On the Translation of Style in Oral Narrative

Author(s): Dennis Tedlock

Source: The Journal of American Folklore, Vol. 84, No. 331, Toward New Perspectives in
Folklore (Jan. - Mar., 1971), pp. 114-133
Published by: American Folklore Society
Stable URL: http://www.jstor.org/stable/539739 .
Accessed: 28/06/2014 21:26
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 support@jstor.org.

American Folklore Society is collaborating with JSTOR to digitize, preserve and extend access to The Journal
of American Folklore.

http://www.jstor.org

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

DENNIS

TEDLOCK

On the Translationof Style in Oral Narrative

A DISCRIMINATING
READER, hoping to find collectionsof AmericanIndiannarra-

tives which are at one and the same time thoroughlyauthenticand respectable
as literature,is likely to be disappointed.When he exploresthe narrativespublished before the field methods of Franz Boas were widely employed,he may
decide that their style seems more Victorian than Indian. If he then turns to
moderncollectionsbut still avoidspublicationsintendedonly for the use of scholars, he may find his prospectivereading described,as in the case of Jaime de
Angulo's Indian Tales, as suitable fare for both children and adults.' Such a
volume will seem about as promisingto him as a movie rated "G" for general
audiences.
If our readerdares to venturebeyond dust jacketsand back coversto reada
preface,he mayfind a commentaboutan authorsimilarto the following one from
Oliver La Farge'sprefaceto TheodoraKroeber'sThe Inland Whale: "She...
turnedwriterand retoldthe stories,a dangerousprocesshere successfullyapplied.
The retelling, one might say, is ethnologicallyhonorable.The storieshave not
beenprettified,elaborated,or ladenwith pseudo-literary
trimmings."This sounds
reassuring,but there is more: the stories "have simply been put into a familiar
idiom, with restraintand good taste, and in some cases purged of the insistent
repetitionsand clutteringdetailsthatprimitivepeople often stuffinto theirstories
for ulteriorpurposes."2
Wishing for greater authenticity,our reader may turn at last to the vast
But he will soonwonder
scholarlycollectionsproducedby Boasiananthropologists.
whetherthe original style of these narrativeswas as choppyand clumsyas that
of most English translations.If he takes these translationsto represent,as Boas
claimed,"faithfulrenderingof the nativetales,"3and if he remainsdisappointed
with popularizations,he may end by agreeing with La Farge, who said, "The
literaryvalue of a greatdeal of primitiveliterature,whethermythsor tales,is nil.
1 Such an evaluation is made in one of the blurbs on the back cover of the paperback edition of
this book (New York: Hill and Wang, 1953). In the preface de Angulo says, "I wrote these
stories . .. for my children" (p. 5).
2 Theodora Kroeber, The Inland Whale (Berkeley, Calif., 1963), 8-9.
3 Franz Boas, Race, Language, and Culture (New York, 1940), 451.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

115

That of muchof the rest is apparent,in the rawform, only to connoisseurs,while

thosewho undertaketo retell someof it often achieveonly emasculation."4
Unless it is true that manyof the oral narrativesof non-Westernpeopleshave
little or no literaryvalue, and that what value they do have is untranslatable,
then somethinghas gone wrong along the way from the oral performanceto the
printedpage. Throughthe close examinationof a single widely-publishednarrative tradition,that of the Zufii Indians,5I hope to show that something has
indeed gone wrong, and to suggestwhat might be done differentlyin the future.
I.
The Zufii narrativescollectedby FrankHamilton Cushingin the i88os have
alwaysattractedmore attentionthan any others: "The Beginning of Newness"
has been anthologizedby Astrov and Thompson, "The Poor TurkeyGirl" by
Thompsonand Greenway,and "The Cock and the Mouse" by Greenwayand
Dundes."But the apparentattractivenessof Cushing'swork is anythingbut a
measureof its reliabilityas a representationof Zufii literature."The Beginning
of Newness," together with the rest of Cushing's"Outlinesof Zufii Creation
Myths,"has long been a problemfor studentsof Zufii culture.Cushinghimself
says that these "outlines"are just that and not directtranslations,'but it is his
additionsto the narrativesratherthananydeletionswhichhavecausedthe trouble,
for, as Bunzel has written, the work "containsendless poetic and metaphysical
glossing of the basic elements, most of which explanatorymatter probably
originatedin Cushing'sown mind."8The "metaphysicalglossing" referredto
includesstrongovertonesof monotheism(also found in Stevenson'swork) which
reflectthe theoreticalpreoccupationsof nineteenth-century
anthropologyrather
than Zufii belief.
4 Kroeber, 7.
In using "Zufii" rather than "Zuni," I follow the practice of this journal. But the Englishspeaking residents of the Zuiii area, including bilingual Zuiiis, use "Zuni" in both spelling and
pronunciation. Academics frequently render "Zuiii" as "zoonyee" (rather than the Spanish
"soonyee"), so that the final result after retaining the fi is still an English corruption of what is
already a Spanish corruption of the Keresan corruption of the Zufiis' word for themselves, which
is Shiwi.
6 Cushing's translations may be found in Frank Hamilton Cushing, "Zufii Fetiches," Annual
Report of the Bureau of American Ethnology, 2 (1883), 13-19, 21-24; in "Outlines of Zufii
Creation Myths," Annual Report of the Bureau of American Ethnology, 13 (1896), 379-447; in
Zufi Folk Tales (New York, I90o; New York, i93i); and in Zufii Breadstuff, Museum of the
American Indian, Heye Foundation, Indian Notes and Monographs, 8 (1920), 20-54, 58-124,
270-288, 395-515. Two additional Cushing interpretations were recorded by men who visited
him in the field: John G. Bourke, Diary (unpublished MS in the library of the U.S. Military
Academy at West Point, 1881), 2565-2585; H. F. C. ten Kate, "A Zufii Folk-Tale," JOURNAL
OF AMERICANFOLKLORE,
30 (1917), 496-499. "The Beginning of Newness," from "Outlines
of Zuni Creation Myths" (pp. 379-381), is reprinted in Margot Astrov, The Winged Serpent
(New York, 1946), reprinted as American Indian Prose and Poetry (New York, 1962), 240-242;
and in Stith Thompson, Tales of the North American Indians (Bloomington, Ind., 1929, 1966),
17-19. "The Poor Turkey Girl," from Cushing, Zudi Folk Tales (54-64), is reprinted in
Thompson ( 225-231) and in John Greenway, Literature Among the Primitives (Hatboro, Pa.,
228-234. "The Cock and the Mouse," from Cushing, Zuifi Folk Tales (411-422), is
1964),
reprinted in Greenway, 151-158, and in Alan Dundes, The Study of Folklore (Englewood Cliffs,
N.J., 1965), 269-276.
7 Cushing, "Outlines of
Zufii Creation Myths," 375.
8 Ruth L. Bunzel, "Zufii
Origin Myths," Annual Report of the Bureau of American Ethnology,
47 (1932), 547.
5

