The Polysemic Analysis of Oral Literature (1985)

The Polysemic Analysis of Oral Literature
Author(s): Erik Schwimmer

Source: Style, Vol. 19, No. 2, Orality and Rhythm (Summer 1985), pp. 213-226
Published by: Penn State University Press
Stable URL: https://www.jstor.org/stable/42946195
Accessed: 13-10-2019 14:13 UTC
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
Penn State University Press is collaborating with JSTOR to digitize, preserve and extend
access to Style
This content downloaded from 196.224.27.30 on Sun, 13 Oct 2019 14:13:00 UTC
All use subject to https://about.jstor.org/terms
Erik Schwimmer
The Polysemie Analysis

of Oral Literature*
Recent advances in semiotics enable us to analyze certain cultural pat-

terns of preliterate and newly literate societies that were not seriously studied
in the past. I am referring to polysémie discourse, that is, discourse in which
many messages on different levels and planes of the discourse are ambiguously
organized (Eco 271). While such discourse occurs, of course, in all cultures as
an "aesthetic idiolect," it tends to be more general and more widespread in
small face-to-face communities where language education is left to family groups
rather than to a school system. Even those who have no firsthand knowledge
of those societies will recognize the pattern from the "colorful speech" in
certain peasant novels and in literature written in major western languages by
African, Caribbean, and Oceanian authors who convey the "poetic" quality
of the language spoken by their characters.
What is said here of language is equally true for other aesthetic message
systems- pictorial, musical, and choreographic- and includes all message sys-
tems that normally form part of artistic forms such as theater, opera, and
ballet. As far as Papua-New Guinea is concerned, these theatrical forms should
perhaps be the very first to be considered as by far the greatest productive
effort goes into these multimedia forms reserved for the great ceremonial events.
"Oral Literature" covers only a small and minor part of this field. Re-
garded from the vantage point of Papua-New Guinea for example, oral liter-
ature does not appear in these great ceremonies except in chants and spells
and sometimes brief dialogues. Most Papua cultures have a rich store of nar-
rative oral literature, but this is produced outside the big ceremonies, even
though it very frequently refers to them.
In addition to their literary value, these texts are of much epistemological
interest to anthropologists. While ethnographers do rely on visual information,
on brief items of news, on answers to their questions, their valuable insights
* This paper was presented at a plenary meeting at the International Semiotic Institute
Summer School, Session 1985, at Bloomington, Indiana. It was revised after this discussion with
summer school participants following suggestions by Ludomír Doležel. The research in Orokaiva
oral literature reported in this paper was undertaken in 1966-67, 1970, 1973, 1981, with the
financial support of the SSHRC of Canada.
STYLE: Volume 19, No. 2, Summer 1985 213
214 Erik Schwimmer
depend a great deal on

