, we can think of it as a short, 31-syllable poem, sometimes referred to as tanka, “short poem”), a considerable portion is contained in the “Anthology of Poems Ancient and Modern” (hereafter, Kokinshū),1 completed around 905 CE (Itasaka 254). As the first poetry collection to be commissioned by the Emperor himself, the Kokinshū holds an inviolable tenure in the history of imperial patronage of the arts in Japan and is largely responsible for the establishment of the limited thematic and rhetorical framework within which, exclusively, the composition of waka was to take place from that time onwards. It acted also as a handbook for the critical analysis and interpretation of later works; many private (lyrical?) poems have almost certainly been lost to posterity because they did not confirm closely enough to the ideals of the Kokinshū and thus did not make it as far as any of the multimedia environments in which waka could be 'published' – uta-awase, the byōbu-e, the hyakushu-uta, etc. (Shirane passim). Perhaps it is because of the ubiquity of its stylistic and hermeneutic legacy that the Kokinshū, once esteemed so highly, attracted a good deal of criticism from commentators schooled in “modern” Western European literary theory, which arrived in Japan, along with no insignificant amount of ideological baggage, in the late nineteenth century. All of a sudden, its artifice, word play and stylistic conventionality are found to be stifling to lyrical expression – and this cannot be a good thing!

Amongst the various attacks to which the Kokinshū has been subject over the course of the last century and a half, one of the most vitriolic is to be found in Meiji critic and poet Shiki Masaoka’s “Letters to a Tanka Poet”, in which he describes it as “third-rate” and its chief compiler as “a hopeless poet”, while mocking those of his contemporaries who, before changing their views in accordance with the fashion of the time, had participated in an “imbecilic” veneration of the text (n. pag.). His views were shared by much of the early twentieth century Japanese intelligentsia, including exponent of free verse and critic Sakutarō Hagiwara, for whom the collection was “just one lacklustre poem after another…not worth reading” (n. pag.).

Criticised for the dull mediocrity of its content and for its overdependence on conventional motifs, the Kokinshū was considered inferior to other poetry collections by modernist2 (including nationalist) Japanese scholarship until the Second World War and – while essays in praise of the it are to be found among the work of Japanese scholars of a more traditional waka-compositional bent – it has received a generally frigid reception in post-war scholarship too, both inside and outside Japan.

Of the many criticisms levelled at the Kokinshū, that with which we are concerned here was made by American scholar, Helen McCullough, in her study of the text, Brocade by Night: Kokinwakashū and the Court Style in Japanese Classical Poetry, first published in the early 1960s. McCullough suggests that the poetry of the Kokinshū is too heavily reliant on established conventions and, therefore, in contrast with that of its thirteenth century descendant, the Shinkokinshū, or “New Anthology of Poems Past and Present”, it lacks the credentials necessary to qualify as “lyric poetry” (8). So, what reason should we have to question this?

McCullough’s argument, I would suggest, is based on two overarching assumptions. Firstly, that the waka found in the Kokinshū are “trammelled” (Bownas and Thwaite lxi) within the confines of socio-rhetorical and prosodic convention and, secondly, that the lyric poem is not. As we shall demonstrate later, the first of these assumptions can be effectively countered using evidence from the text itself. As for how to deal with the second, we might look to yet another Western “ism”, postmodernism, for inspiration.

Postmodernism is, of course, an appellation we like to give to a variety of schools of thought that emerged as a reaction to the dominance of modernism and which burgeoned in influence, throughout the Humanities, during the late twentieth century. Postmodernist, post-colonialist and post-structuralists have attempted to reformulate the parameters within which literary analysis took place, avoiding what they perceived as interstitial lacunae arising from the inflexibility of modern science/non-science nomenclature, instead turning towards narratology, reader response, deconstruction and a miscellany of other recipient-centric approaches. Texts now need to be stripped to reveal their underlying ideologies in order that, ostensibly at least, they can be considered on their own terms. This is why McCullough's preoccupation with “individuality and originality (or their absence)” (Okada 36) has been criticized by other scholars, notably Richard Okada, who notes, in his review of Brocade by Night: “The search for originality...goes against modern-day critical practice in the West. Such notions as ‘originality’ and...‘authorship’ have been vigorously put into question. There has been a shift in emphasis from author to ‘reader’ and ‘text,’ both terms having undergone considerable refinement in the past two decades” (37). In this paper, I hope to contribute to the debate over McCullough's treatment of the Kokin poetry, focussing particularly on the problematic concepts of lyric and conventionality – to which adequate attention has not be dedicated, by Okada and other reviewers so far.