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

I16

DENNIS TEDLOCK

"The Poor TurkeyGirl" and "The Cock and the Mouse" are cited by the
anthologistsas classicexamplesof the AmericanIndian adaptationof European
tales. Cushingrelatesthe historyof "The Cock and the Mouse" as follows: he
had told an Italianversionof it to some Zufiishe had broughtto New England;
abouta year later, backat Zufii, he heardone of these samemen tell (in Zufii)
a considerablyadaptedand expandedversionwhichwas laterpublished.9Exciting
though the Zufii versionmay be, it is not clearwhat the original Italianversion
used by Cushingwas like, for, as Dundes has pointedout, the Zufii versioncontains some distinctlyEuropeanmotifs which are lacking in the Italian version
printed beside it in Cushing's book.10There are further problems: Cushing
necessarilytold the story to his Zufii audiencein the Zufii language (the three
men were monolinguals), and some of the "Zufii"alterationscould well have
originatedwith Cushingin the processof the telling. Moreover,as will be seen in
detail shortly,Cushingwas given to elaborationswhen renderingZufii tales in
English, and there is no reasonto believe he restrainedhimself in the present
case.
Whateverthe specialproblemswith "The Beginning of Newness" and "The
Cockand the Mouse,"the opinionhas been widely held that the qualityof Cushing's translationsis quitegood. The novelistMaryAustin is extravagantwith her
praise,writing that Cushing "is the only Americanwho notablybroughtto bear
on [primitivelore] adequateliteraryunderstanding,"and that Cushing'sis "the
best-sustainedtranslationof aboriginalAmericanliterature,"and, still further,
that Cushingmade no effortto "popularize"his stories."MargotAstrov,in the
introductionto her anthology,lists Cushing as one of those ethnologistswho
have best met "the two requirements"of the translator:"linguisticfidelityto the
original" (short of strictly literal translation) and the communicationof the
"culturalmatrix"of the original.'2But Hymeshas recentlyshownhow far Astrov
has gone wrong in judging the quality of song translations,'3and in a similar
spirit I hope to show here that narrativetranslations,too, are not alwayswhat
they seem.
Among the more curious things in Cushing's major collection, Zuni Folk
Tales, are the oathsused by the characters.Austin citesthese as one of the things
she admiresmost and gives "By the delight of death!"as an example;14other
oathsinclude "Soulsof my ancestors!""Demonsand corpses!""Bythe bones of
the dead!""Oh, ye gods!" and "BelovedPowers!"'5But the Zufiishave no such
oaths; they never make profane use of words denoting death, souls, ancestors,
corpses,"Powers,"and gods. They do use a goodly numberof interjectionsin
tales, such as tisshomahhd(dread), hiyahha (fright, female speaking), and
ya' 'ana (disgust, male speaking),16 but there is not a single one of these inter9 Cushing, Zufi Folk Tales, 411.
10 Dundes, 274.
xix-xx, xxvi.
11 Cushing, Zufii Folk Tales
(1931),
12 Astrov, 5.
13 Dell
Hymes, "Some North Pacific Coast Poems: A Problem in Anthropological Philology,"
American Anthropologist, 67 (1965), 316-341; reprinted in Stony Brook, 1-2 (1968), 179-204.
14 Cushing, Zudi Folk Tales, xxviii.
15 Ibid., 134, 182-183.
16 The
orthography used for these and other Zufii words herein is as follows: vowels should be
given their Continental values; double vowels (aa, etc.) are like the long vowels in Greek.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

117

jections which has any denotation other than the emotion it is supposed to express.
In this case, then, Cushing's translations do not represent "linguistic fidelity to
the original," and, further, they misrepresent the "cultural matrix" of the tales.
Perhaps the most serious difficulty with Cushing's Zuhi Folk Tales is that he
embroiders the tales with devices, lines, and even whole passages which are
clearly of his own invention and not mere distortions. Similes are totally lacking
in all other translations (and in texts as well), but they abound in Cushing's
tales: for example, a young man attacked by a swarm of mosquitoes was "crazed
and restless as a spider on hot
and a person outdoors at night saw a
ashes,""
"light that was red and grew brighter like the light of a camp fire's red embers
when fanned by the wind of the night-time.' 'is These passages may have literary
merit in English, but they do not even have literary existence in Zufiii.
Another kind of embroidery, not so serious as some of the others, is Cushing's
insertion of explanatory material for the benefit of his readers. For example, he
begins one tale with a lengthy explanation of the geographical location and
appearance of its setting,19 whereas a Zufii narrator would take his audience's
knowledge of local geography for granted. In another example, Cushing describes how a suitor ate very little when given a meal at a girl's house (which a
Zufii narrator would do) but then adds, "You know it is not well or polite to
eat much when you go to see a strange girl,''20 again a case in which a Zufii
narratorwould take his audience's knowledge for granted. Of course it is possible
that some of this explanatory material was inserted by Zufii narrators for Cushing's own benefit, but whatever its origin it does misrepresent normal Zufii
practice.
The most distressing of all Cushing's inventions are his moralistic passages. As
I have shown in detail elsewhere, the didactic content of Zufii tales is usually
either implicit or addressed by one tale character to another, and it is never addressed by the narrator directly to his audience.21 But Cushing begins one tale
this way: "Listen, ye young ones and youths, and from what I say draw inference.
For behold! the youth of our nation in these recent generations have become less
sturdy than of old; else what I relate had not happened.'22 In some other cases
he points out the moral in the third person, but his tone is still excessively moralistic, as in this example from the end of an Orpheus tale: "But if one should live
as long as possible, one should never, in any manner whatsoever, remembering
this youth's experience, become enamored of Death."23
It should now be sufficientlyclear that Cushing frequently violates the linguistic
Consonants should be pronounced as in English, with the following exceptions: p and t are not
aspirated; lh is like English and h and 1 pronounced simultaneously; double consonants (kk, 11,
etc., except that ch becomes cch, lh 11h, and sh ssh) are like those in Italian; and ' is the glottal
stop, which, when it follows ch, k, kw, ky, or ts, is pronounced simultaneously with these sounds.
Stress is on the first syllable of a word; exceptional words are marked with '.
17 Cushing, Zudi Folk Tales, 6.
18Ibid., 24.
1oIbid., 203
20Ibid., 3.
21 Dennis
Tedlock, The Ethnography of Tale-Telling at Zufi, (Ann Arbor, Mich.: University
Microfilms, 1968), chapter iii.
22 Cushing, Zufi Folk Tales, 185.
23
Ibid., 53

