Translation and Difference--A Review Article

Brocade by Night: "Kokin Wakashū and the Court Style in Japanese Classical Poetry. by Helen
Craig McCullough; Kokin Wakashū: The First Imperial Anthology of Japanese Poetry (with
"Tosa Nikk" and "Shinsen Waka"). by Helen Craig McCullough
Review by: Richard H. Okada
The Journal of Asian Studies, Vol. 47, No. 1 (Feb., 1988), pp. 29-40
Published by: Association for Asian Studies
Stable URL: http://www.jstor.org/stable/2056358 .
Accessed: 18/06/2014 21:12

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Association for Asian Studies is collaborating with JSTOR to digitize, preserve and extend access to The
Journal of Asian Studies.

http://www.jstor.org

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
 Translation and Difference
A Review Article
RICHARD H. OKADA

THROUGHOUT HER CAREER Helen McCullough has enjoyed a solid reputation as

a meticulous translator of premodernJapanese texts. She has, on the whole, chosen
to deal with material that leans toward the historicoliterary, rather than the more
strictly literary, spectrum. In Japan such texts are commonly known by the terms
gunki monogatari,"tales of war," or rekishimonogatari,"tales of history. " Her numerous
publications include a translation of Gikeiki, a text depicting the life of the tragic
warrior Minamoto Yoshitsune (1159-89) and more recently, in conjunction with
William McCullough, a rendering of Eiga monogatari,a narrative history of tenth-
and eleventh-century Japan.
By 1968 McCullough had turned her considerableenergies from chronicles writ-
ten in prose to a mid-tenth-century text of poetry plus prose, Ise monogatari,an uta
monogatari,"tales of poems." An exceedingly difficult text to classify in terms of its
literary nature and its status in Heian letters, IseaffordedMcCullough the opportunity
to tackle the thirty-one-syllable tanka and its development in Heian society. Of the
209 poems (according to some versions) that constitute the Ise text, 62 appear in
Kokinshz7 (or, as it is known by its full title, Kokinwakashui,"Collection of Japanese
poems old and modern").
Now, seventeen years later, McCullough has again focused her attention on the
Heian waka. She has produced not only a complete translation of Kokinshfi(ca. 905),
the second to appear in two years (see Rodd and Henkenius 1984), but also her own
version of Tosa nikki (ca. 935), Ki no Tsurayuki's diary written in the persona of a
woman, as well as the first English translation of Shinsenwaka (ca. 940), a collection
of 360 waka-282 of which are taken from Kokinshz7-personallychosen by Tsurayuki
in accordancewith an imperial command.

RichardH. Okadais AssistantProfessor,Departmentof EastAsian Studies, Princeton

University.
Brocade byNight.'"KokinWakashz" andtheCourtStyleinJapanese ClassicalPoetry.By Helen
CraigMcCullough.Stanford:StanfordUniversityPress, 1985. xii, 591 pp. $55.00.
KokinWakashz7.' TheFirst ImperialAnthologyofJapanesePoetry(with "TosaNikki" and
"Shinsen Waka").Translatedand annotatedby Helen CraigMcCullough.Stanford:Stanford
UniversityPress, 1985. x, 388 pp. $49.50.
The authorwishes to thank the followingfor their assistancein the publicationof this
article:AnthonyChambers,Norma Field, EdwardFowler,HarryHarootunian,MasaoMi-
yoshi, Tetsuo Najita, William Sibley, and all those signersof a petition presentedto the
editorialboardof theJournalat the Boston meeting of the Associationfor Asian Studiesin
April 1987.
TheJournalof AsianStudiesVol. 47, No. 1 (February1988):29-40.
?) 1988 by the Associationfor Asian Studies, Inc.

29

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
30 RICHARD H. OKADA

The 1, 11 1 poems in Kokinshz7 represent the first imperially commissioned waka

anthology in Japan. The appearanceof the anthology in the early years of the tenth
century was an event of the utmost significance for the Japanese literary tradition,
displacing as it did the three previous imperially commissioned kanshi(poetry written
in Chinese) collections. It also heralded the new prestige of the tanka form and pro-
vided not only source material for writers of later texts but also a standard of poetic
composition for centuries to come. The study of Japanese literature in the west has
long needed a readily available English version of Kokinshz7, and it is indeed fortunate
that there are two to choose from now.
McCullough's translation itself is, for the most part, carefully executed. As the
Ise volume demonstrated, her English versions of waka, in contrast to the prose of
Yoshitsune,may not always read fluently, much less poetically, but they do display
a kind of clinical accuracy in that most of the elements in the Japanese are closely
attended to. McCullough chooses, as Rodd-Henkenius did, to employ a 5/7/5/7/7
English syllabic pattern. Use of the pattern helps to maintain economy of expression
and gives the reader some sense, over a long series of poems, of the rhythms of the
collection. A few examples must suffice. KKS 192 is characteristicof McCullough's
style:
sayonaka to The deepening night
yo wa fukenurashi seems to have come to midpoint,
kari ga ne no for the moon poises
kikoyuru sora ni at the zenith of the sky
tsuki wataru miyu echoing wild goose calls.
Although the English is accurate, the words "poises" and "zenith" and the awk-
wardness of the last line create a mechanical effect that is at odds with the poet's
realization at the cries of the geese.' KKS 474 is more successful, although "absolute
stranger" has the ring of cliche and is probably an overreading of yosonite mo:
tachikaeri Over and over,
aware to zo omou like white waves from the offing,
yoso nite mo my fond thoughts return
hito ni kokoro o to an absolute stranger
okitsu shiranami who has carried off my heart.2
Other examples of less than felicitous wording are "ripped into shreds" (23), "Must
you refuse again / to grant us satiety?" (61), "Brown leaves" (320), "going island-