give them at length an
the ethnographer's sup
ethnographer ought to
own premeditated sch
ethnographer what we
full or fragmentary.
A major problem in
tation of these texts. V
nographer. He could, o
understand, but he is
After a sufficiently l
reduced. On the other h
his own text, in which
text is not a mere repl
from protracted dialo
the narrator while stu
matic in view.
Now this interpretation tends to use the informant's text much as a
literary critic may use quotations from a work he studied. The critic quotes
only what he "needs" and rightly so, as the reader can refer to the work studied.
It is different in the case of the ethnographer, because the work studied is made
up of his own unpublished field notes which may contain the complete literary
and ethnographic works of the people he calls his "informants." Nobody ever
hears the latter.
Several problems arise out of this, partly ethical and partly scholarly. On
the ethical side, we note rapid improvement because the cultures in question
are mostly much less colonized, much more able to lay down the law to the
ethnographer and because the informants are much more educated and, as the
various populations begin to produce their own academics, more able and
eager to do their own studies.
On the scholarly side, intensive experimentation has begun with the form
of ethnographies. This is a serious matter as ethnography has become a literary
genre in its own right, with its own rhetoric and its own authority. This genre
would be profoundly transformed if, say, texts collected in Papua were to be
given a prominent place in it. It would certainly be of little use to simply
reproduce them verbally without detailed commentary, since these texts, even
when translated into a western language, would make little sense to a western
reader. In fact, they made just as little sense to the ethnographer when he first
arrived on the scene, and the main result of his protracted stay in the "field"
is that at the end of his stay those texts did make sense to him.
Fieldwork, then, is semiosis, and the interpretation of the texts in question
is the quintescence of that semiosis. One of the key activities of semiotic
Oral Literature 215
anthropology is there
ples studied. This inv
with most traditiona
evidence of polysémi
because such analysis
the reader of the eth
he cannot rely on his
As so often happens
of a discipline the s
scholar faced with
solution with a litera
that the problem is
been made.
Malinowski discusses in much detail the problems polysemy poses to the
ethnographer and insists that it is unrelated to "mental confusion, poverty of
language, wanton or careless usage" (72). His own chief contribution to this
field was in the analysis of the magical word and the magical formula.
With regard to the Lévi-Straussian structuralist method, there is no doubt
it is polysémie. The fullest example we have of structuralist polysémie analysis
is the paper on Asdiwal, where a number of levels of conceptualization are
postulated and demonstrated; the story moves at all these levels at once. Mean-
ings were suggested at each level separately, supported each time by the kind
of comparative analysis characteristic of the method. Finally, it was shown
that the message on each of the levels formed, in the end, a coherent total
message in which contradictions in social structure were presented as logically
prior to all the others. In later work, Lévi-Strauss often used similar methods,
but he never treated polysemy in the same detail.
One interesting point, in his attempts as well as all others at polysémie
analysis, is that the levels of conceptualization never arise spontaneously from
the text. They follow standard anthropological practice: kinship and marriage,
agriculture, fishing, and hunting, the geographical and weather factors, eth-
nozoology and ethnobotany, ethnoastronomy, and so on are related to religious
and ritual codes, to sorcery practices and in that sense there is a plurality of
levels or conceptualizations. An obvious logical query about such a procedure
would be to say that the culture in question may not use such a classification;
and that there is congruence between the various levels, so that in the end
there may have been no polysemy except in the eye of the investigator.
Such a query is in principle valid, and we must assume that Lévi-Strauss
would not have proceeded as he has if he had not known that the kind of
cultures he was analyzing really do have something like the cognitive processes
he ascribes to them. Certainly, such processes have been very well described
by Maurice Leenhardt for Melanesia, not only in the analysis of narratives but
also in a wide range of everyday discourse. He rejected the theory of "mystical
216 Erik Schwimmer
participation" of Lévy
"mythic." Here a num
particular totem, a par
bark and all its other
spondences also appear
culture learn them fro
that make the narrativ
the narratives in that w
but not knowing the c
will point them out as
while before the infor
something so simple. T
culture but hard to lea
mythical." It implies
mental categories, no
More recently, at leas
in New Guinea myths
analysis somewhat sim
building blocks" by tri
erated by man's rene
polysémie because each
but another order is a
The second attempt,
aesthetic insights, is Y
Leenhardt, all the leve
of the hero serve as a
leading informants. M
their lives and to the
enter history as the p
The mythical, historic
of meaning.
II
After these importan

ysis of polysemy in N
oral literature cannot b
to which it is closely rel
is in fact the only aest
the great ceremonies.
The principal sculptu
music- to mention onl
Oral Literature 217
give rise- all serve th