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

DENNIS TEDLOCK

Is8

and culturalrequirementswhich Astrovsets for translators,and that a good deal

of what Austin calls "color... so delightfullyrendered"(includingthe oaths)24
looksmorelike Victorianquaintnesson closeexamination.
The work of MatildaCoxe Stevenson,a contemporaryof Cushing,avoidshis
stylistic embroideries,but her major compendiumof narrativesin The Zuni
Indiansis not a translationof actualZufii performancesbut rathera descriptive
summaryin her own words.25 Much of the apparentorderin these materialsis
her own: she ignores the possibilityof alternateversionsand attemptsto place
each storyin a chronologicalsequencewhich reflectsher own Westernpreoccupation with historymore than actualZufii practice.Elsewherein the samevolume,
however,she does presentone narrativewhich (though ratherabbreviated)appearsto be a directtranslation.26
Beginning in the second decade of the present centurya veritablearmy of
Boasianfield workersdescendedupon Zufii. The firstmembersof this armyto
publish translationsof Zufii narrativeswere Franz Boas himself, Elsie Clews
Parsons,and EdwardL. Handy;27hard on their heels cameRuth L. Bunzel and
Ruth Benedict.28 Only Parsonsand Bunzel publishednative-languagetexts, and
onlyBunzelpublishedtextsin anyquantity.29
Members of the Boasian school, at Zufii and elsewhere, typically valued
translationsthat were "direct"or "close" or "literal,"publishedwith as few
changes as possible from the sort of English used by interpretersor bilingual
narrators.Thus Parsonscould write, in introducinga collectionwith which she
was particularlypleased, that the tales "interpretedby Lare as close to the
and when
in
I
as
it
is
to
narrative,"'3
think,
originalZufii,
possible get English
she showedthese translationsto A. L. Kroeber,who had trainedher interpreter,
he said, "In readingthem, I can hear LspeakingZufii."3` One can indeed
"hearLwhen
awkwardchoicesof Englishwords
speakingZufii,"especially
are preservedor when Englishwords are organizedaccordingto Zufii grammar,
as in these passages:"The strapsthe man carriedwood with, in the other room
24 Ibid., xxviii.
25 Matilda Coxe Stevenson, "The Zufii Indians," Annual Report of the Bureau of American
Ethnology, 23 (1904), 23-6I.
26
Ibid., 135-137.
27 Franz Boas, "Tales of Spanish Provenience from Zufii," JOURNAL OF AMERICAN FOLKLORE, 35 (1922),
62-98. Elsie Clews Parsons and Franz Boas, "Spanish Tales from Laguna
and Zufiii, New Mexico," JOURNALOF AMERICANFOLKLORE,
33 (1920), 47-72. Elsie Clews
Parsons, "Notes on Zufii, Part II," Memoirs of the American Anthropological Association, 4
(1917), 302-327; "Pueblo-Indian Folk-Tales, Probably of Spanish Provenience," JOURNALOF
"The Origin Myth of Zufii," JOURNAL OF AMERICAN
AMERICAN FOLKLORE, 31 (1918),
216-255;
FOLKLORE,36 (1923), 135-162; "The Scalp Ceremonial of Zuiii," Memoirs of the American
Anthropological Association,
43 (I930),
1-58. Edward

31 (1924),
L. Handy,

28-34; "Zufii Tales," JOURNAL OF AMERICAN FOLKLORE,

"Zufii Tales," JOURNAL OF AMERICAN FOLKLORE, 31

451-471.
(1918),
28 Ruth L. Bunzel,

"Zufii Origin Myths"; "Zufii Katchinas," Annual Report of the Bureau of

American Ethnology, 47 (1932),
837-I086 (narratives are scattered throughout this work);
Zudi Texts, Publications of the American Ethnological Society, 15 (I933). Ruth Benedict, Zuni
Mythology, Columbia University Contributions to Anthropology, 21 (i935).
29 Parsons, "Zufii Tales" (texts are given for only two of these narratives); Bunzel, "Zufii
Origin Myths" and Zuhi Texts.
30 Parsons, "Zufii Tales, " 2.
31 Ibid., 2 (quoted by Parsons).

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

19

he would hang up," "This way you were going to do to me," and, incredibly,
"Thenone of his legs he threwup."32
The literalismin most other translationsof Zufii narratives,including those
of Boas and Benedict,does not reachthe absurdextremesof Parsons.Benedict
followed the usual practiceof her contemporariesin asking her informantsto
give "literal"translations,but it was her statedintention to smooth out "their
inadequateEnglish" in her published verisions.33She did indeed eliminate
obvious grammaticalerrors,but stylisticinadequaciesremainincluding a choppiness and lack of grammaticalcomplexitycommonto muchof the work of this
period. Zufii narrators,like many others, frequentlykeep a story in motion by
combiningstringsof clausesinto long sentences,and by joining these sentences
with parallelism.But one would neverknow this from readingBenedict'stranslation:
Her eyes were almost shut. She was skin and bones. She was too weak to sit up and she
scratched herself all the time. He jumped up. He ran to the house of Pekwin's son. His
wife was just as old. She had gray hair and was bent double. The two young men were
angry. They would not talk to their wives. They drove them away. The two old women
went off leaning on their canes. They were too weak to travel. There was no rain. The

peoplewerehungry.34
Such a disasterprobablyresultsnot only from informantEnglish, but also from
the stops and startsof the dictationprocessand from a tendencyto treat parallelisms as not worth preservingin print. But whatevertheir sources,Benedict's
distortionsare not purely the result of dictation:Bunzel's translations,which
were based on dictatedZufii ratherthan dictatedEnglish, have a very different
character:
They laid the deer down side by side. They laid them down side by side and they made the
boy sit down beside them. After they had made him sit down they gave the deer smoke.
After they had given them smoke they sprinkled prayer meal on them. After they had
sprinkled prayer meal on them the people came in.35

Probablyas a resultof dictation,the parallelismhere (A, AB, BC, CD, DE) is

more mechanicalthan the parallelismin my own tape-recordedZufii narratives,
and the sentencelength (as elsewherein Bunzel's work) fails to reachthe extremes possible in uninterruptednarration.Despite these flaws the text translations of Bunzel displaythe qualitiesof oral performancebetterthan any of the
otherZufiiworkof this period.
Aside from their frequentlack of parallelism,the narrativesof the Boasian
school tend to be condensationsof what a performerwould tell in a normal,
and DuBois, and Gladys
spontaneoussituation.A. L. Kroeber,Demetracapoulou
Reichard, all of whom recognized this problem in their own collections of
American Indian narratives,place most of the blame on the tediousnessof
dictationandthe consequentabsenceof a responsivenativeaudience.36
Substantiat32Ibid., 6, 30.
Benedict,vol. I, xxxviii.
34Ibid.,vol. I, 219.
35Bunzel,Zuhi Texts, 109.
36 A. L. Kroeber, "A Mohave Historical Epic," University of California Anthropological
and CoraDuBois, "A Studyof Wintu Mythology,"
Records,ii (i951), 133; D. Demetracapoulou
33