The lyric, for modernists, was essentially a short poem composed of the original and personal sentiments of an “I”, an individual poetic persona usually identifiable with the poet himself but always with the individual. Even where that is individual was interpreted as a reader-construct, as in some models, the lyric was seen nevertheless as executing “the unifying role of the individual subject” (Culler 170). Conventional elements – the dictates of taste and society – impede the expression of this subject and, therefore, should be avoided. Ideally, a lyric poem would exist only as a stream flowing in the mind of the individual, channelled by its emotions and, with its occult, fluid power shape the actions and characters of its protagonists. As Morris notes, with more than a touch of cynicism, “the lyric method almost demands that a high degree of energy and ingenuity be diverted into the elaboration of romantic characters with complex emotional sensibilities” (563).

While post- and anti-modern critiques are hesitant to abandon once and for all this lyric schema (I shall avoid the word archetype) – as we come to realize when we encounter comments such Okada's footnote “one of the most obviously contrived in the whole collection, [Ariwara no Narihira's poem] KKS 411 [i.e. Kokinshū 411] is perhaps one of the least lyrical poems one can imagine” (36) – they do provide us with some tools to examine our understandings of lyric, afresh.

Michel Foucault, associated with postmodernism and post-structuralism (although he appears to have disliked both terms), has claimed, from the time of “Les Mots et les Choses” onwards, that all knowledge – and systems pertaining to the classification of knowledge – that have existed in the West since the late eighteenth century form part of the same overarching epistemological framework or, in his words, the “episteme”. In “Les Mots et les Choses”, he famously concludes: “… the entire modern episteme…was formed towards the end of the eighteenth century and still serves as the positive ground of our knowledge… [It concerns] man’s particular mode of being and the possibility of knowing him empirically” (385).

According to Foucault, the late eighteenth century saw an intellectual sea change take place in Europe and North America. It was the late eighteenth century when the Discourse was displaced by empirical objectivity and it was also then that the Arts – distinct from but adjunct to this new brand of knowledge – were forced to fashion for themselves a fresh mould in the self-justifying otherness of subjectivity. The lyric was to emerge from this mould altered and invigorated, to the point that, for the first time since Callimachus, critics were ready to “assert its superiority over epic and drama”. The unadulterated and succinct expression of emotion that characterized this new form of lyricism became, as Johnson puts it, “not only beautiful but also the only beautiful” (82). In other words, the lyric became a poetic standard in its own right – and it is by this standard that McCullough and scholars like her judge the waka poetry of Early Heian Japan.

Thus, Foucault (who I have quoted here only as an example of the post-Saussurian thinkers that have shaped the course of late-twentieth and early-twenty-first-century criticism) has provided us with a justification for re-evaluating the criteria by which poetry is included in and excluded from the lyric bracket and, more generally, for questioning any superimposition of analytical conditions defined within our episteme on the literature of its antecedents or concurrent but unrelated epistemes. This paper aims to form part of such a re-evaluation.

Having established the raison d’être of our discussion, we are now in a position to put aside our “isms”, at least for the time being. It must be stressed that this is not a postcolonial or a poststructuralist reading. It is not intended as a post-anything reading. No “deconstruction” will be attempted here. Nor will we jump headfirst into the discussion of immanent power-relations present in Kokin poetry (although, no doubt, they do exist and would make interesting subject matter). Instead, and in response to McCullough’s remarks, we shall restrain ourselves to thinking about the extent to which the environment in which the Kokin poems were created might have rendered them conventional and the implications this could have for their status as lyric poetry– in the context of our new understanding of both as contested concepts. Implicit in such a discussion is the corollary, and more general, question – whether the concepts of conventionality and lyricism are polarized, as the modern axiom dictates, or whether elements of both can be juxtaposed, and coexist, within a single poetic entity. And, since all our terminology is alien to the pre-modern Japanese critical lexicon – and, as we have shown, it eludes succinct definition even in English – it will behove us to look to examples from the so-called “lyric traditions” of other regions for the purposes of comparison.

Therefore, the remainder of this paper will be structured in a bipartite manner, beginning with a discussion of prosodic conventions and thematic precedents observable in Kokinshū waka, before moving on to look more generally at how elements such as these affect our consideration of poetry and the generic classification to which we assign it. During the course of our discussion we will, hopefully, come closer to a working definition of “lyric” but we must nonetheless remain mindful of imprecision of the nomenclature at our disposal throughout.