These societies have
sages in all media ar
abundance on the in
plenished during th
ceremonial occasions
exchange and recipr
ogy, we might speak
symbolic production
As far as this total
constraints to its ac
oral literature which
production cycle. It
great ceremonies but
social life and natur
temporary historical
Before turning to
to summarize briefly
most traditional wes
sculturally, the class
many difficulties. H
conventionally calls
tinguish their "myth
so. Detienne, Vernan
genre generated by w
Greeks.
Yet there is genera
traditional Oceanic
symbolic correspond
manence of the wh
genre, as, for instan
might say, to a styl
The first rule, prop
ner, but most clearl
folktales are divisibl
contains what one m
closures corresponds
sion of how such seg
such rules of narrati
Wagner suggests int
order while in folkt
A change in the or
in the tale's message
218 Erik Schwimmer
not remain meaningfu

but the outcome and t
The second rule has t
Greimas, and Meletin
distinguishing between
ciate. Anthropological
us a formalist idea of t
and folktales are cultur
who knows how to mak
text, distinguish from
As soon as a tale appr
elements of the struct
This may confuse the
derstand." Quite apart
tales are apt to be poly
the action by type, be
is often impossible to s
Finally, there tends to
of myth progress with
mythic character like
depiction of transform
aspects of world order
personal "character."
The third rule has to
"mythic." Meletinsky h
the adversaries are usu
the interests served by
are personal. In a myt
he is typically a victim
a world where men an
is not yet instituted, a w
without marriage, whi
The fourth constitue
digmatic one, referred
the oppositions present
world. If we did no par
not be revealed and we
what our informants me
We believe Wagner's m
Here it is again the r
richness of the codes.
naments, marriage par
of construction of the
Oral Literature 219
paradigms of folktal
such as high/low, ow
tiplicity of meanings
The fifth constituent of the code of oral literature is what one sometimes
calls the "game rules" binding upon the hero. They are of three kinds. First,
they flow from the general rules of social exchange which the hero invariably
obeys and which the villain does not obey. Second, they are specific rules laid
upon the hero by helpers and by reliable authority figures: they define the tests
the hero must undergo. The third type of rules flows directly from our con-
notations of the status of the hero. Picking the highest and most dangerous
fruit from the tree represents in Meletinsky's excellent phrase, "a kind of
implication for the hero" (106). Such rules are more universal than any others
and exist equally in myth and folktale, but Meletinsky believes that rules of
behavior have the character of game rules "to a much higher degree" (31) in
folktale than in myth. Here again, the difference may be one of style as the
game rules seem to us as constraining in the one case as in the other, but it
is true that in myth they are often implicit and common knowledge to members
of the culture.
Ill
We need not repeat here our earlier demonstration (Schwimmer "Myth")

of how the collection of myths is an essential part of the ethnographer's dis-
covery procedures and how myths and literature give answers that can be
obtained from no other source to questions of social anthropology. But we
need to add one point not made in that earlier essay: the full information held
by an informant about a particular mythic theme is perhaps never stated in
any one performance of a myth. Every performance is shaped by the situation
in which it occurs: inside the family, it may be transmitted to pass on magical
knowledge or knowledge about land rights or about the family's present tally
in an old vendetta. Outside the family, some details may never be revealed:
secular names are substituted for sacred ones, and so on. On the other hand,
literary embellishments may be added, and general meanings may be developed
more clearly.
When the public includes an ethnographer, themes that are usually left
implicit may be explained in more detail so as to help the stranger understand
some cultural meanings. In other contexts again, as when a man inquires about
his land rights, the myth's territorial implications may be more fully spelled
out than usual. The presentation, then, is always influenced by the pragmatics
of the situation- the question asked, the status of the person addressed, a
particular result envisaged, a right asserted or denied.
220 Erik Schwimmer
Once such a pragmat

