Review

Reviewed Work(s): Japanese Linked Poetry: An Account with Translations of Renga and
Haikai Sequences by Earl Miner
Review by: Hiroaki Sato, Earl Miner and Earl Miner
Source: The Journal of the Association of Teachers of Japanese , Sep., 1979, Vol. 14, No.
2 (Sep., 1979), pp. 181-193
Published by: American Association of Teachers of Japanese

Stable URL: https://www.jstor.org/stable/489044

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 support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

American Association of Teachers of Japanese is collaborating with JSTOR to digitize, preserve

and extend access to The Journal of the Association of Teachers of Japanese

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
 181

Journal of the Association Vol.XIV, No.2

of Teachers of Japanese

EPISTOLARY REVIEW

JAPANESE LINKED POETRY: AN ACCOUNT WITH

TRANSLATIONS OF RENGA AND HAIKAI SE-
QUENCES, by EarZ Miner. Princeton,
N.J.: Princeton University Press,
1979, xix, 376 pp. $20. 00.

Hiroaki Sato
with Earl Miner

IN JAPANESE LINKED POETRY Professor Miner

picks up where he and Professor Robert
Brower left off in JAPANESE COURT POETRY,
published nearly two decades ago. The
earlier volume, dealing with the period
from the beginnings to the fourteenth cen-
tury, naturally emphasized tanka, the pre-
dominant poetic form of that time. The
new volume deals with renga, which grew
out of tanka and its tradition. The renga
consists of two to a hundred alternating
"stanzas" of 5-7-5 and 7-7 syllables, and
is essentially collaborative poetry that
has "disjunctive linking" (p. 168) as the
basic rule of its composition. In this
book Professor Miner describes the history,
principal poets, and canons of this unique
form of poetry with his customary thorough-
ness, and translates six sequences in their
entirety: one with Sogi (1421-1502) as one
of the participating poets, one by S6gi
alone, three with Basho (1644-1694) as a
participant, and one with Buson (1716-1783)
as a particiDant. Just two of his more re-
markable accomplishments in the book are ac-
cording one of the "four grades of impres-
siveness" (p. 362) to each stanza and deter-
mining the "distance" between any two stanzas
--things most Japanese annotators have not
done. In these and other resDects, JAPANESE
LINKED POETRY is destined to become a clas-
sic, just as JAPANESE COURT POETRY has.

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
182

In agreeing to review this book, I

asked if I could make the review one that
incorporates the author's responses. The
review editor's reply was "yes," and Pro-
fessor Miner agreed to respond to my com-
ments. So, here are the questions I raised,
and Professor- Miner's answers to them.

SATO: Your observation that "the renga

poets do not allude to earlier renga in
writing their poems... (Haikai poets also
do not allude to earlier haikai)" (D. 14)
troubles me somewhat. Even if one regards
renga as distinctly seDarate from haikai
and ignores Basho's hokku alluding to S6gi's
(pp. 32-33), Bash6's "old pond / frog"
hokku (p. 96) was alluded to by later haikai
poets, Buson among them. Buson's "blue
heron" hokku (p. 130), which topped a com-
pleted kasen, appears to allude to the 33rd
stanza in "At the Tub of Ashes," which you
cite as one of your "favorite stanzas by
Bash6" (p. 116). Is your observation based
on - rule?

MINER: First of all, my thanks for being

able to respond to your questions. This is
a first experience for such a review of any
book of mine, and I do not remember ever
seeing it done before. Since it is tradi-
tional for reviewers to have the last word,
you may wish to append some further comment.
To your first question. I have not
seen in any shikimoku or other rule book a
ban against echoing earlier renga in a later
renga sequence. I put the question to Konishi
Jin'ichi. He agreed with me, saying that an
earlier renga is never used for honkadori.
He added that it was not so much a rule as
a practical consideration. There were not
widely circulating versions of the major renga
collections (TSUKUBASHO, SHINSEN TSUKUBASHO)
as there were of waka collections. To make
any sense (as he said), allusion would have
to be.caught and therefore even a borrowing

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
 183

would not be noticed. Haikai differs

somewhat, as all of us have seen. The
last stanza (ageku) of the last kasen in
"Fuyu no Hi" echoes the first (hokku) of
the first; the hokku and the second (waki)
in "Shi Akindo" are almost repeated in the
35th stanza and the ageku. In this con-
nection, I think we must consider that
there are different decorums involved for
renga and haikai no renga. Renga came to
be established as a "ga" art, whereas haikai
was generally regarded as a "zoku" art.
Basho certainly had higher ambitions, may-
be sometimes got to a kind of ga-zoku status,
but Teimon and Danrin styles were certainly
zoku. The point: renga used waka language
(utakotoba); haikai did not.