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

DENNIS TEDLOCK

120

ing their view is the fact that the narrativesin my own Zufii collection,related
in all instancesto at leasta smallnativeaudienceandtakendownby taperecorder,
averagenearlytwicethe length of the narrativesin Benedict'scollection.37
After the 1930s, collectionof AmericanIndian narrativeswent into a rapid
decline. Texts and translations(other than Zufii ones) continued to appear
sporadically,but many of these later collections, such as Jacob's Clackamas
ChinookTexts,38were delayedreportsof field work done duringthe mainperiod
of Boasianactivityratherthan reportsof anythingnew. In the Zufii case, the
thirty years which separatedthe appearanceof Benedict'sZuhi Mythology (in
1935) from the beginningof my own field work saw the publicationof only one
minor collectionof fresh narratives.39
Generallyinsteadof fresh materialsthere
appearedanalyticaltreatmentsof old ones that reflectedthe two main currents
in modern narrativetheory: Bert Kaplan in "PsychologicalThemes in Zufii
Mythologyand Zufii TAT's" sees Zufii myths as possible projectionsof "the
in "The
repressedunconsciousprocessesof the id,"40 while ClaudeLUvi-Strauss
StructuralStudy of Myth" finds Zufii myths (among others) exhibiting the
Hegelian dialectic,which he believesto be a substratumin all humanthought.41
II.
While advancesmay have been made in the analysisof oral narrativecontent
since the 193os, the artof translationhas seen no substantialgains since the turn
of the century.The tape recordershouldimprovethis situation,but its full possibilities have yet to be exploited. It has been a practicaland accuratefield instrument for only a short time, and the theoreticalinterestsof many of its users
(or potential users) are centered on "content"which they presume enjoys
a certainindependencefrom the fine points of "style"and translation.John L.
Fischer,for example,saysthat in sociopsychological
analysisthe primaryconcern
"is with the semanticsof folktale;with the messageor 'tale picture'which can
be transmittedby the codes of variouslanguages,or by variousequivalentconstructionsin a single language."42 Levi-Straussholds a similarview, though his
particularanalyticalinterestsdiffer from those of Fischer:"The mythicalvalue
of the myth remainspreserved,even throughthe worst translation.. ... Its subJOURNALOF AMERICANFOLKLORE,45 (1932), 400; Gladys A. Reichard, "An Analysis of
Coeur d' Alene Indian Myths," Memoirs of the American Folklore Society, 41 (i947),
5.
37 Tedlock, "The Ethnography of Tale-Telling in Zufii," 279-330; "Pelt Kid: A Humorous
Zuiii Tale," Human Mosaic, I (1966), 55-65; "The Boy and the Deer: A Zufii Tale," The Kiva,
33 (1967), 67-79; "The Boy and the Deer: A Narrative Poem of the Zufii Indians," Stony Brook,
5-6 (in press); Finding the Center: Narrative Poetry of the Zuli Indians (New York, in press).
38 Melville Jacobs, Clackamas Chinook Texts, Publications of the Indiana University Research
Center in Anthropology, Folklore, and Linguistics, 8 (Bloomington, Ind., 1958) and ii (1959).
39 Anna Risser, "Seven Zufii Folk Tales," El Palacio, 48 (i941), 215-226.
40 Bert Kaplan, "Psychological Themes in Zufii Mythology and Zufii TAT's," in The Psychoanalytic Study of Society, ed. Warner Muensterberger and Sydney Axelrod, vol. II (New York,
1962), 255-262.
41Claude Levi-Strauss, "The Structural Study of Myth," JOURNAL OF AMERICAN FOLKLORE,
68 (1955), 428-444; reprinted in Claude Levi-Strauss, Structural Anthropology (Garden City,
N. Y., 1967) 202-228.
42 John L. Fischer, "The Sociopsychological Analysis of Folktales," Current Anthropology,
4 (1963), 237.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

121

stancedoes not lie in its style, its original music, or its syntax,but in the story
whichit tells."43
Even when a scholar does show interestin the stylistic aspectsof narrative
traditionsthere is no guaranteethat he will give much thought to translation.
Melville Jacobs,for example, though he promisesthat his analysisof style or
form in ClackamasChinooknarratives"will greatlyenhanceenjoyment"of that
literature,offers translationswhich are typicallyBoasian in being "almost literal."44Despite the literaltranslation,the readerdoes not experiencedirectlythe
of Clackamas
"terseness,"which is supposedlyone of the principalcharacteristics
style, for Jacobshas made hundredsof "explanatory"parentheticalinsertionsto
rescuehim fromthatterseness.
In some cases the neglect of translationis doubtlessrelatedto a belief that
style, or at least the betterpart of it, is simply untranslatable.Franz Boas and
A. L. Kroeber,for example,held thatstyle (or "literaryform") was so boundup
with the peculiaritiesof particularlanguagesthat it was unlikelyto survivetranslation.45If theirview of styleis combinedwith the view thatcontentsurviveseven
badtranslation,then thereis no roomat all for an artof translation.It maybe that
no one scholarhas ever held both theseviews simultaneouslyin their pure form,
but manyscholarsof the past four generationsmight as well have done so.
Somecollectorsof AmericanIndiannarrativeshavetakenissuewith the narrow
linguisticview of style. Demetracapoulouand Du Bois even go so far as to say
that in the Wintu case, given an interpreteror narratorwho is fluentin English,
a translationinvolvesno distortionat all.46Jacobsfindsin the Clackamascasethat
all but a very few featuresof narrativeform areindependentof the particularities
of Clackamaslinguistics,4'the implicationagain being that translationproblems
should not pose any great difficulty.The Zufii narrativetraditiondisplaysmore
stylisticmanipulationin phonology,lexicology,and syntaxthan Jacobsindicates
for the Clackamas,but once more a large part of style lies outside of what is
traditionallythought of as linguistics,and I would add that even the linguistic
featuresof Zufii style do not createinsurmountable
translationproblems.
On the phonologicallevel, Zufii narrativestyle involvesonly two commondistortionsof normalpatternsand both of these also occurin everydayspeech, although they are more frequent in narrative.One of the distortionsinvolves a
combinationof stressshift andvowel lengthening:a tale charactermaystartoff an
ordinarygreetingwith somethinglike hom nana, "My grandfather,"with stress
on the firstsyllableof nana (as is normal), but if the occasioncallsfor exceptional
formalityor seriousness,he will shift the stressand lengthenthe final vowel as
follows: hom nana-. It might be hard to get a similar effect by shifting the
stress on "grandfather"in translation,but a syntacticshift to "Grandfatherof
mine"succeeds,I think,in reproducingthe originaleffectof formality.
The othermajorphonologicaldistortionin Zufii narrativeinvolvesa combina43 Levi-Strauss, 430.
44 Melville Jacobs, The Content and Style of an Oral Literature, Viking Fund Publications in
Anthropology, 26 (1959), 3, 6.
45 Boas, Race, Language, and Culture, 452; A. L. Kroeber, 133.
46 Demetracapoulou and DuBois, 386.
47 Jacobs, The Content and Style of an Oral Literature, 7-8.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

DENNIS TEDLOCK

122

tion of intonation change and vowel lengthening: "Thus they lived on" would
ordinarily be intoned as follows, with the lowest pitch at the end:

lesnolh 'aateya'kya
But the length of time involved may be emphasized by shifting the highest pitch
to the final syllable and drawing out the final vowel for as much as two or three
seconds:
2

lesnolh 'aateya'kyaThe same operation may be performed on a verb like 'akya, "he went," to indicate a long distance (but not necessarily a long time). Such forms might be
translated as "Thus they lived on and on and on," and "He went and went and
went," but in Zufii this sort of repetition usually indicates repeated action rather
than drawn out action (or state of being), as in lines like, "And all the people
who had come killed the deer, killed the deer, killed the deer." To translate
drawn out Zulii verbs as repeated ones would mean collapsing two stylistic devices into one. A more direct translation seems a better solution: "Thus they lived
" (in which the o's should be held). This
on," and "He went on
rendition may seem strange on the printed page, but comparable lengthening does
occur in spoken English, as in, "It's been such a lo-ng time."
There are no grammatical differences between everyday speech and formal
narrative in Zuiii, except for a greater tendency to construct long sentences in the
latter. The following, in strict syntactic terms, is a single sentence (each line
break indicates only a slight pause):
Towayalan 'ahayuut 'aachky'akwap,
he'shoktan 'aatoshle
'aachi
ky'akwap,
'itiwan'an ihuwal'ap,
pinnaawanihuwal'ap, ky'ak'iimalhuwal'ap,
lesnolh lhuwalaa'ullapnap,taknankwayilep,taknankwayilenakwa'ky'ak'aawina'ma.48
There is no translation problem here: given as a single English sentence, this runs
as follows:
At CornMountainthe two 'ahayuutahad theirhome,
at He'shoktathe 'aatoshle
the two of them
had theirhome,
at the Middle therewerevillagers,
at Winds' Placetherewerevillagers,at Ky'ak'iimatherewerevillagers,
there were villagers all aroundgoing out to gather wood, and when they went out to
gatherwood they did not comehome.
This is somewhat cumbersome by the normal standards of written English prose,
48 Tedlock, Tale H-9, personal collection.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