essentially theatrical n
literature analysis afte
guage to the profane a
to the profane, he cho
meaning to the profan
myths are called polys
is direct and explicit an
only.
In my study "Power and Secrecy," I described this pattern of ambiguities
as it emerged from published data of the initiation ceremonies of a number
of Papua-New Guinea cultures. Ambiguity always involves pretending that
there is a spirit which remains hidden while, on another level of reality, the
spirit is visible and an impersonation. Revelations staged for the profane al-
ways have their counterpart in secret theatrical props and carefully rehearsed
deceptions from inside the cult house. Oral literature is thus a reflection of
this dual nature of symbolic production in societies practicing initiation.
There is, however, no reason to believe that such societies- like the Bak-
taman described by Barth- actually deprive all but a small minority of their
citizens from adequate information about the "mythic" content of their own
culture. First, a scrutiny of Barth's statistics shows that even by his reckoning
all adult males except one who was mentally retarded were receiving this
knowledge. Second, a comparative study of twenty-two Papua-New Guinea
societies shows that the knowledge withheld does not relate to the basic ideas
constituting the "mythic" but rather to the specific cult secrets which are either
magical formulae or else the kind of secrets analogous to what theater people
would call the "backstage" secrets of a performance.
The difference between the western and Melanesian "backstage" or the-
atrical illusion lies in the way the two traditions conceive of the personages
portrayed on stage. For the western theater, these personages are ultimately
part of theatrical illusions. Some great dramatists like Shakespeare have indeed
maintained that the stage was the whole of life, that these personages represent
the whole of reality, but even Shakespeare made that suggestion only in the
context of a play and of pronouncements of particular stage characters. Me-
lanesians, on the other hand, believe this to be literally and totally true: the
spirits they put on stage represent to them an ultimate reality. Hence, they are
not mere impersonators but at the same time the only possible vehicles for
this ultimate reality to be presented to a public. Their position differs therefore
intrinsically from the actor's position as described in Fischer-Lichte. They
believe in no ultimate reality other than the one their illusions present on stage.
We have completed several analyses of oral literature where we applied
the above model of ambiguity. The first of these was a paper on Orokaiva
"plant emblems," carried and exhibited by individuals in numerous contexts
Oral Literature 22 1
as "icons of identity
showing the explicit o
texts appear to treat
gradually become awa
plant emblems are so
A second possible re
events in the world o
of the living to that
but something akin t
in which the spirit of
new patrilineage with
A second applicatio
dealt with another ca
New Guinean male co
linked by a symbolic
or gestural exchange
are severely excluded
designating no other
point of simulating c
In a third application
tale with some intert
bolic exchange occurr
fies, or perhaps betw
a male couple there
profane were worse t
a wrong turning and
manages to open them
efficacy of these the
drawn into political a
behavior but they are
of the institution of
society, and so on.
We may now furthe
cultural transformat
did in a recent paper
on "Political Discour
Orokaiva informants and one of their relatives who has become a historian.
They converse in various forms of backstage chatter but the end product of
this is a kind of public performance: the historian explaining the Taro Cult to
an audience made up of academics and Papua-New Guinean intellectuals from
other tribes. In principle this would be a comprehending and sympathetic, but
non-Orokaiva, audience.
222 Erik Schwimmer
This paper is a furthe

starts with the overt a
the face of it, purely c
As analysis proceeds, w
the taro field becomes
When we come to th
production technique
code. It is only at the
an historian born in t
the level of political or
tional characteristics o
Far from being a mer
trained Orokaiva inter
seems to us a foreshad
ford; Schwimmer "Tas
munications. He reduc
pects where his auth
incorporates anthropol
The presence of the a
the model of the adeq
foreign anthropologi
reconstruct the ideolo
one example where th
The text, it should be
of sequences selectivel
such an ideological pro
with the considerable
essentially the discour
IV
Finally, we must exp

thetics of oral literatu
creators to confine ou
semiotics of oral litera
constraints upon the ae
the "mythic." These c
sistency, religious or
though these constrain
literature from produ
object."
Oral Literature 223
The chief problem