SATO: SDeaking of 'Issats masterpiece," you

say it is "not one of his three kasen" (p.
103), giving the impression that Issa's
kasen totaled three. Even though the Iwa-
nami edition has a selection of three, the
Shueisha edition (ISSA SHU) prints two hun-
dred and six kasen, and more are known to
exist.

MINER: You are right on Issa's kasen. I

was woolgathering. He was of course enor-
mously productive, not just in kasen and
hokku but in diaries as well. It's a pity
that more of his work is not readily avail-
able.

SATO: On Basho's following hokku, incor-

porated in THE NARROW ROAD:
Fleas and lice
and the sound of horses pissing
next to my pillow
Nomi shirami
wrea no bari suru
makura moto

you say, "There is an incongruity in the fact

that the greatest Doet of the time should be
in such filthy quarters. Bash6 saw it and
was not amused" (p. 122). It has been pointed

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
184

out that Bash6 was far from being a domi-

nant force in his lifetime (Imoto, for
one, says taking a view to the contrary
represents a tendency to idealize every-
thing about someone who has come to be re-
garded as a great man), and it is doubtful
that he should have considered himself any
better than one of the more prominent
tenja. More important, it was during that
trip to the north that he sought "lightness."
Also, the hokku refers to the name of the
place where it-was written, Shitomae ("in
front of pissing"), giving it a light twist.
These, and the tone of the hokku make me
lean toward the interpretation advanced by
Yamamoto and Otani that Bash6 meant the
hokku to be humorous. It may be that I tend
to try to see what haikai meant to its prac-
titioners, rather than follow "the kind of
thinking encouraged by Masaoka Shiki" (pp.
137-138), which is to say an emphasis on
shasei and shajitsu (o. 105). BashO's hokku
in question seems especially misleading in
its apDarent "realism."
MINER: By the general standards of his time,
Basho was certainly not taken to be an im-
portant, "ga" poet. Some hackneyed Rezei
poet of nothing other than historical inter-
est today would have been felt to be far more
important, as would some cliche-ridden samu-
rai poets in renga. On the other hand, that
is not to Pay that Basho and his followers
did not take them seriously. For various
kinds of evidence, let me mention in passing
Yasui's 35th stanza in "Shimotsuki ya no Maki";
the incidental parts (Kikaku's preface,
Shinken's Chinese verse and prose, the KIU
NIKKI, and J6so's postscript) of SARUMINO;
and KAREOBANA (a collection for Bash6 com-
Diled by Kikaku). And more to the Doint of
his own view of himself, it is striking how
he interweaves into his prose in OKU NO HOSO-
MICHI echoes of Doets such as Tu Fu and Li Po,
and even verses in Chinese and waka as well,
they having the same function as his own hokku.
Taken in context, his reported remarks have a

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
 185

similar import. As when he says that

"high art" (fuga, as I recall) is "all
one in nature, whether in Saigyo's waka,
in Sogi's ren.a, in SesshQ's painting, or
in Rikya's tea ceremonial." As the con-
text and the last reference suggest,
haikai may aspire to the same heights. As
you say, this claim has been taken by Jap-
anese more than seriously enough. Nobody
presently doubts that he is the greatest
poet of his age, just as Saikaku and Chika-
matsu make up the rest of that splendid tri-
umvirate, although at the time'they were
considered beneath haikai.

So how to interpret that hokku at

Shitomae? For one thing (to speak in your
support), it is not a real hokku but rather
one like many in the hokku sections of
SARUMINO, formally a hokku, but not of suf-
ficient elevation to qualify as a Basho-
style opening stanza. It is a kami no ku
sort of hokku, of the kind we might expect
to find in the development section of a
kasen. In short, I think you are quite right
for wanting to see comedy.
But that does not conflict, in my view,
with seeing that Bash6 was also annoyed. One
can call the joke verbal (kotoba), the an-
noyance conceptual (kokoro). I see some-
thing similar in the DaishOji episode. Sora
has left, going on ahead because of his ill-
ness. BashO is unhappy alone, worried about
Sora, and Dut out by the priests who urge him
to write a hokku. He treats them with sar-
castic wit in his hokku (which he may or may
not have given them), even while being lonely
and out of sorts without his traveling com-
panion. In any event, such interpretations
were but one of two things for which Professor
Konishi explicitly praised my translation
of OKU NO HOSOMICHI. I would like to think
the work the more interesting for such com-
plex tonalities, and BashO the finer for be-
ing both amused and annoyed--I mean for show-
ing his warts as well as his ambitions.