'Comparethe Rodd-Henkeniusversion, which more closely capturesthe spirit of the

Japanese:
as night deepens it
seemsmidnight must be near for
only the cries of the
wild geese are heard and the moon
has climbedhigh in the darksky
2 The Rodd-Henkenius
versionspecifies,perhapsunnecessarily,the persona:
I turn back to her
as the white wavesroll to shore
fromthe offing no
matterthe distance my heart
is held captiveby her charms

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
 TRANSLATION AND DIFFERENCE 31
hid" (409), "I exist, it seems, I only that whenevening falls" (815), "rice ear" (822),
and "in deep mountain fastnesses"(954). Problems of interpretation are at issue in
such poems as KKS 366, in which "journeying where" might have been something
closer to "journeyingfrom," and in KKS 497, in which "show my colorsI as blossoms"
might have been "show colorsbrightas blossoms."
There are occasions when attempts, valiant as they are, to match the puns in
Japanese produce English that borders on the unintelligible, as in KKS 649:
kimi ga na mo Let us keep gossip
wa ga na mo tateji away from your name and mine.
naniwa naru You must not mention
mitsu to mo iu na the "seeport"at Naniwa,
aiki to mo iwaji nor I the "met" net fishing.
The note provided for the poem helps to decipher the wordplay, but the translation
itself cannot be called a success. There are other cases, such as KKS 307, when a
different metrical configuration might have prevented the somewhat comical ambi-
guity of referent (here, of "its"):
ho ni mo idenu As I stand guard
yamada o moru to in the mountain paddy field,
fujikoromo its ears still unformed,
inaba no tsuyu ni my coarse robe is wet daily
nurenu hi wa nashi by dewdrops from the rice plants.
One wishes that both McCullough and Rodd-Henkenius had used a system of
romanization approximating the orthography of late ninth- and early tenth-century
Japanese, which would have preserved the numerous instances of phonological and
orthographic ambiguity and play (KKS 376, or the extreme case of KKS 955, where
a different system would have allowed the reader to see that no two syllables are
repeated). One wishes also that McCullough had provided more supplementary notes
to allow the readerto follow a greater number of the wordplays and the crucial links
from poem to poem. Finally, both translators, in their 5/7 patterns, largely ignore
the phenomenon of ji amari, lines containing 6 or 8 syllables.
Difficult as it may be to translate Heian waka, however, it is even more difficult
to discuss them, as the second volume under review demonstrates. In what is a signal
departure from the usual "introduction-plus-translation"format one finds in other
English translations of Japanese texts, McCullough has vastly expanded the remarks
made in the introduction to her earlier study of Ise into a separate, 591-page com-
panion volume to the translations, thereby creating a major event forJapanese studies.
As she did in her work on Ise, the author continues to derive most of her method-
ological perspectives from the work of the Japanese scholarKonishi Jin'ichi, especially
his 1959 article on Kokinshui, which she later translated(Konishi 1978). In that article
Konishi argued that it was the Six Dynasties style, especially the style which appeared
at the Liang court, that exerted the major influence on Heian poets (Konishi 1978:66).
In McCullough's words in BrocadebyNight: "It is to the late fifth, sixth, and seventh
centuries, when aristocratic courts were the centers of Chinese poetic activity, that
we can most profitablylook for correspondences.. . . The current tendency in Japanese
scholarly circles is to focus on those centuries, and we shall do likewise" (p. 9).3 Her

3This is not quite the case. RecentJapanesescholarship(for which no citationsappear

in BrocadebyNight),in additionto its ongoingstudyof the SixDynastiesperiod,hascontinued