psychologically cred
universe on many d
idiolect," and it is ac
pluses of expression
notion of 'cosmicity
art" (273). Now this
pation," the "pre-log
Even if this applies
the state of mind of
a taro cult. For our t
in a garden, to speci
precisely a theologic
Such texts are "my
as the reader cannot
larity. The "mythic"
close to what Eco cal
differences between
clearer if we compar
in much detail in Bar
of his method depend
codes are seen to int
as Stephen Heath ca
Some of Barthes' fi
and have often been
of myth, equating
according to which my
recognize the "refere
that the narrative could be reduced to that code.
Similarly, the "symbolic" code, or as Barthes calls it, the antithetical
code, is well-known from Lévi-Strauss's structuralist analyses of myths. There
is no doubt that these logical antitheses are an important part of the "meaning"
but again they do not exhaust it. The same is true for what Barthes calls the
"actional" code, again familiar to us from the work of scholars such as Propp,
Meletinsky, Greimas, and many others. An "aesthetic" reading of the text,
then, recognizes the relevance of all of these but sets them side by side as
readings that ought to be made conjointly and simultaneously.
Barthes adds two further codes or "readings" that will be recognized by
literary theorists as supremely important and that have often been undervalued
by anthropologists and folkorists. These are the "hermeneutical" and "semic"
codes. The former, it will be remembered, is a term used by Barthes in a very
particular way. The "hermeneutic" question of S/Z was the apparently simple:
"Sarrasine, what is that? a common noun? a proper noun? a thing? a man? a
224 Erik Schwimmer
woman" (24)? The wh

discovery of the ans
respond extremely w
myths, how the prese
ually emerged from a
One can certainly rea
gradual emergence of
that such a reading ad
value of the texts.
The most difficult problems are raised by Barthes' "semic" reading: that
is, the analysis of the connotations of the Orokaiva (in our case) words of the
texts. These semes, says Barthes, are unstable and dispersed throughout the
text, like particles of dust, like meaning flashed by a mirage. Such a description
of semes is well fitted to Balzac's prose but in Orokaiva oral literature (and
other oral literatures we are familiar with) such devices are rare indeed.
Under what circumstances would a Papuan storyteller subtly hint (and
he is very capable of subtle hints- there is no doubt of that) at the "moon-
like" quality of a particular character, of his "sélénité" (Barthes 31)? This could
happen if the character turned out later to be the moon, or if he went to live
on the moon, or if he was a spirit identified with the moon. Even then the
Orokaiva word for moon suggests menstruation, agricultural activity, pollu-
tion, sequestration, or perhaps a girl's seclusion ceremony (at her first men-
struation). This is not because the Orokaiva does not value romantic land-
scapes, but because such landscapes have to him all the socioreligious referents
just mentioned. More generally, most Orokaiva semes tend to be read simul-
taneously as "references" or allusions to institutions of one sort or another.
Is it possible then that such literatures are lacking in "connotations" in
the western sense of that term? Both Elli Köngäs Maranda and Erika Fischer-
Lichte have suggested that such cultures have little room for connotation as
they are highly homogeneous,- too homogeneous to have idiolects. Such a
view ignores, however, that these cultures are in fact full of idiolects in dancing,
in painting, and in many other forms. In Papua, New Guinea, there is no lack
of individualism or freedom to express individual experience.
It is therefore wiser to make the more limited observation that mythic
literature, because of its particular cognitive constraints mentioned above, does
not have the same degree of freedom as some other media and that, for this
reason, one cannot just use a word like "moon" as a literary image, because
it is far too loaded with precise and irrelevant symbolic referents.
We have ourselves begun to explore this difficult field of Papuan aes-
thetics, and two articles are now in press that try to resolve the paradox that
on the one hand Papuan art in all media is principally concerned with bringing
spirits to life and making them visible or perceivable. At the same time there
is much emphasis on mimesis : that is, of copying accurately the appearance
Oral Literature 225
and behavior of entit

the Papuan also are n
As a provisional co
ganization of messag
spirits are present. S
and its literary efficac
device elaborated by
sistent with the cogn
Works Cited
Barthes, Roland. S/Z. Paris: Seuil, 1970.

Barth, Fredrik. Ritual and Knowledge among the Baktaman of New Guinea. Oslo:
Uni versitets vorläget, 1975.
Ben-Amos, Dan. "Analytical Categories and Ethnic Genres." Genre 2 (1969): 275-301.
Clifford, James. "De l'autorité en enthnographie." L'Ethnographie 79.2 (1983): 87-1 18.
Coseriu, Eugenio. "Sistema, norma y habla." In Teoria del lenguaje y lingüistica general.
Madrid: Gredos, 1962.
Detienne, Marcel. "Une mythologie sans illusion." Le temps de la réflexion 1 (1980):
27-60.
Eco, Umberto. A Theory of Semiotics. Bloomington: Indiana Univ. Press, 1976.