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
186

SATO: Annotators seem to agree that the

kasen, "At the Tub of Ashes," was com-
posed not at the Genjuan, but after
Basho left it (p. 316). Also, your expla-
nation of akuoke (p. 320) seems different
from that usually given.

MINER: You are certainly right. Why I

misremembered GenjQan for Mumyoan would
be more of a mystery if I did not find my
powers failing in other respects as well.
Such factual corrections are always wel-
come, and I wish you had reminded me of
this earlier, so that I could have entered
a change for the second (and paperback)
printing. But it is never too late to get
things right.

SATO: Evidently you thought and experi-

mented a good deal before deciding on the
arrangement of giving each stanza twice,
except the ageku, in the translations and
transliterations (.D ix). Still, the ar-
rangement makes me uncomfortable. BashO is
said to have observed, "A kasen may be com-
pared to thirty-six steps. One must not
think of'going back a single step" (SANZO-
SHI). In practice that dictum must have
been nearly impossible to follow, with so
many complex rules restricting the partici-
Dants. But I wonder if your arrangement
does not unnecessarily slow down the read-
ing of a finished linked Doem. I also won-
der if it doesn't intensify the temDtation
to translate each stanza more differently
the second time than is necessary.
MINER: I have always heard renga recited
so that the maeku is included with the
tsukeku. (Sometimes peoDle mistake, giving
the kami no ku and then the shimo no ku,
even when the maeku was in fact a shimo no
ku, but renga students consider that a sole-
cism.) Historical evidence includes the fact
that early renga collections, such as the
TSUKUBASHU,-not only put hokku at the end
but for each instance (for examDle, what
would'be termed 'haru no renga stanzas")
would give the maeku before the tsukeku se-
lected for its quality.

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
 187

The passage in the SANZOSHI is one I

therefore interpret to forbid going far-
ther back than the maeku, although as I
noticed there are exceptions to that in
BashO-style haikai. I know of none in
renga.
You may be right that the repetition
of the maeku can lead to various tempta-
tions for a translator (a transition to
the next query), but nothing else seems to
me to represent the way we must understand
renga and haikai.

SATO: You've been translating classical

Japanese poetry over two decades, and I'm
sure I'm not the first to notice your ten-
dency to add words in translation. In my
experience, a straightforward English trans-
lation of a text in Japanese comes, on the
average, to 70% of the original in sylla-
bic count, but your translations tend to go
over 100%. Thus, the hokku of the six se-
quences you have translated come to 103%;
the ageku, to 112%; the sho-ori of "Three
Poets at Minase," to 102%; and the sho-ori
of "Throughout the Town," to 116%. Though
I know you add words to approximate the ori-
ginal syllabic patterns and to present
translations "in natural English" (p. 78),
I wonder if that practice doesn't often
sentimentalize the effect, limit interpre-
tations, dilute the images, and induce
courtly stateliness where it is undesirable.
In the 17th and 18th stanzas of "Three
Poets at Minase," for example:
Even in clearing moments
the sleeves wet with the cold drizzle
on the traveler's clothes
my wet grass pillow of the journey
gives a weakened image of the moon
Haruru ma mo
sode wa shigure no
tabigoromo
waga kusamakura
tsuki ya yatsu san
C(. 191)

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
188

the addition of two "wets'" seems to make

the translation "wet." In the 20th stanza
of the same sequence,
the wind rustles upon the reeds
begrudging the brief dream of love

yume ni uramuru
oai no uwakaze
(p. 193)
the added words, "brief" and "love," may make
the transition to the following non-love stanza
awkward:

all those people

once seen often in the capital
are gone without trace
mishi wa mina
furusatobito no
ato mo nashi
(p. 193)
(Incidentally, in the 21st stanza is it pos-
sible to translate furusato "the capital"?)
Addition of words seems to bring about special
problems in haikai no renga. In his letter to
Sensen and Shikin, dated. 10th of the 4th month,
1690, BashO told themto "try not to be heavy-
handed and roundabout" in haikai and hokku.
In your translation of the hokku and waki
of "Even the Kite's Feathers," for example:
Even the kite's feathers
have been tidied by the passing shower
of early winter rain
stirred about by a gust of wind
the withered leaves grow still again
Tobi no ha mo
kaitsukuroinu
hatsushigure
hitofuki haze no
ko no ha shizumaru
(p. 283)