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
32 RICHARD H. OKADA

present study can be seen as a faithful acceptance of Konishi's proposal: "The whole
matter of Chinese influence on Japanese literature requires new investigation; and
one of the foundation stones of such researchmust be the identification of whatever
Chinese elements are present in the Kokinshz"(Konishi 1978:170).
A large part of Brocadeby Night consists of just such an attempt to identify
"Chinese elements." McCullough discusses a variety of texts from both the Chinese
and Japanese traditions and supports her points with an impressive quantity of ma-
terial. She includes useful information on such texts as the Chinese poetry anthologies,
the early Heian kanshicollections, Kudaiwaka, and Shinsenman'yoshz7. Even specialists
who can go straight to the Japanese sources will find it an accessible if cumbersome
reference work. It is clearly not for the general reader, however.
This reader was struck by two aspects of the companion volume in particular:
first, the sheer number of poems, most of them translated by the author herself, given
as illustration. With over 50 percent of the 1,111 Kokinshuipoems reappearing in
BrocadebyNight, what McCullough has produced is, in fact twovolumes of translation.
The emphasis on translation is not coincidental, but an important part of the author's
methodology. Second, despite its breadth of coverage and its tone of reasonableness,
Brocadeby Night is filled with perplexing statements, and the work seems founded
on questionable assumptions concerning history and literature. Indeed McCullough's
remarks often mystify, rather than clarify, such broader issues as the complexity of
intercultural relations-particularly such issues as "influence" and literary value.
They also raise further questions concerning the role of a translator-interpreterof a
non-western culture and the discourses(including the never-innocent act of translation
itself) that characterize the resulting interpretation. Although detailed coverage of
all the issues raised is beyond the scope of this review, the size of the study, written
by a scholar of impeccable credentials, makes it imperative to address at least some
of the more important ones squarely. Let it be said here that, in view of the author's
long and productive career, I hesitate to criticize what is obviously the result of tireless
labor; it is in the hope of contributing to an alternativeapproachto Japanese literature
that the following comments are.offered.
In a brief introduction attached to chapter 1, "The Chinese Heritage"-there is
no independent introductory chapter-McCullough provides the raison d'etre for
Brocadeby Night: "It is now widely agreed that an adequate appraisal of the artistic
accomplishments of the Kokinshz7 poets and compilers requires, at the very minimum,
a knowledge of Chinese poetry and its influence on Japan; a familiarity with Heian
aesthetic conventions; an awareness of the aims of poets and compilers, and of the
expectations of their intended audience; and an acquaintance with the workings of
Heian society" (p. 5). Aside from the implication that such knowledge is readily
available to a researcher-translator,the phrase "appraisalof the artistic accomplish-
ments"4 stands out to this reader: the goal of the study appears to be evaluation

to focuson the poets of the earlyand mid-Tangperiod, and especiallyon Bo Juyi, who, in
McCullough'swordsin Brocade byNight(againfollowingKonishi),"hadno significanteffect"
(p. 9) on Kokinshz.KojimaNoriyukiis one notablescholarwho, in his voluminouswritings,
hasalwaysemphasizedthe roleplayedby Tangpoetry.Japanesescholarshavealsoincreasingly
scrutinizedthe nativeJapanesetradition,both kanshi(poemswrittenin Chinese)and earlier
waka-McCullough's discussionof the nativekanshitraditionis skewedin this regard.See
Bungaku1985:vol. 53, no. 12, a specialissuedevotedto essayson the relationbetweenkanshi
and Kokinshz7.
4 Cf. her 1968 statement:"Howoriginalis even the bestJapanesecourtpoetry?Without

some notion of the extent of Chineseinfluencein all these matters[originality,socialrole of

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
 TRANSLATION AND DIFFERENCE 33
(scarcely present in Konishi). In other words, McCullough is interested in an act of
judgment-"appraisal," which carries a familiar connotation of "connoisseurship."
One might question the wisdom of this preoccupation with determining literary
value, an act that is nowadaysperformed, if at all, with great caution even in studies
of western literature-not to mention in studies of a culture for whose members the
question of "accomplishment" was consistently dispersed within the broader, inter-
personal matrices of group effort and communal reception.
In addition to following Konishi, McCullough adheresclosely to George Sansom's
classic description of Heian society as governed by a "rule of taste": "Here we may
merely note that the famous 'rule of taste' was an aspect of the moral code, and that
it dictated, among other things, the nature of the Heian approach to poetic compo-
sition" (pp. 2-3); "Muchof this book will be devoted to determining what constituted
a good poem, which is to say a tasteful one, in early Heian eyes" (p. 4). To characterize
a society as having evolved a "rule of taste" is one thing; it is quite another meth-
odologically to determine, much less adopt, as the author does, such a criterion in
one's own literary and cultural exegesis or, in this case, appraisal.
McCullough is also concerned with the question of autonomy: "Kokinshzu repays
serious critical attention, both as a repository of individual poems and as an inde-
pendent literary creation" (p. 6). But such terms as "individual" and "independent
literary creation" (also "repository,"which denotes "museum," "vault," or "tomb")
suggest a preoccupationwith what might be called essentialist or monolithic concerns
that are not completely congruent with a text produced in tenth-century Japan.5
FurthermoreMcCullough adopts a problematic view of historical development: "We
need to understand not only what the rule of taste entailed, but also how it evolved"
(p. 5); "When Man'y&shzu, Kokinshz7, and the literary activity of the period between
the two anthologies are considered as a whole against the backdrop of the Chinese
literary tradition, it becomes apparentthat the entire period from the seventh-century
Omi court to Kokinshfiwas one of gradualadaptationand assimilationof importedmodels,
and that Kokinshzu must be viewed not as a sudden, radical departure from unsullied
ancient norms, but rather as the culminationof 250 years of experimentation with
Chinese poetry and Chinese aesthetic concepts" (p. 7; emphasis added). Such state-
ments, with an overemphasis on China, suggest a view of history and literature that
pays insufficient attention to the active side of cultural interplay in which differences
can only culminate in ever-changing instabilities. It also neutralizes the differences
between an earlier tradition, whose originary "norms" are hardly ever "unsullied"
(whether or not Kokinshz7 represents a radical break or a "culmination"), and later
interpretations of that tradition-whether Heian Japanese or contemporary western.
To view history as progressing in a neat, linear fashion "of gradual adaptation and
assimilation" is to succumb to the seduction of teleology and to ignore forces of
"discontinuity" and "disruption."6

poetry,Japanesevalues,etc.], one can neitherappraisethe poetryin Talesof Isenor properly

understandthe culturalmilieu fromwhich it emerged"(McCullough1968:14).
'Chapters 6 and 7 are called, "The 'Kokinshui'Style and the Individual Voice" and
'Kokinshui' as LiteraryEntity," respectively.
6 See the "Introduction"by TetsuoNajitain Najitaand Koschmann1982:3-21. Najita's
essay containsthis remarkby HaydenWhite: "The Historianservesno one well by con-
structinga speciouscontinuitybetweenthe presentworld and that which precededit. On
the contrary,we requirea historythat will educateus to discontinuitymorethaneverbefore;
for discontinuity,disruption,and chaosis our lot" (p. 6). Otherwriterswho haveaddressed
the issue includeMichel Foucault(1976:pt. 4, chap. 5) and ThomasKuhn (1962:2-3 and
chap. 9).