123

but such length would not be extraordinaryfor an oral narratorin English (unless he were readingfrom a writtentext) or for a largenumberof Englishpoets.
Most of the remaining"linguistic"manipulationin Zufii style involves the
choice of lexical items or formulaicphraseswhich would be rare or absent in
completelyneutraleverydayspeech.As Newmanhas shown,Zufiivocabularyruns
along a continuumfrom items labelledas slang (penaky'amme)to items labelled
as sacred (tewusu), with variousshadingsand a largeunnamedneutralcategory
excludedfrom
in between.*"
Anythingclearlyrecognizedas slangis systematically
is
used:
formal narratives,but at least one slightly substandardterm
'okyattsik'i,
which Zufiis translateas "old lady." A hideous old ogress named 'aatoshle,for
example,may be referredto irreverentlyas 'aatoshle'okyattsik'i;translatingthis
simplyas "OldLady'aatoshle"preservesthe originaleffectquitewell.
Exceptfor esotericoriginstories,Zufii narrativesdo not includemanywordsor
phrasesthat are clearlysacred,but they do includea fair numberof items,mostly
archaisms,whichfall betweenthe neutralandthe trulysacred.Among theseitems
are the formulas used to open and close a fictional narrative,son'ahchiand
tee
semkonikya,which Zufiis neveruse except as storyframesand which
declare
to be untranslatable.The opening formula, son'ahchi,might be
they
renderedas "Onceupon a time," which is itself a sortof untranslatable
formula,
but "Onceupon a time" suggestsa children'sfairy tale and is thereforewholly
inappropriateto most Zufii narratives.It seems best to leave these framing devices untranslated;their positionsin an otherwisetranslatednarrative,together
with a note to the effectthat they indicatefiction,should maketheir "meaning"
clear enough. A numberof past translatorsof Zufii narratives,includingBenedict, have chosento omit these formulas,but that is like leaving the coversoff a
book.
Most prominent among the longer archaicformulas used in narrativesare
greetingin Zuiii is kesshi, whichhas
greetingexchanges.The usualcontemporary
the effectof "Hi," and the replyis the sameor tosh 'iya, "So you'vecome."But
a tale character,on entering a household other than his own, may say, Hom
'aatacchu, horn chawe, ko'ndto tewanan 'aateyaye? and someone will reply,
K'ettsanisshe, ho'naawan cha'le, tosh 'iya, s'iimu. A straightforward translation of

this exchangepreservesits stiltedqualityand even a touchof its archaicconnotation: "Myfathers,my children,how have you been passingthe days?""Happily,
ourchild,so you'vecome,sit down."
The archaicinterjectionsused by charactersin serioustales are difficultto translate: as was mentionedearlier,these are not oaths, but simplygive directexpression to emotions.English interjectionshaving only covertreligious referenceor
lackingsuchreference,suchas "Wow!" "'Mygoodness!"and "Dearme!" sound
ludicrousin the mouth of a heavytale character,and those which are archaicin
additionsound even worse, "Gadzooks!""Zounds!"and "I'll be switched!"for
example. Probably most of the Zufii interjections in serious contexts should be
left untranslated; even at that most of them would require little explanatory notation, for contexts usually make their meanings fairly clear. When a young man
49 Stanley Newman, "Vocabulary Levels: Zufii Sacred and Slang Usage," Southwestern Journal
of Anthropology, 11 (I955), 345-354; reprinted in Language in Culture and Society, ed. Dell
Hymes (New York, 1964), 397-402.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

DENNIS TEDLOCK

124

who has just been turned into an eagle because his wife failed to demonstrate her
love begins his lament with hanahha! or when a father who has just been told that
his son plans to exchange bodies with a bloody dead man replies with tisshomahha!
the reader is not likely to go far astray in judging the feeling tone of these interjections; at least he will not be likely to think them equivalent to "Gosh!" or
"Good grief!"
Not all archaisms are serious-some are used to embellish humorous tales. It
is difficult to place these on Newman's slang-sacred continuum. The fact that they
are old should make the terms highly valued, but, in fact, they are employed to
make a characterseem foolishly old fashioned rather than serious or scared. They
are probably not of slang origin, but hearing these archaic phrases mouthed by
foolish characters is somewhat like hearing someone use out-of-date slang. This
makes them easier to translate than serious archaisms. A noodle named Pelt Kid,
who has just gotten married but knows nothing about sex, suddenly remembers
his grandmother's instructions and says, in his hoarse voice, 'a'ana ha'la! Hom to'
kwili yalaa teshunholh hakky'akkya, ha'holh shiwaya kwayip yam shuminnkya
kwatoky'anaknanna.50The beginning interjection, 'a'ana ha'la! is an archaism
rarely heard even in tales, and an archaic term is used for "penis" (shuminne,
sandhied and run together with another word in the quotation). The following
translation, which takes these archaisms into account, conveys Pelt Kid's ridiculousness well enough: "Golly whizz! You told me to look for two hills, and if it's
steamy there I should put my dingie in."
The onomatopoeic words in Zufii narratives may be considered a part of linguistic style since they are used more frequently in narratives than in everyday
speech, though unlike archaisms they are neutral where the slang-sacred continuum is concerned. Context usually makes the reference of onomatopoeic words
obvious enough that it is unnecessary to attempt to translate them, as in this passage (again, each line break representsa slight pause) :
'an suwe kululunanpololo
n teyatip,
(low, hoarsevoice) tu-u
'an papawilo' 'ananpololo, wilo' 'ati
w teyatikya.
(low, hoarsevoice) too
Sekwatlo'lii pottikya.
Laky'antolhihiton 'iya.
Lhiton 'ikya,ikyas
'isshakwakwahish ky'aptom'el'ikya.
His youngerbrotherrolledthe thunder
(low, hoarsevoice) tuu - n it began,
his elderbrotherrolledthe lightning,lightningstruck
w it began.
(low, hoarsevoice) too
Now the cloudsfilled up.
Here comesthe rain.
The raincame,it came
the waterreallydid comedown.51
"isshakwakwa
50 Tedlock, "Pelt Kid," (personal collection).
51 Tedlock, Tale H-9.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