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
 189

the far too explanatory phrasing of

"Passing shower of early winter rain"
for hatsushigure and the addition of
"stirred about," "withered," and "again"
might be just the sort of things Basho
would have wanted avoided.

MINER: Yes, people have commented on

substantially what you say. More than
that, there are many faults of other
kinds to be found in my translations over
the years, and over those same years I
have noticed what I have taken to be
faults with the translations by others.
It surely must be important to consider
what should be the nature of translations
of various kinds of Japanese poetry. Why
else would people spend so much time do-
ing so? But surely there is no single
right answer. For example, I find it im-
possible to use rhyme, which is not popu-
lar among most contemporary poets or
younger readers, in any case. But if a
real poet came along and rhymed, it would
work. (I do not consider myself a real
poet.) That is to say that we may each
be right or wrong about our conceptions
of the nature of translations from Japanese
--whether prose, plays, waka , renga, hai-
kai, or modern free verse. But the results
from a fine theory may be terrible, and
the results from a weak theory may be ap-
pealing. Each of us had to decide on what
is feasible to Number One and run the hazard
of alienating some readers. When your and
Burt Watson's collection of Japanese poetry
is about to come out, you'll need some skin-
thickener, like the rest of us.

You also raise questions about three

specific translations, and since specifics
differ from theory, they deserve attention.
1) On the two "wet's" in MINASE, 17 and 18.
I agree with you, two "wet's" is a crowd,
even by standards of classical Japanese po-
etry. I wish that I had noticed and changed,
so that you would have had nothing to notice.

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
190

2) Whether "yume" means "dream of love"

in "Minase," 20. On p. 84, I quote from
UBUGINU on "yume." "Generally speaking
it implies love." With "uramuru" in the
same line, I do not see how it means any-
thing else. 3) Whether there is an awk-
ward transition to the next stanza, with
the interpretation of "yume" as love,
takes more consideration. But first, does
"furusatobito" imply (people of) the ca-
pital? This seems to me not as clear, as
certain as the preceding. But as any kogo
jiten will show, "jfurusato" tends to mean
the capital, or a capital, when in ruins
or when loss is felt (ato mo nashi). If
we consider KOKINSHU, 2: 90, by the "Nara
no Mikado" (Furusato to / narinishi Nara no /
miyako ni mo, etc.), which is cited by many
kogo jiten for "furusato," I think that my
translation is one that is defensible. Others
are also, but I think all an individual need
do is show evidence for what that person has
done. Now, if we consider that "furusato"
does mean a capital gone without trace, then
the longing implied for what was lost sorts
very well with love (omou, kou) as a yearn-
ing for what one does not have, the essence
(hon'i) of love in renga. (Again, see UBUGINU,
or the book, pp. 83-84.) This does not prove
that my translation is the sole right one,
or that there is any such thing for renga,
but I think it shows that I was not using
whole cloth.

SATO: I'm glad to come to my last point,

which is about lineation, because you more
than once expressed a desire to discuss the
subject with me in relation to my one-line
translations of tanka. (I have yet to pub-
lish my translations of renga, but as you
can expect, I give one line to each of the
5-7-5-and 7-7- syllable Darts.) I have two
reasons for translating tanka in one line.
One, I don't feel comfortable giving one
line to each of 5- and 7-syllable units;
here, at least Professors Seidensticker
and Teele seem to agree with me to a degree,