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
34 RICHARD H. OKADA

The task of appraisalraises first and foremost the question of criteria: "Goodtaste
[for the Six Dynasties poets] required. . . urbanity, wit, and a tactful reluctance to
let political problems or deeply felt personal concerns intrude" (p. 63; emphasis
added). To "urbanity" and "wit" McCullough adds "allusion" (p. 64), "figurative
speech" (p. 66), "metaphor" (p. 67), "feigned confusion" (ibid.), "reasoning tech-
nique" (p. 68), "personification"(p. 70), and "aestheticism"(p. 130). The problem
of terminology quickly arises. How, for example, is "figurative speech" to be dis-
tinguished from "metaphor"?And terms like "urbanity," "wit," and "aestheticism,"
with their genteel ring, are apt to be slippery and falsely comforting; they can be
made to apply to practically any Kokinshuf poem. It is unusual to find them the focus
of a study at a time when scholars have been attempting to achieve precision in the
critical reading of literature. The use of such imprecise terms gives a writer tremen-
dous leeway in their application but it also puts that writer in the privileged position
of being as much an arbiter of "taste" as the society being described.7
McCullough then proceeds to identify influence. A sampling of the type of state-
ments in which the study abounds: "It seems best to think of it [KKS 781] as an
imitation of the Chinese palace style" (p. 215); "The rest of Shinsenman'yoshualso
strongly reflects the influence of Chinese poetry in general and the Liang salons in
particular"(p. 266); "As might be expected, all four [of the following poems] exhibit
the characteristictraits of the Chinese court style" (p. 288); "Manyof his [Tadamine's]
waka ... show conspicuous Six Dynasties influence" (p. 391). She even finds that
Kokinshupoems can be grouped statistically on the basis of detectable influence: "Most
love poems use the Man'yo style of direct affirmation(and some the jo [introductory
segment]), but Chinese influence is strong in approximately one-third" (p. 189).
McCullough's discourse on influence is worth a closer look. First, what is the
larger significance of having identified Chinese influence if not simply as a means of
judging the level of development of Japanese poetic skill or success of a particular
Japanese poem? The conflation in McCullough's discourse of the search for influence
and the goal of evaluation turns into a procedure by which the extent of perceivable
influence becomes the standardused to assess the worth of the poetry. As noted above,
however, Heian waka almost never occupied isolated moments of composition but
arose out of specific, although changeable, contexts. Second, and more important,
she implies not only that Chinese elements are importable but also that they remain
distinctly identifiable, in their reified forms, over space and time and across languages
and cultures. Yet recent studies of language and meaning, revitalizing the seminal
work of Ferdinand de Saussure, have tended more and more to emphasize the dia-
critical nature of signification. Linguistic concepts, in the words of one critic, are
"purelydifferential . .. determined not by their positive content but by their relations
with the other terms in the system" (Belsey 1980:40). In other words, the terms or
signifiers of a given linguistic system do not possess any essential properties or values
("positive content"); they generate meaning only as they relate to and differ from
other terms or signifiers of that system. When two or more cultures are in question,
the relations between them can then be problematized as each is granted its own
temporal and spatial integrity as a linguistic or cultural system of differences. "Chinese

7See Konishi 1978:71. The devices,whicharenot evaluativetermsin themselves,derive

in large part from Konishi'suse of the Chineseterm i-bang,"oblique,"found in Chinese
texts, to characterizethe "indirect"style of Six Dynastiespoetry. Such designationswere
alreadyproblematicin Konishi'sessay:all poetic language,of course, is indirect in some
fashionor other.

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
 TRANSLATION AND DIFFERENCE 35
elements," for example, are not natural objects but must be seen as part of a network
of relations within a specific culture in a given historical epoch. Attention to such
issues might have shown McCullough the relevanceof recent notions of language and
meaning production for the classical Japanese linguistic and literary situation as it
confronted the Chinese examples, notions that might also have allowed her to describe
more compellingly the Japanese literary scene, especially the supremely "relational,"
fluid contextualizations of Heian waka (of which the Kokinshzi anthology is an excellent
example). It would also have allowed a differentperspectiveon the searchfor influence.
McCullough's treatment of influence in BrocadebyNight demonstrates the hazards
of intercultural interpretation: it leads to a consistent valorization of China (the men-
tor) and devaluation of Japan (the pupil)-"We should not be surprised that the
Kaifzisiuse of parallelism [does not achieve) the urbanity of the Chinese court style"
(p. 93). Such implicit sinocentrism prevents the native Japanese tradition from man-
ifesting itself as anything but direct, derivative, primitive, marginal, and even "cheer-
ful,"8 of which the phrase "unsullied ancient norms," cited above, may be taken as
emblematic. According to McCullough, the early (pre-Man'yo)poetry was direct:
"Much more common, however, is the direct, straightforward mode of expression
found in folk literature"(p. 121). Among the Man'yjshzu poems themselves, she finds
some with and some without Chinese influence. The following reveals the author's
general attitude toward Man'yjshf: "We may dispense with most such considerations
[understanding the rule of taste) in the case of Man'yoshui,an anthology that is ac-
cessible to the modern readerbecause it is, to a large extent, the product of a different
and more readily intelligible poetic tradition" (p. 5). This is a remarkablestatement.
As the innumerable volumes of commentary and centuries of scholarship devoted to
them testify, the 4,500 poems in Man'yJshz7 are anything but "accessible"or "readily
intelligible" to modern readers;it is the author's presupposition of evolutionary de-
velopment (and the power of translation?)that makes them so. Finally, by her account,
Kokinshgrepresentsthe culmination of a process: "A deep gulf separatesthe two [early
Japanese song and Kokinshf);for one is, after all, the product of an illiterate, isolated
society and the other that of an advanced civilization heavily indebted to borrowings
from abroad" (p. 85).9 McCullough does seem to believe that the early, "native