The poetic material of which the Kokinshū is comprised was, for the most part, composed at a time when the judicative, administrative and fiscal infrastructures that maintained the Early Heian oligarchy were already firmly established – allowing courtiers a life relatively free of mundane concerns, in which they were able to develop and cultivate highly sophisticated aesthetic tastes, drawing on Chinese example for inspiration (Rodd and Henkenius4). That the Kokinshū emerged from this courtly aesthetic tradition, there can be little doubt, since not only was the collection commissioned by the Emperor himself, but the poet most heavily involved in its compilation, Ki no Tsurayuki, was considered by his contemporaries at court to be a chef de file of fashionable waka composition.

Early Heian waka can be sub-classified into two groups according to time of composition. The first dates from the time of the Rokkasen (mid-ninth century) which, together with older Manyō period verses, comprise the “ko”, or “Ancient” element in the Kokinshū’s title. The “kin” or “new” part refers to the poetry of Tsurayuki and his contemporaries, roughly a generation later.

During the ninth century, a vogue for all things Chinese gripped the Japanese court. Chinese culture and literature had played such a fundamental role in the public and aesthetic lives of the courtiers, in the period leading up to the compilation of the Kokinshū, that it has been described, by Japanese literature specialists (seemingly with a nationalist agenda) as a Koku fūan kokuji dai (“Dark Age of National Style”).

The Japanese literati of the ninth and tenth centuries were thus undoubtedly familiar, to a greater or lesser extent, with a considerable array of Chinese literary sources. However, perhaps because of socio-cultural parallels between the early Heian court and the Six Dynasties salons in China, it seems to have been poetry composed during the Six Dynasties period that most impressed poets of the two eras mentioned above and, therefore, exerted most influence on the formative development of the Kokinshū (McCullough 327). Its impact is most visible in two areas – firstly, the direct borrowing of subject matter, thematic material and motific elements from Chinese verse and, secondly, an increasing tendency within Japanese poetry toward epideictic display and the use of linguistic devices akin to the conceits found in the poetry of the European Renaissance.

With regard to subject matter, there can be little doubt that the Japanese poetry found in the Kokinshū was heavily influenced by Chinese precedent. It is well documented that situations or stimuli perceived as possessing insipient poetic worth within Six Dynasties verse (particularly the so-called yongwu or ‘poems about things’) were transferred directly into the Kokin waka (Brower and Miner6) – “migrating geese, scattering leaves…gentle breezes, new willow growth, and lingering snow”, to name but a few (McCullough 72).

Although the tone of the two poems is quite different, a comparison of the following short verse penned by Southern Qi Poet, Fan Yun (McCullough 66) with an example taken from the first book of the Kokinshū illustrates for us the borrowing of “conventional” poetic material, as well as the Japanese adoption of the concept of “elegant confusion” from China (Rodd and Henkenius 51):


Fan Yun (451-503) “A Song of Parting”

Then, when I left, the snow was like flowers;
Now, when I return, the flowers are like snow.


Topic Unknown



so longingly have I


awaited the fresh flowers


of spring     that they have

kieaenuyuki no

dyed my soul     and I see snow

hana to miyuran

as clustered blooms on branches


During the Six Dynasties period, court poetry (mostly shi) was, by and large, conceived as an elevated means of social interaction, or “polite discourse”, in much the same way as the prose-poem fu had been during the Han Period (McCullough 48). In order to demonstrate his linguistic virtuosity and copious literary knowledge, in the context of a banquet or court function, the Six Dynasties poet could be said to have striven toward familiarity, rather than originality, by reworking phrases and concepts from older poems and alluding to historical, geographical and socio-cultural points of reference, which were recognizable to his (educated) audience. For example, when court poet Su Ting was required to compose a verse on an “Imperial Visit in Early Spring to Princess Taiping’s Southern Villa”, he describes the Imperial family as having “spread gold”, and “supported a loom” – allusive references to the Buddha’s retreat and Qi Xi,3 an astrological phenomenon associated with meeting, respectively (ibid. 50). To illustrate the fact that Chinese-style allusion was favoured by the Kokin poets as a rhetorical device, we may note that the very same myth is taken up in poems 173 to 183 of the Kokinshū. In almost every case, the story is referred to obliquely, acting as a vehicle for the expression of pain caused by the transitory meeting, and subsequent separation, of lovers or friends. Poem 174 provides a clear example of this:


Topic Unknown


hisakata no

over the vastness of

ama no kawara no

Heaven’s River –


oh ferryman there on the bank –


should my love  cross to me,


would that you hide away your oars!