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
 191

translating tanka intwo lines. Two,

since the Western concept of lineation in
poetry came to be known in Japan during
the nineteenth century, some tanka poets
have experimented with lineation, though
they have never become the majority. I
should note that to me the experiments
and the fact that most poets have written
tanka in one line are equally important.
In the case of renga, the arrangement of
writing each of the 5-7-5- and 7-7- syl-
lable parts as one line seems to have be-
come standard by Bashu's time, so visual-
ly too, I feel uneasy about breaking up
the syllabic units.
MINER: The point of my response to the
preceding question applies to you here.
If, as I think, you are mistaken about
tanka being one line, that does not prac-
tically mean that you cannot translate
well. By the same token, if you are right,
you do not have a fail-safe recipe for
translation. So go ahead, do what you
believe in, buying a second bottle of
skin-thickener. (Knowing your tastes, I
may mention that the '74 red Bordeaux
have suddenly come on.) Let me also add
that Mark Morris (leaving Yale for Ade-
laide) has ideas rather like yours on
waka and ku as single lines.
Now to the matter of fact. It is
true that Japanese did not have the sense
of isolated lines as radical constituents
such as we find in Western poetry. The
ways of writing on shikishi and tanzaku
like the ways of printing today give ample
testimony to that. There was and probably
is ambiguity as to what "ku" means. Take
the term in "kami no ku" to mean both "the
first half" (that is, 17 syllables) and
also a plural--"first 3 ku" (5, 7, 5 syl-
lables or 3 lines, even if not clearly cut
in definition, as said).

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
192

Unless we grant that Japanese have

always had some sense of line, we have
to come up with some strange explana-
tions. Why do pillow words (makurako-
toba) never appear in what might be called,
for the sake of argument, 7-syllable
"lines?" How can one account for the pro-
hibition against hypometric "lines." And
even with my tin ear, I sense a very dif-
ferent rhythm in dramatic 7-5 syllable
combinations from the opposite 5-7 alter-
nations in choka and tanka. With your
keener ear you must do so far better.
Finally, on this matter of fact, is it
an accident that we can genuinely scan
waka, renga, and haikai--as we cannot mod-
ern free verse--in terms of fives and
sevens as lines?

I also see a practical problem in

translating older poetry. Leaving aside
sedoka, what about choka? If the one-
line-for-the-whole principle holds in
waka, translations of choka will look very
strange and probably require a scroll
rather than book. Also, see KOMPARU
KADENSHO SHUSEI, edited by Omote Akira,
Tokyo: Wan'ya Shoten, 1969, p. 526, "On
the matter of 5-7, 5-7 lines." For de-
cisive evidence, see Murakami's acrostic
poem in EIGA MONOGATARI, in which the hid-
den message can be understood only by
reading ten syllables--the first and last
of the five (forgive me) lines.
I think you have a bee in your bonnet
on this and have overreacted to certain
simplicities. But go ahead. As I've said
twice, theory does not decree success or
non-success in practice. Run your risks
and lay in your wine.
I must thank you for letting me re-
spond to your questions, which really do
raise the state of the art. The ideas of
all of us change in the course of time, SQ
no doubt my present ones will differ ten

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms
 193

years from now, as my present notions

about translation differ from those a
decade ago. My current ruling passion
holds that one should get into English,
when possible, the order of lines
(whoops! phrases) of the original. I
could go on, but that's enough. Domo
ski ni arigato.
SATO: Kochira koso. But may I add a
word or two. That the five and seven syl-
lables are the basic units is not in ques-
tion. Their very existence shows that. In
response to your citation of the Emperor
Murakami's tanka, I may point to acrostic
examples using only the first syllable of
each tanka, such as those written by Izumi
Shikibu and Fujiwara no Teika, and to
rhymed examples, such as those by Teika,
where only the last part of each tanka is
used. In renga, acrostics, at least those
cited by Fukui KyQzO in RENGA NO SHITEKI
KENKYU, use only the first syllables of the
5-7-5 and 7-7 syllable parts. The question
is how to lineate a tanka in English trans-
lation. Here, one clue may be the ambig-
uous use of the word ku you mentioned, and
another clue, the equally ambiguous use of
the word toku in the MAN'YOSHU; to me, they
seem to show gropings for larger entities
than the two syllabic units. Of course,
nothing conclusive can be said. If your
question about ohOka is apt, it may be
equally apt to ask how you propose to trans-
late, for example, Miyazawa Kenji's tanka
that are written in one to five lines (and,
to go a step further, Kitahara HakushQ's
"lines" consisting of various combinations
of 5 and 7 syllables). In this, my one-
line translation is no more than a way of
expressing my discomfort with the five-line
approach.

This content downloaded from

130.123.99.4 on Tue, 06 Oct 2020 11:20:11 UTC
All use subject to https://about.jstor.org/terms