"Just how "cheerful"(p. 177) or "happy"arepoemsin Man'y&shk? Let us examineMYS

2421, which McCulloughcites as an exampleof "happylove"(p. 122): "I hope there will
be / no hills with rocks to treadon / along the way:/ what if the horse should stumble/
bearingthe one I await?"Dependingon the context, the poem could easilybe readin other
than "happy"ways:as an expressionof anxiety(line 4), loneliness("Iawait"),frustration(yet
anotherexcusefor a lover'sabsence),etc. By the sametoken, althoughshe states that most
love poems-love being one identifiablenative topic-are expressionsof painful love, the
poems she cites containlittle in themselvesto suggestpain (see, for example,the discussion
of MYS 2680 [p. 143)). McCulloughalso ignoresthe implicationsof her own conclusions
thatthe laterMan'yJpoemsareindistinguishable,in manycases,frompoemsfoundin Kokinshui
(ibid.). The deeplycontextualnatureof wakaand the problemof poetic persona-that is to
say, the fictivenatureof Japanesepoetry-must be kept in mind when analyzingindividual
poems.
9 On influence,see MichaelBaxandall:" 'Influence'is a curseof art criticismprimarily
becauseof its wrong-headedgrammaticalprejudiceabout who is the agent and who is the
patient.... To think in terms of influenceblunts thought by impoverishingthe means of
differentiation.... Worse it is shifty. To say that X influencedY in some matteris to beg
the questionof causewithout quite appearingto do so." In short, it is better to regardY,
whetherit be Picasso(vis-'a-visCezanne)orJapan(vis-'a-visChina),as activelyrespondingto
specificcircumstancesand makingan "intentionalselectionfroman arrayof resources"(Bax-

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
36 RICHARD H. OKADA

tradition" constituted part of the "influence" on Kokinshfu, but her with/without

scheme of analysis precludes the possibility of confronting it directly. 10
Taken together with her concern for appraisal, McCullough's inscribing of China
and Japan into what may be termed a primary/secondaryor central/marginalpolarity
is disturbing. It suggests that Japan is unable to stand on its own as a literary culture.
Instead, the Chinese tradition provides the requisite standardsfor ennobling the de-
rivative culture and serves to mediate our understanding of Japan. But Chinese tra-
dition too needs to be translated and the appropriateelements selected and applied.
The elements, once found, were designated by the author using terms impossible to
define. The final "selector" thus always becomes the translator herself, and her own
notions of "taste" end up having been unquestioningly imposed on the subject of
study from the beginning.1
Concomitant with the search for standards of taste and influence is the author's
focus on individuality and originality (or their absence). Throughout, she attempts
to locate evidence of "an author'sunique genius" (p. 368): Henjo is "urbane, elegant,
witty-and, within safe bounds, original" (p. 217), Narihira makes "convincing
lyrical statements" (p. 208), and "Komachi is thus a poet capable of remarkable
resourcefulness,originality, and subtlety, a distinctive voice in her own era"(p. 230).
This quest for originality seems to run counter to a poetic form that depended heavily
on context and convention and to poets who respondedin regulated ways to practically
ritualized situations, whether public or private. That is not to say that syntactic
patterns cannot be perceived (but what does it mean to say that a poet is original
"within safe bounds"?) or that they should not be looked for, but a constant search
for originality goes against one's sense of the artistic and textual environment of the
Heian period, which gainsays any approach centered on conceptions of individual
authorship or on such terms as the "lyrical." As the very act of anthologizing dem-
onstrates, poems were freely appropriated, repeated, revised, recopied, and recon-
textualized. 12 Although McCullough devotes one chapter to the structure of the an-
thology, that discussion is so marked by her typical search for the superior and the
inferior and for norm and deviance that the open-ended, playful, inter- and intra-
discursive movement of the Heian poetic language, as well as the inventive manip-

andall 1985:58-59). Baxandall'scommentsareequallyapplicableto literaryhistoryandcrit-