As elaborate word play, orthographic and semantic rebus came to be perceived as fundamental to courtly expression within the Six Dynasties salons,4 so were elegant witticisms, contrived syntactic constructions and double-entendre incorporated within the courtly poetry of the Kokinshū. Similar social functions fulfilled by poetry in both courtly societies – polite communication and a means of entertainment at social functions. In Japan, this extended to the formalized presentation of verses at competitions sponsored by the Empress – the so-called uta-awase, poetry from which features prominently within the Kokinshū (Rodd and Henkenius5).

The various epideictic devices that developed within these social contexts are described by several Japanese technical terms, the most commonly discussed being the jo no kotoba (‘prefatory phrase’), the kakekotoba (‘pun’), ‘or extended epithet’ and the engo (‘related word’). The latter two phenomena will feature later in our discussion.

Before moving on, it is worth noting that, thus far, we have attempted to demonstrate that the aesthetic norms which governed the composition of many Kokinshū waka drew heavily on Chinese example and often led to poetry being ornamented with arcane, and formulaic, rhetorical devices. However, this does not necessarily constitute evidence of their conventionality.

Brower and Miner suggest that conventionality in poetry arises out of an imbalance between the impersonal and the personal (21) – that is to say, an over-eagerness on the part of the poet to either fulfill preconceived social functions within the context of given aesthetic parameters or to enunciate his response to emotional stimuli. Too much of either, according to these critics, can result in insincerity and artifice. If this is the case, then even verse replete with shows of linguistic ostentation and courtly wit can remain free of conventionality, provided it maintains equilibrium by also giving voice to the poet’s concerns – be they particularistic or universal.

English literature abounds with examples of such poetry. Many commentators claim, for instance, that it was by means of “conceit” – contrived phrases and allusive language – that Donne and other metaphysical poets were able to discover “occult resemblances in things apparently unlike” (Wellek 100). Indeed, a brief examination of the first stanza of “A Nocturnal upon S. Lucy’s Day, being the shortest day” reveals Donne’s expertise at manipulating language in an elegantly witty fashion:


‘A Nocturnal Upon St. Lucys Day, Being the Shortest Day’ (Extract), John Donne

'TIS the year's midnight, and it is the day's,
Lucy's, who scarce seven hours herself unmasks;
The sun is spent, and now his flasks
Send forth light squibs, no constant rays;
The world's whole sap is sunk;
The general balm th' hydroptic earth hath drunk,
Whither, as to the bed's-feet, life is shrunk,
Dead and interr'd; yet all these seem to laugh,
Compared with me, who am their epitaph.

In “A Nocturnal”, Donne talks of the Sun sending forth “squibs”, which could refer both to “beams of light” or “altercations” and “sap”, which could be interpreted as “vital fluids” or, in other contexts, “to undermine foundations”. His use of the interrogative “whither” might have called to mind the verb “to wither”, tied semantically to the earlier “sap” (Donne 72). Thus, it can be seen that Donne employed techniques closely resembling the kakekotoba and engo, in an attempt to juxtapose – and relate – ideas, for the consideration of his reader.

There are also elements within the Kokinshū that seem to suggest poets’ purposeful manipulation of semantically pregnant language, to create a meditative atmosphere, or to raise questions regarding the nature of animate and inanimate states. Perhaps some of the best examples of this are to be found within the work of Ariwara no Narihira, whose poetry has been described as exuding a “deep emotionality, touched directly by human existence” (Aoki 16).

The following waka, for example, makes use of the kakekotoba, “nagame”, meaning both “Spring rains”, and “Staring, lost in thought” (Ozawa 254) to associate the restless emotions of the poetic speaker with the meteorological characteristics of Spring (Cranston, 66). Having set the tone of the poem as a response to a highly personal issue in the kotobagaki (prose introduction), Narihira evokes a hazy, synaesthetic image which precipitates contemplation of the relationship between emotional and environmental stimuli (i.e. sexual frustration and incessant, hazy drizzle).


Composed towards the beginning of the Third Month on watching the haze of drizzle – and sent to a girl to whom he had previously spoken fond words in secret.