125

One might render'isshakwakwaas "it splattered"and the thundersounds as

"boom"or "rumble,"but no claritywould be gained and the readerwould not
have his experienceof onomatopoeiaenrichedby the Zufii words.
III.
While it may be that past translationsof Zufii narrativeshave sufferedsomewhat from neglectof the "linguistic"featuresof style discussedabove,they have
sufferedmuch more from neglect of "oral"or "paralinguistic"featuressuch as
voice quality (tone of voice), loudness,and pausing. Boas wrote long ago that
"the form of modernprose is largely determinedby the fact that it is read, not
spoken,while primitiveproseis basedon the artof oraldeliveryand is, therefore,
more closely relatedto modernoratorythan to the printedliterarystyle."52 He
might have added, had he not so easily labelled primitivenarrativeas "prose,"
that it is also relatedto thatportionof modernpoetryin which attentionis given
to "the art of oral delivery."But Boas and his followers,in translatingoral narratives,have treatedthem as if they were equivalentto writtenproseshortstories,
except in caseswhere the originalswere sung or chanted.Jacobshas called for
a t"dramatistic"
approachto oralnarrativesandhasmadeextensiveuse of dramatic
but his translationsfollow the familiarshortstorypattern,except
terminology,53
for occasionalnotationsof voicequality.
The presenceof the tape recorderhas so far failed to wean post-Boasiansfrom
the short story approach.Systematicschemesfor the notation of paralinguistic
but such notationis not yet in wide use;
featureshave been proposedrecently,54
and no one seems to have given much though to preservingthese featuresin
translations.Yet suchfeaturesare, at leastin the Zufii case,highly "translatable,"
and it is possibleto representthem withoutmakingthe resultlook as formidable
as a symphonicscore.The necessaryliteraryconventionshavebeenthereall along,
but they are to be found in dramaand poetryratherthan in prose.Pausing,as in
two of the narrativepassagesalreadypresented,canbe representedby line breaks
as in writtenpotery;unusualloudnesscan be representedby exclamationpoints,
doubledto representextremeloudness;and unusualsoftness, togetherwith unusualvoice qualitiesand variousotherfeatures,canbe noted in parenthesesat the
of these
left-handmargin,as is commonlydone in plays.The straightforwardness
a
reader.
proceduresplacesminimalbarriersin the pathof potential
The control of volume in Zufii narrativecan be illustratedby a pasticheof
twenty of the loudest and softest lines from a storyof more than five hundred
lines; these twenty lines revealthe skeletonof the story,completewith opening
and closing formulasand the momentsof greatestemotion.For the sake of simplifying their presentationhere, I have indicatedthe soft lines with parentheses
ratherthanwith marginalnotes:
Son'ahchi!
(The little babycameout.)
Boas, Race, Language, and Culture, 491, from an article originally published in 1925.
53 Jacobs, The Content and Style of an Oral Literature, 7.
54George L. Trager, "Paralanguage," Studies in Linguistics, 13 (1958), 1-12; reprinted in
Language in Culture and Society, ed. Dell Hymes, 274-279. Robert E. Pittenger, Charles F. Hockett,
and John J. Danehy, The First Five Minutes (Ithaca, N.Y., 196o), 194-206.
52

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

126

DENNIS TEDLOCK

("Where is the little babycrying?"theysaid.)

(He was nursed,the little boy was nursedby the deer.)
("I will go to KichinaVillage, for he is withoutclothing,naked.")
(When she got backto herchildrentheywere all sleeping.)
"He saw a herdof deer!
But a little boywas amongthem!"
"Perhapswe will catchhim!"
Then his deermothertold him everything!
"Thatis whatshe did to you, she justdroppedyou!"
(The boybecame
veryunhappy.)
And all the people who had come killed the deer,killed the deer, killed the deer!
(And his uncle,dismounting,
caughthim.)
"Thatis what you did andyou aremy realmother!"
(He put the quiveron andwent out.)
(There he died.)
This was lived long ago! Leesemkonikya!5

The extremesof loudnessand softnessoverlapin functionin thattheyboth draw

special attentionto a line. The softness of "He was nursed,the little boy was
nursedby the deer" seems more appropriatethan a loud rendition,and the line
aboutthe killing of the deerseemsproperlyloud, but someotherlines couldhave
been renderedeitherway, "Buta little boy was amongthem,"for example.
The manipulationof voice quality in Zufii narrationhas a diversitywhich I
have only begun to explore;only a few examplescan be given here. One of the
narratorsrepresentedin my collection delivers the opening lines of his tales,
including formulasand the namesof the majorcharactersand the placeswhere
they live, with a formalitywhich approachesthat of a chant: his stressesare
heavier,his enunciationmore careful,and his pitch controlgreater (but not as
great as in singing) thantheywould be in his normalnarratingvoice;as he moves
into the first events of the story this formalityslowly dissolves, over the space
of eight or ten lines, until his voice is normal.The only other fully predictable
manipulationof voice quality on the part of this same narratorinvolves the
quotationof story characters:the words of the 'ahayuuta(twin boys, the war
gods), for example, are usually deliveredin a high, raspyvoice, and most female characters,except where their speechesare long, are given a tense, tight
(but not high) voice. Since a native speakerof English might prefer to render female voices in narrativesby raising his pitch, one might "translate"the
Zufii "tight" voice into an English "high" one. Neither of these practicesis a
more objectiverenditionof the femalevoicethanthe other;both representa selective imitation of the common properties of female speech.6""
There are many less conventionalized (and less common) uses of voice quality
55 This and all further Zufii narrative quotations are from the Stony Brook version of "The Boy
and the Deer." The full story will appear in translation only, but for the present excerpts I have
supplied a parallel Zufii text wherever this seemed necessary.
56 Tightness, or "squeeze," is more common among women than among men, according to
Pittenger and others, 202-203.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

127

in Zufii narratives, two examples of which will suffice here. When a characteris
trying to pull some tough blades loose from a yucca plant, the narratormay render
"He pulled" with the strain of someone who is trying to speak while holding his
breath during great exertion. When a passage involves intense emotion, the narrator may combine the softness mentioned earlier with a break in his voice, as
if he felt like weeping. The use of this voice technique is exemplified in the following passage in which a man is killing three deer who are the foster mother
and siblings of his nephew:
The thirduncle
(softly, voicebreaking) droppedhis elder sister
his elderbrother
his mother.
Loudness and voice quality are obviously worth noting, but it seems to me that
pausing is foremost among the paralinguistic devices that give shape to Zufii
narrative and distinguish it from written prose, and the same could probably be
said of many other oral narrative traditions. Stravinsky has said, "I dislike the
and he could have
organ's legato sostenuto, . . . the monster never breathes,"''57
said the same thing of written prose. The spoken word is never delivered in the
gray masses of boxed-in words we call prose; indeed, according to GoldmanEisler, as much as half the time spent in delivering spontaneous discourse is
devoted to silence, and "pausing is as much a part of the act of speaking as the
vocal utterance of words itself."'8 But of all the past anthropological collectors
of so-called prose narratives, only one, Paul Radin, seems to have shown any real
sensitivity to pausing. For several passages from Winnebago texts he marks
pauses of three different lengths; he also breaks these passages into lines. Here
his intention is unclear: each line break coincides with a pause, but there are also
pauses within lines."9 Unfortunately he preserves neither pause marks nor line
breaks in his translations.
In dealing with the pauses in Zufii narrativesI have found it best to divide them
into two types: "ordinary"pauses, represented by line breaks, and "long" pauses,
represented by double spaces between lines. I initially spotted pauses only by ear,
running through the tape of a half-hour narrative several times. An oscillograph
of the same tape later revealed that my "ordinary" pauses ran from four-tenths
of a second to two seconds, with the average at three-fourths of a second. The
longer pauses ran from two to three seconds. Some other listener might come up
with slightly different boundaries for his ordinary and long pauses than I did, or
he might want to make more than two distinctions; but in any case, given a reasonably good ear, he could probably make fairly consistent notations without the aid
of an oscillograph.
Intonation poses no great problems where Zufii pausing is concerned. Except
for the special intonational device used to lengthen time or space (discussed
7 Igor Stravinsky, album notes to Symphony of Psalms, Columbia Records 6548.
Frieda Goldman-Eisler, "Discussion and Further Comments," in New Directions in the Study
of Language, ed. Eric H. Lenneberg (Cambridge, Mass., 1964), 118-119.
59 The Culture of the Winnebago: As Described by Themselves, Memoirs of the International
Journal of American Linguistics, 2 (1949), 42-44, 61-62, 103, io6-io8. I infer from Radin's
remarks on p. 42 that these pauses were reconstructed rather than recorded in the field.
58