icism. Such a notion as "intentionalselection"would allow one to begin to describemore
accuratelythe Japanesepoetic impulseitself and to grapplewith such questionsas why the
Japanesechosewhat they did from the Chineseliterarytradition.
10The questionof the vital role playedby the newlydevelopedkanawriting systemthat
enabledHeianpoets to write freelyin Japanesein the firstplaceand the questionof how that
act of writing relatesto native transformations of Chinesepoetry are only given a passing
comment by McCullough(p. 172). Aside from descriptionsof "auditoryeffects,"she also
neveraddressesthe criticalquestionof the "oral"natureof wakapresentationand reception.
" See the symposiumon EdwardSaid'sOrientalism in TheJournalof AsianStudies34, no.
3 (May 1980):485-517.
12As McCulloughherselfstates, "By and large, if a poet was good enough to gain ad-
mittanceto Kokinshi,his style was, by definition,basicallyanonymous"(p. 285). She con-
cludes, "In a sense, therefore,we must returna negativeanswerto the basic question we
raisedat the beginningof the chapter['Canone competentKokinshu- poet really be distin-
guishedfromanother?')" (p. 418); the "negativeanswer"one findsheresuggeststhat different
questionsmight havebeen askedmoreproductively.On McCullough'suse of the term "lyr-
ical": KKS 411, the famouskarakoromo poem by Narihira,is called a "convincinglyrical
statement"because"the artificeis so well concealed"(p. 208). One of the most obviously
contrivedin the whole collection,KKS411 is perhapsone of the least lyricalpoemsone can
imagine. For McCulloughto call its artifice"well concealed"is to displaya bold disregard
for the workingsof rhetoricallanguage.

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
 TRANSLATION AND DIFFERENCE 37
ulation of waka by the Kokinshucompilers, is barely discernible. What the study of
Heian waka needs is not an approachbased on influence and evaluation but rather a
sustained examination of what it meant, in literary, historical, and political terms,
to compile poetry collections-an act that was repeatedagain and again in premodern
Japan-and how that act related to the question of individual/social waka compo-
sition. A closer examination of such questions might have prevented McCullough
from bemoaning the fate Mitsune and others receive in Shinsenwaka, where Tsura-
yuki's "editorial policies ... have resulted ... in the virtual disappearanceof an au-
thentic individual voice . . [and) his misrepresentationof two earlierpoets, Komachi
and Narihira, is truly astonishing" (p. 531).
The search for originality also goes against modern-day critical practice in the
west. Such notions as "originality" and, by extension, "authorship"have been vig-
orously 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. In order to avoid the ethnocentric pitfalls that accompany unexamined uses
of terminology, it would help to keep in mind the relation of writers and texts to
the forces of power and control at work in any given society.13 The concern for
originality and individuality ultimately does little more than satisfy a need for single,
determinate meanings, for guarantees of "truth," which in turn are necessaryfor the
act of appraisal.The Heian tradition itself-based on Shinto and Buddhist worldviews
that emphasize a social nexus and do not posit one supreme deity-as well as modern
critical procedures point in a different direction.
The preoccupation with standardsand the binary thinking thereby produced-
foreign/native, oblique/direct, and original/stereotyped-extends in McCullough's
discourse to her conception of literary language and rhetoric. Here the polarity that
emerges may be termed main statement / ornament: "More often, the jo [an 'intro-
ductory' segment that provides a lead-in to the rest of the poem} makes a contribution
to the main statement" (p. 393); "To function as an engo [a word semantically as-
sociated to another to intensify poetic adhesion in a poem}, a word must have two
meanings: a primary one contributing to the main statement,and a secondary one
establishing relationship with somethingelsein the poem" (p. 222; emphasis added).
What is that "something else"? How does it relate to the "main statement"? It is
just as misleading continually to speak of "main statement" as somehow distinct from
rhetorical devices as it is to speak of "Man'yodirect styles" as distinct from "Chinese
elements." Moreover the notion of poetry as "propositional" remains a highly de-
batable issue in criticism today and, in any case, is not to be taken for granted.
Makurakotoba("pillow words," or epithets conventionally associated with certain
words) and jokotoba,which are said to derive from earlier times (earlyJapanese poetry
could, of course, be indirect too), and engoand kakekotoba("pivot words," or puns
playing on phonic/orthographic ambiguity), considered later developments, are all
part and parcel of the poet's response, that is, the "statement" itself.14 To put it
another way, the rhetorical itself is often the "statement" a poem makes. A similar

13
As RolandBartheshasput it, "We knownow that a text is not a line of wordsreleasing
a single 'theological'meaning,(the 'message'of the Author-God)but a multi-dimensional
spacein which a varietyof writings, none of them original,blend and clash"(Barthes1977:
146). Or, in EdwardSaid'swords:"To what extent is a text so discontinuousa series of
subtextsor pre-textsor paratextsor surtextsas to beggarthe questionof an authoras simple
producer?"(Said 1975:58).
14
McCullough'sdiscussionof KKS 113 mentionedin the next paragraph,whereshe notes
the potentialfor multiple interpretation,actuallydemonstratesthe impossibilityof keeping
the elementsdistinct, but herconcernwith "originality"obscuresthe total linguisticpicture.