Ariwara no Narihira


not sleeping, not waking

nemosedeyoru o

I lay ‘til dawn came,


and today I while away

haru no mono tote

– an object of spring? –


my stare long, long like the rain


Therefore, although much of the poetry of the Kokinshū is ornamented with contrived witticisms and, indeed, derives inspiration from Chinese example, it is precisely through the skilful manipulation of allusive and polysemantic turns of phrase (which modern critics find artificial), commonly understood by the poet and his/her audience, that it transcends superficiality and attain sushin (“meaningfulness”). Furthermore, while Kokin poetry undoubtedly contains conventional elements – a facet it arguably shares with all poetry – according to Brower and Miner’s definition, it should not be termed conventional in itself, because it maintains a balance between the personal and the impersonal.

Having established this, we are now well placed to progress to the next stage of our discussion – to determine the extent to which the early Heian waka contained in the Kokinshū may be described as lyrical and how, if at all, this relates to elements of socio-poetic convention that can be perceived within them.

Lyricism is, in the words of Lindley, “elusive of definition” (1). As it has already been pointed out, it was usual for modernists to view the lyric as a primarily individualist and unconventional genre, centred on an individual-specific, rather than society-based poetic persona. In doing so, they exclude from the lyric bracket all poetry that is impersonal or has been created to fulfill a specific social – and by extension, conventional – function. As noted earlier, there is considerable evidence to question this but, for the purposes of argument, we shall maintain the assumption at present and examine the nature of Kokin poetic composition accordingly.

We have already seen that much of the poetry contained in the Kokinshū was created as a means of social discourse and, primarily, functioned within the parameters of courtly aesthetic norms. However, this does not necessarily indicate that the poets’ creative objectives were obscured, nor that the flow of his emotions was in any way constricted. Many of the Kokin waka, particularly those from the several books of love poetry, can be appreciated as terse evocations of individual emotional experiences. In this period at least, dialogue through the poetic language of waka removed need for humble and honorific inflections of everyday court speech and so, it could be argued, facilitated a more genuine expression of the poet’s emotions than would have been possible otherwise (Brower and Miner7). At the very least, it offered the impassioned courtier or jilted lover a more emotionally expressive medium than Chinese, in which most official business was conducted, by virtue of being composed in their first language. The following love poem, particularly interesting for its entirely imagistic approach, demonstrates the depth of emotion present in some of the Kokin waka:


Topic Unkown

Fujiwara no Kachion

shiranami no

the vessel  traversing

atonaki kata ni

where not a trace of wake is left –


even she may rely

kazezotayori no

upon the winds


to mark for her a passage!


This waka incorporates highly intense metaphorical language to describe the emotional tumult associated with romantic frustration – language that is akin to the natural similes of Thomas Wyatt’s lyrical sonnets. It might even point towards the poet’s dissatisfaction with the abstracted notion of fate, as he muses on why “the winds” have failed to guide him safely through his amorous encounter.

Ariwara no Narihira’s work, as we have already noted, is particularly contemplative. It often evokes an apostrophic state of heightened emotional or philosophical perception, triggered by a particular external stimulus – reminiscent of Dostoyevsky’s ‘lyrical prose’ style, such as is to be found, for example, in his novelette, “The Gentle Spirit”. Narihira’s considerable talent and mastery of poetic diction notwithstanding, such is the intensity of emotion expressed in his work, (at the risk of sounding Poundian), it is even tempting to suggest that the natural parataxis of the Japanese language also played a role in the compositional process – in that can channel complex strings of utterance into a single syntactic stream in ways that other languages cannot.

In the following poem composed for an official’s birthday party by Narihira in 875, the falling of the cherry blossom seems to initiate an individual response from the poetic persona, causing him to muse on the transience of human life and, perhaps more interestingly, on the prospect that emotional intensity stimulated by one natural process might be able to alter the course of another:


Composed when the fortieth birthday celebrations of the “Horikawa” Minister were held at his villa on the Ninth Avenue.

Ariwara no Narihira


O, cherry flowers!


dance down to cloud

oiraku no

the path whence they say

komu to iunaru

comes age – that he may


never find his way there


Having shown that the Early Heian waka is both conventional and demonstrably lyrical by modernist standards, let us return finally to the original premise of our discussion – that the modernist hypothesis is a product of its time and place of origin and that is, therefore, inappropriate to apply it in other contexts.  As we shall now see, there is much evidence in pre-modern European sources to indicate that poetic individualism has never been synonymous with lyricism and that, in any case, the adoption of a conventional poetic idiom should not necessarily detract from either.