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

128

DENNIS TEDLOCK

earlier) and a few other, rarer deviations, Zufii narrative patterns can be covered
by a general rule rather than marked for each line: the boundary between one
intonation contour and the next is strongly marked where a change of phrase or
sentence corresponds with a pause or where a quote begins, and less strongly
marked where a change of phrase or sentence occurs within a line or where a
pause occurs within a phrase. This pattern seems close enough to the normal
tendencies of an English speaker so as to create no translation problems. As far as
the internal details of the contours are concerned, the typical Zufii contour does
not happen to be very different from that of a declarative sentence in English,
but it should not matter if the two contours were very different: what is important
in translation, except for deviations from typical patterns, is the boundaries between contours.
The following passage, with silences and intonation contours as indicated above,
will serve to illustrate most of the properties of pausing in Zufii narrative ("they"
in the first line refers to a herd of deer) :
Yam telhasshik'uushinayalhtookwin'aawanuwa'aayemakkya.
'aayemaknalesnolh
chimkwat 'iskon 'aateyatom sunnhaptutunaapaniye.
'aateya'kya - koholhlhana
'ist
'an lhuwal'an
'an kyakholh
'imatlhatakky'an'aakya.Lhatakky'an'aana
'imatpaniinas'ist
'uhsi lak 'ist
wi'ky'al'anholhlesnapaniinauhsist lak

o10

k'uushinyalhtan'uhsitewuuliyalhtookwinholh'imatky'alhkonholhyemakna.
They went backup to their old home on the PrairieDog Hills. Having gone up
theywereliving thereandcomingdown only to drinkin the evening.
They lived on------ for sometime
until
from the village
his uncle
went out hunting.Going out hunting
he camealong
down around
Worm Springand from these he went on towards

10

the PrairieDog Hills and cameup nearthe edge of a valleythere.

The problems encountered in preserving the original pauses in English are minimal. Occasionally Zufii word order makes the transposition of lines or part of
lines desirable, but this can usually be done without serious distortion of the effect
of the original: in the above passage no transposition seemed advantageous,
except that "down" in line 9 of the translation is a partial rendition of paniinas
in line 8 of the text, but elsewhere in the same story a literal rendition would

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

129

produce lines like, "Her clothes / she bundled," and "His kinswoman / he beat,"
which call for transposition.
Where the length of lines is concerned, it would be difficult and foolish to
slavishly follow the exact Zufii syllable counts in translation, but it is possible to
at least approximate the original contrasts in line length. The importance of such
an approximation may be seen from the fact that the length of lines-or, to look
at it in another way, the frequency of pauses-is the major source of apparent
variations in the rate at which human speech is delivered. Passages with short lines
(many pauses) will seem slow, while those with long lines (few pauses) will
seem fast.60 In the above passage, the narrator rapidly tells of the deer-herd's
residence on the Prairie Dog Hills (lines 1-2), then slows down, with suspenseful effect, as the man goes out hunting (lines 4-9), and finally speeds up again
with the excitement of the man's arrival at the Prairie Dog Hills (line ii).
Preserving such patterns in narrative pace obviously precludes the insertion of
any but the smallest bits of "explanatory" material by the translator: where the
Zufii word lapappowanne means "a headdress of macaw tail-feathers worn upright at the back of the head," for example, he will have to settle for something
like "macaw headdress" in his translation and leave the rest to a note or a picture,
though Cushing might have done otherwise. And where it is frequently unclear
which charactersare responsible for quotations, as in Clackamas (but not Zufii)
narratives, the translator may find it best to place the names of the speakers outside the main left-hand margin, as in a play.
One of the most striking things about the lines in Zufii narrative is that they
are not always dependent on the major features of syntax. In the above excerpt
some of the pauses do correspond with changes of phrase or sentence, but five of
them (the pauses following lines 4, 5, 6, 9, 10o) leave the hearer hanging, syntactically speaking, thus adding to the suspense already noted for this passage.61
The longer pauses in Zufii stories often correspond to sentence boundaries, but
in the present excerpt they occur between two phrases of the same sentence (after
line 3) and in the midst of a phrase (after line io). The first of these pauses is
a sort of paragraph marker between the affairs of the herd and the hunting
expedition of the man; its location within a sentence keeps the listener on the
string in much the same way that the placement of a chapter division within an
episode (instead of between episodes) keeps the reader of a novel on the string.
With the second of these pauses the narrator keeps the listener dangling for a
moment and then suddently lets him know, in the first words of the next line,
that the hunter has arrived at the Prairie Dog Hills, where the herd is.

IV.
The treatment of oral narrative as dramatic poetry has a number of
analytical
advantages. Some of the features of oral narrative which have been branded
"primitive," on the basis of comparisons with written prose fiction, can now be understood as "poetic" instead. It has been said, for example, that while most of our
60
Goldman-Eisler, i20; she adds that the rate of syllable articulation (between pauses), by
contrast with the rate of pausing, is almost constant.
61 One-third of the lines produced by my principal Zufii narrator involve this kind of phrase
splitting, which is twice the proportion of splitting (or "necessary enjambment") reported for
Yugoslav epics, by Albert Lord in The Singer of Tales (New York, 1965), 54.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

DENNIS TEDLOCK

130

own prosenarrativeis highly "realistic,"primitivenarrativeis full of fantasy:a

stone moves about like an animal, an animal speakslike a man, a man jumps
through a hoop and becomesa coyote. Yet when we encountergross and unexplaineddistortionsof realityin Yeats, for example,we are apt to call them not
"primitive"but "dream-like"or "mystical"and to regardthem as highly poetic.
It is also said that "primitive"narrative,again unlike written prose fiction,
seldom describesemotionalstates.This is true enough, but the comparisonwith
prosemissesthe point: what oralnarrativeusuallydoes with emotionsis to evoke
them ratherthan describethem directly,which is preciselywhat we have been
taught to expect in poetry.In the Zufii case, such descriptionsof emotionsas do
exist arevery simple, "The boy became/ veryunhappy,"for example,but evocationsaremyriadandsometimesquitesubtle,asin thispassage:
He went out, having been given the quiver, and wandered around.
He was not thinking of killing deer, he just wandered around.
In the evening he came home empty-handed.