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
38 RICHARD H. OKADA

dichotomy noted earlier, norm/deviance, can be seen at work in the low opinion
McCullough holds of certain Kokinshucategories- "Eccentric," "Names of Things,"
and much of "Miscellaneous."Poems in the last category receive some of the author's
harshest judgments.15
A further look at McCullough's discussion of rhetoric and rhetoricaldevices: "The
last three [of five] poems may represent an evolution from simile to metaphor" (p.
120 is metaphor higher on the tanka evolutionary scale than simile?); Komachi's
famous poem (KKS 113) "is neither metaphoricalnor symbolic; rather, it is wordplay
of the highest order" (p. 223). The terms metaphor, symbol, and wordplay have
become almost meaningless here. The problem extends to her discussion of Tsura-
yuki's famous kokoroand kotobadistinction: kokoroincludes "topic, theme, tone, wit,
mitate, conceit, and conception," whereas kotobaincorporatesmakurakotoba, jokotoba,
engo,kakekotoba,and imagery or, in McCullough's words, "diction, rhetoric, imagery,
syntax, and auditory effect" (p. 326). McCullough equates kokorowith "wit" and
kotobawith "beauty"(p. 329)-again not clearly definable(or distinguishable) critical
terms. And although she rightly defines the important term mitateas "a blanket term
embracing figurative language of all descriptions" (p. 66), she does not seem to see
its close kinship to other tropes but suggests that it is somehow qualitatively different
from both metaphorand simile. The latter two belong in her scheme to kotoba,whereas
mitategets subsumed under "conception," that is, kokoro:"If we are justified in treat-
ing mitate as an aspect of kokoro.... (p. 329). In the end, one is never sure exactly
where mitatebelongs or what the author's idea of figurative language, rhetoric, kokoro,
and kotobamight be. Another trope, metonymy, perhapsthe best way to decribe such
figures as mitate, never enters the discussion at all. 16 In short, the relation between
a term McCullough lists in one category and a term in the other is not self-evident,
and the terms themselves are not identical (the Heian preferencefor ambiguity itself
shows that kotobaand kokorowere hardly stable entities) but undergo change (and
change of value) according to the nature of linguistic difference and their particular
compositional contexts. Traditional binarisms like kokoroand kotoba(or form and
content) are metaphors that often conceal as much about linguistic interplay as they
reveal. That the Japanese themselves believed the situation to be more complex than
McCullough's discussion implies can be seen in the existence in waka discourse of
such terms as sama and sugata(signifying something like "rhetoricalmode" and "sit-
uational effect," respectively) which McCullough does not treat.
Difficulty with the intricacies of rhetoric is symptomatic of the state of most
waka criticism in the west today and is by no means unique to McCullough's discourse.
As their earliest recorded poems and folksongs show, the Japanese have always dis-
played an affinity for wordplay-possibly a combination of their belief in the power
of words, kotodama,impulses gained from Chinese precedents, and an inclination

15"KKS455
16
[is) negligible as literature"(p. 481).
I
See, forexample,KKS 27: "It twists together leafythreadsof tendergreen/ and fash-
ions jewels/ by piercingclear, white dewdrops- the willow tree in springtime."To see
dew as beads or jewels and willow leaves and branchesas threadsis to indulge in figural
languagewhose components,kokoroand kotoba,cannoteasily be divided. For McCullough
the poem shows that "kokoroitself becomesa sourceof beauty[kotobal as well as of wit"
(p. 329). Asidefromthe problemsinvolvedin attemptingto maintainessentialistdistinctions,
the mitateseen in KKS 27 may be bettercharacterized by the term metonymy,ratherthan
eithermetaphoror simile: the willow-threadand dew-jewelpairsbecomeassociatedconven-
tionallyinto boundfigures.The samecanbe saidforplum/cherry-snow, dew-tears,deer-bush
clover,etc. In a recentarticle,SuzukiHideo clearlyarguesthat mitateought to be considered
a rhetoricaldevice (Suzuki 1986).

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
 TRANSLATION AND DIFFERENCE 39
arising out of the phonological nature of the Japanese language. Instead of treating
rhetorical devices as somehow distinct from "main statement" or from expressions of
lyricism and "authenticity of suffering"(p. 395), new methods for tropological analy-
sis must be developed to resituate Konishi's work"7and to attempt to integrate the
various facets seen above-reasoning and confusion (although such terms along with
wit, urbanity, and aestheticism are perhaps better discarded), personification, mitate,
makurakotoba, jokotoba,engo,kakekotoba,utamakura(poetic toponyms), and kago (con-
ventionally employed poetic words)-into a view of Heian poetic composition that
can more fully describe the propensity for wordplay that lies at its heart, encompass
poems that might otherwise be seen as deviant or anomalous, and delineate the socio-
political forces that crisscross it. 18
The issues raised by the volumes under review provide an important, albeit be-
lated, lesson for literarystudies ofJapan. The act of translation, which initially dreams
the possibility of a seamless (or even a complete or nearly complete) transferof "con-
tent" from a source language into a target language, must not further mislead the
translator into assuming by analogy a possibility of natural transferals for areas of
research such as history, literature, and politics. In reading Brocadeby Night, one
experiences a time-warp sensation. Unaffected by the implications of recent research
in the humanities and social sciences for the study of Japanese waka, and otherwise
unable to view the poems in their situational complexity, it is a text out of the past.
Hence the tale told by BrocadebyNight becomes an old one. There is a source language/
culture that is deemed deficient in its ability to shine, so to speak, by itself. It must
be illuminated, explicated, re-presented in an altered (translated) guise. In this case
the guise is enhanced by the rhetorical fabric of Chinese civilization and elevated
terminology and by McCullough's authoritative text. McCullough's comment on KKS
297, from which the title of the companion volume derives, sums it up: "KKS 297
satisfies the kanajo criteria as fully as any tanka in the anthology. Its conception, a
model of courtly wit, demonstrates precisely the right degree of originality, its im-
agery evokes a scene of great visual appeal, and its carefully ordered diction creates
elegant aural effects.... We might say, indeed, that both KKS 297 and Kokinshu
itself resemble 'brocadeworn in the darkness of night.' The weaver's skill cannot be