Without wishing to bore the reader with stories he or she will have heard time and time again, I feel it is necessary here to return to the source of our understanding of lyric (and many other generic categories) in the West – which is, of course, Ancient Greece. In the introductory section of to “Poetics”, Aristotle as follows refers to what has generally been understood as “Greek lyric poetry” simply as “κιθαριστικῆς”, verses composed to be sung to accompaniment of the harp-like cithara (Aristotle3). His choice of terminology is relevant to our discussion since the cithara – unlike the closely-related λύρα (lyre), from which, of course, the English word “lyric” is derived (Bowra, Greek Lyric Poetry3) – seems to have been performed exclusively by troupes of professional musicians, in places and on occasions of special social significance. An Athenian definition of lyric was, thus, not confined to issues of individual poetic personae, or to streams of consciousness. A discussion of characterizations of the genre in Greek discourse is not required here – but, since it is irrefutably the case that Greek lyrics often fulfilled specific socio-political, religious and collectivist functions and employed complex epideictic language, it is worth us mentioning briefly one concrete example of such poetry: Pindar’s Nemean Odes.

We know the Odes were commissioned by influential patrons and were occasional – created to fulfill specific social demands and not just to satisfy the poet’s individual creative impulse. Pindar throughout makes use of standard “lyrical and elegiac” imagery, punctuating his verse with witticisms. (Bowra, Pindar 241). This is evident, for example, in lines 48-53 of the Fifth Ode, in which Pindar plays around with phonological resemblances, such as athletai and Athanai (Race 53). It is worth quoting here:


ἴσθι, γλυκεῖάν τοι Μενάνδρου σὺν τύχᾳ μόχθων ἀμοιβὰν
ἐπαύρεο: χρὴ δ᾽ ἀπ᾽ Ἀθανᾶν τέκτον᾽ ἀθληταῖσιν ἔμμεν.
εἰ δὲ Θεμίστιον ἵκεις, ὥστ᾽ ἀείδειν, μηκέτι ῥίγει: δίδοι
φωνάν, ἀνὰ δ᾽ ἱστία τεῖνον πρὸς ζυγὸν καρχασίου,
πύκταν τέ νιν καὶ παγκρατίῳ φθέγξαι ἑλεῖν Ἐπιδαύρῳ διπλόαν
νικῶντ᾽ ἀρετάν, προθύροισιν δ᾽ Αἰακοῦ


Moving on (very quickly!), let’s think about the European Middle Ages – where songs called lyrics were sung by troubadours around Provence and elsewhere. These troubadour-lyricists, I would argue, remained true to the Pindaric tradition. Being professional poets, they sought primarily to impress their wealthy clientele and, in order to do so, aspired to the witty “manipulation of standard ideas” and the “artful deployment of conventional motifs” (Lindley 53). Although, superficially, their work may appear to evoke the emotions of a one individual (romantic love for a partner, loss and sadness after the end of a relationship, etc.), the reality is that it was generally occasional, intended to be performed in public – just as Heian uta-awase poetry had been – and made extensive use of conventional thematic motifs and prosodic devices. Thus, we see that modern understandings of generic categories like lyric – to which McCullough and others cling with such tenacity  – are “inapplicable to the poetry of the [European] Middle Ages” (Burrow 67-8).

A few hundred years later, again, in England, court poets such as Wyatt would seek to embellish simple “lyrical” song with elegant cadence, taking it “in the direction of wit and logical development”, to which their elite audience were able to relate (Lindley9). Wyatt, for example, was able to objectify even the most intensely personal of themes – such as the discovery of a lover’s unfaithfulness – by looking to neo-Platonic concepts such as “goodness” and “truth” to provide an “evaluative commentary of…[his] subjects” (Ricks and Michaels 363). Such philosophical quips are common throughout his oeuvre but a particularly good – and oft-quoted – example is to be found in the first few lines of his well-known ode, “A Revocation”, given below (Quiller-Couch 62):

What should I say?
―Since Faith is dead,
And Truth away
From you is fled?
Should I be led
With doubleness?
Nay! Nay! mistress.