Fischer-Lichte, Erika. Semiotik des Theaters. Tübingen: Gunter Narr Verlag, 1983.
Heath, Stephen. Vertige du déplacement. Paris: Fayard, 1974.
Herzfeld, Michael. "Lévi-Strauss in the Nation-State." Journal of American Folklore
98 (1985): 191-208.
Hymes, Dell. "The 'Wife' who 'Goes Out' like a man: Reinterpretation of a Clackamas
Chinook Myth." In Structural Analysis of Oral Tradition. Ed. Pierre Maranda
and Elli Köngäs-Maranda. Philadeplphia: Univ. of Pennsylvania Press, 1971, 49-
80.
nication. Eds. Dan Ben-Amos and K. Goldstein

74.
Köngäs-Maranda, Elli. "Five Interpretations of a Melanesian Myth." Journal of Amer-

ican Folklore 87 (1973): 3-13.
Leenhardt, Maurice. Do Kamo , Paris: Gallimard, 1947.
Lévi-Strauss, Claude. Anthropologie Structurale II. Paris: Pion, 1973.
Malinowski, Bronislaw. Coral Islands and Their Magic. Vol. II. Bloomington: Indiana
Univ. Press, 1965.
Meletinsky, Eleazar, et al. Soviet Structural Folkloristics. Ed. Pierre Maranda. The Ha-
gue: Mouton, 1974.
226 Erik Schwimmer
Schwimmer, Erik G. "Th

Oceania" colloquium, Ne
in a volume of proceed
thropologie et sociétés, 1986).
issue about Victor Turner).
and Jean Umiker-Sebeok. The Hague: Mouton,
Posner. Tübingen: Stauffenberg Verlag, 1985
Herdt. Berkeley: Univ. of California Press, 19
Ed. Harry R. Garvin. Lewisburg: Bucknell Univ.
Smith, Pierre. "Positions du mythe." Le temps

Thom, René. Modèles mathématiques de la mor
Vernant, Jean-Pierre. "Le mythe au réfléchi."
Wagner, Roy. Lethal Words. Cornell Univ. Pre
Young, Michael. Magicians of Manumanua. Ber

The Polysemic Analysis of Oral Literature (1985)

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

The Polysemic Analysis of Oral Literature (1985)

Caricato da

Copyright:

Formati disponibili

The Polysemic Analysis of Oral Literature

Author(s): Erik Schwimmer

The Polysemie Analysis

Recent advances in semiotics enable us to analyze certain cultural pat-

STYLE: Volume 19, No. 2, Summer 1985 213

depend a great deal on

After these importan

give rise- all serve th

not remain meaningfu

We need not repeat here our earlier demonstration (Schwimmer "Myth")

Once such a pragmat

This paper is a furthe

Finally, we must exp

The chief problem

woman" (24)? The wh

and behavior of entit

Barthes, Roland. S/Z. Paris: Seuil, 1970.

Eco, Umberto. A Theory of Semiotics. Bloomington: Indiana Univ. Press, 1976.

nication. Eds. Dan Ben-Amos and K. Goldstein

Köngäs-Maranda, Elli. "Five Interpretations of a Melanesian Myth." Journal of Amer-

Schwimmer, Erik G. "Th

thropologie et sociétés, 1986).

issue about Victor Turner).

and Jean Umiker-Sebeok. The Hague: Mouton,

Posner. Tübingen: Stauffenberg Verlag, 1985

Herdt. Berkeley: Univ. of California Press, 19

Ed. Harry R. Garvin. Lewisburg: Bucknell Univ.

Smith, Pierre. "Positions du mythe." Le temps

Potrebbero piacerti anche