Accordingto both the narratorand a memberof his audience,these lines clearly

indicate (to a Zufii, at least) that the person referredto is depressed,and they
regardedthis person'sdeaththreedayslateras a sortof suicide,thoughit was describedin the storyas an accident.
Another distinguishingfeature of "primitive"narrative,accordingto Boas
and manyothers,is repetition,rangingfrom the level of wordsor phrasesto that
of whole episodes.62At least one of the kinds of repetitionin Zufii narrativeis
indeed rarein our own prose (and poetryas well), and that is the linking of two
sentencesor majorclausesby the conversionof the final elementof one into the
initial elementof the next, as in these lines (from the last passagequotedin the
previoussection): "His uncle / went out hunting.Going out hunting/ he came
along ..." But the same device is commonin epic poetry,as in this Yugoslav
example:"And mayGod too makeus merry,/ Makeus merryand give us entertainment!"63Unless we want to call epic poetry"primitive,"this particularkind
of repetitionmustbe properlyunderstoodas "oral"and not "primitive,"and the
samething goes for the repeateduse of stockformulasin both epic poetry (epithets, for example) and Zuiii narrative(greeting exchanges,for example).
When it comesto the repetitionof whole passages,"primitive"narrativemay
be comparedto epic poetryand also to refrainsin songs (from both literateand
nonliteratecultures) and in writtenpoetry.Refrainsare often varied from one
renditionto the next, and the sameis true (althoughin a less structuredway) for
the repeatedpassagesin Yugoslavepic, as shownby Lord,64andin Zufiinarrative.
In the following Zufii passage,a boy's foster motheris quotingto him what he
mustsaywhen he addresseshis realmother,who abandonedhim as a baby:
My Sun Father
made you pregnant.
When you were about to deliver
it was to Nearing Waters
that you went down to wash. You washed at the bank.
62 Boas, Race, Language, and Culture, 491-493.
63 Lord, 32.
64

Ibid., 82.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

13I

But when the boy actuallyconfrontsthis realmotherlaterin the samestory,this is

whatthe narratorhashim say:
My Sun Father
madeyou pregnant.
Whenhe madeyoupregnantyou
sat in thereandyourbellybeganto growlarge.
Yourbellygrewlarge
you
you were about to deliver,you had pains in your belly, you were aboutto give birth to
me,youhadpainsin yourbelly
yougatheredyourclothes
andyouwentdownto thebankto wash.

The remainingkinds of repetitionin Zufii narrativeare of the sort we approvingly call "parallelism"(or somethingof the sort) when we find them in our
own poetry. A line like, "And all the people who had come killed the deer,
killed the deer,killed the deer!"cannothonestlybe calledprimitiveunlesswe call
Shakespeareprimitivewhen he has Hamlet say, "You cannot,sir, take from me
any thing that I will more willingly partwithal: exceptmy life, exceptmy life,
exceptmy life." And not all of the parallelismin Zufii narrativeinvolvessimple
repetition:
Tewuuli kolh nahhayaye.Nahhayap
lalholh 'aksik ts'an 'aksh 'allu' 'aye, kwan lheyaa k'ohanna
Muusilili lheya'kwip 'an lapappowaye.
Lapappowlesnish 'aawanelap,ten 'aktsik'i
'ottsi
ho"i 'akshappa.
In the valleywas the herdof deer.In the herdof deer
therewas a little boygoing aroundamongthem,dressedin white.
He had bells on andwas wearinga macawheaddress.
He was wearinga macawheaddressandwas handsome,surelyit was a boy
a male
a personamongthem.
What all this means, simply stated, is that (remarkably enough) there was a
human being among the deer, but the narrator chooses to explore the fact in
half a dozen different ways.
Repetitions and other poetic features of oral narrative have implications even
for those who focus on content analysis and choose to ignore "style." The implica-

tions for psychologicalanalysis,which is normallybasedon the contentof prose

translations, may be illustrated by the following passage, in which a boy has just
exposed the woman who secretly abandoned him as a baby (parentheses indicate
softer portions) :
At thatmomenthis mother
embracedhim (embracedhim).
His unclegot angry(his unclegot angry).
He beat

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

132

DENNIS TEDLOCK

hiskinswoman
(he beathiskinswoman).
This passagemight haveappearedin a conventionalproserendition(by Benedict,
for example) as, "At thatmomenthis motherembracedhim. His unclegot angry.
He beathis kinswoman,"thus having lost the nuancesand greaterintensitygiven
it by the repetition,the changesof loudness,andthe frequentpauses.
The complicationsof poetic style have especiallystrongimplicationsfor those
who seek to measurethe social and psychologicalcontentof narrativeby means
of word-counts."Killed the deer," repeatedthree times in a line quoted above,
might well have been reducedto a single occurencein the translationsof the
past. Moreover,it seems crudeto give the same weight to a word like "killed"
when it is shoutedand when it is renderedflatly.And the indirectexpressionof
emotion, as in the case of the depressionand suicide mentionedearlier,would
escapea wordcounterentirely.
L6vi-Straussand other structuralists
operatingon an abstractlevel assumethat
almostany translationwill do for their purposes,but poetic subtletieshave a potential for radicallyalteringsurfacemeanings,irony being an obviousexample.
The more concretestructuralanalysisproposedby Hendricks,on the otherhand,
does take the "linguistic"aspectof poeticsinto account,since eachbasicelement
in his systemconsistsof a single semantic"function"which may be servedby
severallower-level"linguistic"(phonological,morphological,or syntactic)elematters,thoughit is prements.65But even Hendricksoverlooks"paralinguistic"
wall between "linguisthat
the
function
at
the
level
of
semantic
arbitrary
cisely
for
In
tics" and "paralinguistics"
narrative,
example,the semantic
collapses. Zufii
functionof markingthe startof a quotationmay be servedby such "linguistic"
devicesas a sharpintonationchangeor the words, "The deer spoke to her son,"
devicesas a pauseor a change
but it may also be servedby such "paralinguistic"
in voicequality.
The treatmentof oralnarrativeas dramaticpoetry,then, clearlypromisesmany
analyticalrewards.It should also be obvious that there are immediateesthetic
rewards.The apparentlack of literaryvalue in many past translationsis not a
reflectionbut a distortionof the originals, caused by the dictationprocess, an
emphasison content,a pervasivedeafnessto oral qualities,and a fixed notionof
the boundarybetweenpoetryand prose. Presentconditions,which combinenew
recordingtechniqueswith a growingsensitivityto verbalartas performed"event"
ratherthan as fixed "object"on the page, promisethe removalof previousdifficulties."Event"orientation,togetherwith an intensifiedappreciationof fantasy,
has alreadyled modernpoets to recognizea kinshipbetweentheir own work and
the oral art of "primitives."As JeromeRothenbergpoints out in Techniciansof
the Sacred,bothmodernandprimitivepoetsareconcernedwith oralperformance,
both escapethe confinesof Aristotelianrationalism,both transcendthe conventional genre boundaries of written literature, and both sometimes make use of
stripped-down forms which require maximal interpolation by audiences.66 This
65 William O. Hendricks, "On the Notion 'Beyond the Sentence,' " Linguistics, 37 (1967),
32-35.
66
Jerome Rothenberg, Technicians of the Sacred (Garden City, N. Y., 1968), xxii-xxiii.

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions

ON THE TRANSLATION OF STYLE IN ORAL NARRATIVE

133

last point recalls the Clackamas"terseness"discussedby Jacobs,and I am reminded of the Zufii who askedme, "When I tell these storiesdo you pictureit,
or do youjustwriteit down?"
The effortpresentedhere is intendedmore as an experimentthan as the final
word on the poeticfeaturesof oral narrativeand theirpresentationon the printed
page. I hope it will encourageothersto makefurtherexperiments.
WesleyanUniversity
Middletown,Connecticut

This content downloaded from 209.181.210.66 on Sat, 28 Jun 2014 21:26:43 PM

All use subject to JSTOR Terms and Conditions