As for the neo-classicist lyrical poets, we need say little here except that, naturally, they sought, as far as is possible in the vernacular, to emulate the work of the Ancients (Crane et al. 412) – and, just as much of that was highly conventional and far removed from the modernist conception of a personal, emotionally-charged lyric, so indeed was the product of its imitation. Indeed, the modernist definition, when applied to any or all of the pre-modern European lyric poetry we have surveyed, simply does not ring true. So, why then should we apply it to the poetry of Early Classical Japan, yet further removed from the intellectual conditions in which it was forged?

It is well known that Wordsworth and his early-Romantic contemporaries sought specifically to liberate English poetry from the yoke of superfluous artifice and, thereby, to create a sincere idiom for the exposition of the poet’s emotion (ibid. 411) but even they must necessarily have drawn on images familiar to the reader, in order to stir within him “a kindred feeling concerning them” (Shawcross 50). Thus, while the Wordsworthian lyric, as typified by the Lucy poems, captured beautifully the dreamlike flow of the individual poetic speaker’s contemplations, by virtue of being poetry – and therefore being heard by others – it could not transcend the framework of shared cultural experiences that tie a poet to his audience. And why? Postmodern critics, at least, would tell us that, in the Romantic lyric – as in any other – the “concept and status of an a priori ‘individual’ are always in question” since the language through which such an “I” is delineated is necessarily contiguous with, and subject to, “a pre-existing system that at once socializes and individuates it” (Blasing5).

We can therefore conclude that, although a lyric standard was undoubtedly created in the late eighteenth and early nineteenth centuries – a standard to which critics have inappropriately compared a great deal of poetry from other periods and regions –Romantic lyrics themselves scarcely live up to it. Or rather, the lyrics that have been best received in the literature do not. Romantic poets with an overzealous commitment to the “Art for Art’s Sake” cause, or else a predilection for intense, introverted personalism – like a few of the contributors to the Kokinshū – run the risk of seeing their work indicted as affected and conventional in its own right, according to Brower and Miner’s definition at least. It is most likely to poems such as these that Blasing is referring to when she talks of “cultivated, induced, pathological, or ‘deviant’ irrationality…which certain poetic practices may invoke” (2).

Definitions of conventionality and lyricism within poetry are multifarious – to the point that their application to any specific works is very difficult. However, as we have observed, the two concepts are not contradictory as the post-late-eighteenth-century paradigm suggests and, indeed in many instances, seem to enjoy a symbiotic relationship. While Early Heian waka undoubtedly exhibits many conventional elements and considerable borrowing from older Chinese poetic forms, this, as we have seen, is by no means a bar to lyrical expression.

So how are we to understand our understanding of the lyric? Having put modernist assumptions aside, we are left, essentially, with the Classical definition – that is to say a poem which is not epic or dramatic and which, in antiquity, was sung to the accompaniment of a plucked stringed instrument. But that gives us very little to go on. It is the opinion of the author that mainstream criticism in the late twentieth and early twenty-first centuries has been heading in the right general direction when it has sought to prove that the “individual” that they associate with the lyric is a linguistic construct – only real in that context – and that the lyric is not merely written in poetic language but that it actually is poetic language, language which“…keeps in view the linguistic code and the otherness of the material medium of language to all that humans do with it”(ibid.).

Through our discussion, we have shown that we might be justified in taking this line of thinking one step further and eliminating the “individual” altogether. In so doing, we reduce the lyric to a kind of “poetic vernacular”, system of linguistic signification which can be simple or, in some instances, more arcane, but one which is necessarily comprehensible to the poet and audience in such a way as to bring about the metagenesis of additional referents, which might be entertaining, thought provoking, emotionally charged, socially critical, or a combination of the above. If such a vernacular exists, there can be little doubt that the Kokin poets were anything less than highly conversant in it.




[1]. The collection’s full title is Kokinwakashūbut I have abbreviated it throughout to Kokinshū, according to Japanese convention.

2. Taking inspiration from Romanticism, Marxism, Imagism, Impressionism etc.

3. Japanese: Tanabata –  lighting conditions cause the Milky Way to appear dimmer than usual, hence creating the impression that two lovers on either side of it (the stars Altair and Vega) are able to meet and consummate their affection.

4. Perhaps the best known examples of Chinese poetry which employ such word play are the palindromic epigrams of Xiao, dismissed by many critics as little more than prandial frivolity. See: (McCullough 61).


Works cited


Aoki, Takako. Chūko no Kajin [Waka poets of the Classical Period]. Tokyo: Nihon Koten Bungaku Kōza, 1960. Print.