17
Konishihimselfwasinterestedin demonstratinga similarityin attitudebetweenChinese
Six Dynastiesand Kokinshu poets; but even more, he was writing a polemic directedat ko-
kugakusha scholarswho had been shortchangingthe role of Chinain Heian poetic practice.
McCullough,in takinghim at facevalue, fails to see the extent to whichKonishimight have
overstatedthe case to makehis point. A reexaminationofJapaneseCourtPoetry,the standard
English-languagetext on classicalJapanesewaka (which also relied on Konishi'swork), is
lonf overdue(see Browerand Miner 1961).
One of the reasonsthat McCulloughincludedTosanikkiandShinsenwakawas that they
might shed light on why certain"ambiguousand overspecificcompositions,""naivesongs,"
or "assertivestatements"were part of Kokinshu7: "We must try to determinewhethersuch
anomalies are due to nonliteraryconstraints... to disagreementamong the compilers. . . or
simply to a lack of firm standards"(p. 494; emphasisadded).Then, after a conscientious
examinationof the two texts comes this conclusion:"Unhappily,we cannot claim to have
identifieddefinitiveexplanationsforKokinshui's editorialinconsistencies.... We lack the data
to go beyondthe realmof conjecture"(p. 534). Suchelementsonly appearto be "anomalies"
and "inconsistencies" when they areseen in termsof faultyassumptionsaboutpoetic compo-
sitionandvalue;in fact, it is elementslike thesethatmayprovidea clue to a differentapproach.
McCulloughalso overlooksimportantwork being done on rhetoricaldevices. She does not
cite names like Suzuki Hideo, one of the best scholarsworking today on Heian poetry,
especiallyKokinshu7 rhetoricaldevices, and KatagiriY6chi, who has been examining,among
other topics, makurakotoba, utamakura, kago,and the relationbetweenvoice and personain
waka. See, for example,Suzuki 1974 and Katagiri 1983.

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions
40 RICHARD H. OKADA

appreciatedwithout proper illumination" (p. 415). The poem, on its own, is neither
superior nor inferior to countless other tanka, but it gets transformedin the illusory
light of McCullough's discourse. Brocadeby Night stands as a mighty attempt to turn
the dream of pure transferalinto its own interpretativeweave. In some cases, however,
rather than proceed in the deceptive clarity of a mystified slumber, might it not be
better to provide illumination more awake to the dictates of difference?

List of References

BARTHES,ROLAND. 1977. "The Death of the Author." In Image-Music-Text.

Trans. Stephen Heath. New York: Hill and Wang.
BAXANDALL, MICHAEL. 1985. Patternsof Intention:On the HistoricalExplanationof
Pictures.New Haven: Yale University Press.
BELSEY, CATHERINE. 1980. Critical Practice.New York: Methuen.
BROWER, ROBERT H., and EARL MINER. 1961. JapaneseCourtPoetry.Stanford:
Stanford University Press.
FOUCAULT, MICHEL. 1976. The Archaeologyof Knowledge.New York: Harper and
Row.
KATAGIRI YOICHI. 1983. Utamakura,utakotoba jiten [A dictionary of utamakuraand
utakotobaj.Tokyo: Kadokawa Shoten.
KONISHI JIN'ICHI. 1978. "The Genesis of the 'Kokinshu' Style." HarvardJournal
of Asiatic Studies38:61-170.
KUHN, THOMAS. 1962. TheStructureof ScientificRevolutions.Chicago: University of
Chicago Press.
MCCULLOUGH, HELEN CRAIG. 1966. Yoshitsune: A Fifteenth-CenturyJapaneseChron-
icle. Stanford: Stanford University Press.
1968. Tales of Ise:LyricalEpisodes
from Tenth-Century Japan. Stanford: Stan-
ford University Press.
MCCULLOUGH, WILLIAM H., and HELEN CRAIG MCCULLOUGH. 1980. A Tale
of FloweringFortunes:Annals ofAristocraticLife in theHeian Period.2 vols. Stanford:
Stanford University Press.
NAJITA, TETSUO, and VICTOR KOSCHMANN, eds. 1982. Conflictin ModernJap-
aneseHistory.Princeton: Princeton University Press.
RODD, LAUREL RASPLICA, trans., with the poet MARY CATHERINEHENKENIUS.
1984. Kokinshi: A Collectionof PoemsAncientand Modern.Princeton: Princeton
University Press.
SAID, EDWARD. 1975. Beginnings:Intentionand Method.New York: Basic Books.
SUZUKI HIDEO. 1974. " 'Kokinshu' hyogen no keisei" [The formation of the Ko-
kinshi style]. Bungaku42, no. 5:63-77.
1986. " 'Kokinshu' no mitate ni tsuite" [Mitate in the Kokinshiu.Bungaku
54, no. 2:167-79.

This content downloaded from 188.72.126.181 on Wed, 18 Jun 2014 21:12:09 PM

All use subject to JSTOR Terms and Conditions