Aristotle, and Donald William Lucas. Poetics. Oxford: Clarendon Press, 1978. Print.

Blasing, Mutlu Konuk. Lyric Poetry: The Pain and the Pleasure of Words. Princeton, N.J.: Princeton University Press, 2007. Print.

Bownas, Geoffrey, and Anthony Thwaite. The Penguin Book of Japanese Verse. Harmondsworth: Penguin, 1964. Print.

Bowra, Cecil. Greek Lyric Poetry: From Alcman to Simonides. Oxford: Oxford University Press, 1961. Print.

Bowra, Cecil. Pindar. Oxford: Oxford University Press, 1971. Print.

Brower, Robert, and Earl Miner. Japanese Court Poetry. Stanford: Stanford University Press, 1961. Print.

Burrow, John Anthony. Medieval Writers and Their Work: Middle English Literature and Its Background 1100-1500. Oxford; New York: Oxford University Press, 1982. Print.

Crane, Ronald Salmon, et al. Critics and Criticism: ancient and modern. Chicago & London: The University of Chicago Press, 1952. Print.

Cranston, Edwin. A Waka Anthology, Vol.2: Grasses of Remembrance. Stanford, Stanford University Press, 2006. Print.

Culler, Jonathan. Structuralist Poetics: Structuralism, Linguistics and the Study of Literature. London: Routledge, 1975. Print.

Donne, John. The Complete English Poems. Harmsworth: Penguin Books, 1997.

Foucault, Michel. The Order of Things: An Archaeology of the Human Sciences. London: Tavistock Publications, 1970.Print.

Hagiwara, Sakutarō. Hagiwara Sakutarō Zenshū [Collected Works of Sakutarō Hagiwara].Vol 7. Tokyo: Chikumashobō, 1976. Print.

Itasaka, Gen, ed. Kodansha Encyclopaedia of Japan. Vol. 4. Tokyo: Kodansha, 1983. Print.

Johnson, Walter Ralph. The Idea of Lyric: Lyric Modes in Ancient and Modern Poetry. Berkeley; London: University of California Press, 1982. Print.

Lindley, David. Lyric. London and New York: Methuen, 1985. Print.

Masaoka, Shiki. Yomini atauru sho [Letters to a Waka poet]. Tokyo: Iwanami shoten, 1955. Print.

McCullough, Helen. Brocade by Night: Kokinwakashū and the Court Style in Japanese Classical Poetry. Stanford: Stanford University Press, 1961. Print.

Morris, Mark. "Waka and Form, Waka and History" Harvard Journal of Asiatic Studies 46.2 (1986): 551-610. Print.

Okada, Richard. "Translation and Difference – A Review Article" The Journal of Asian Studies47.1 (1988): 29-40. Print.

Ozawa, Masao, ed. Kokinwakashū [Anthology of Poems Ancient and Modern]. Tokyo: Shōgakkan, 1971. Print.

Quiller-Couch, Arthur Thomas, ed. The Oxford Book of English Verse: 1250-1918. Oxford: Oxford University Press, 1939. Print.

Race, William, trans. Pindar II: The Nemean Odes, Isthmian Odes, Fragment. Cambridge MA: Harvard University Press, 1997. Print.

Ricks, Christopher B., and Leonard Michaels. The State of the Language. Berkeley: University of California Press, 1980. Print.

Rodd, Laurel, and Henkenius, Mary, trans. Kokinshū: Collection of Poetry Ancient and Modern. Princeton: Princeton University Press, 1984. Print.

Shawcross, John, ed. Coleridge’s Biographia Literaria. Vol 1.Oxford: Oxford University Press: 1904. Print.

Shirane, Haruo. Japan and the Culture of the Four Seasons. New York: Columbia University Press, 2012. Print.

Wellek, René. Concepts of Criticism. New York & London: Yale University Press, 1963. Print.


Unless otherwise stated, all translations from Japanese are the author’s own. Translations of the Kokinshū follow the Nihon koten bungaku Zenshū edition of the text.


L Mann

Laurence Mann is reading for a DPhil in Japanese Literature, supported by the Wolfson Foundation. As an undergraduate at Oxford, he was the recipient of several scholarships and award, and a Gibbs Prize, for outstanding performance in examinations. His doctoral research focuses on the poetics of Shinto liturgies and he is currently preparing a forthcoming paper entitled “Orality in the Engishiki Norito”. He is also a project member of the Oxford Corpus of Old Japanese.