Funcions Anfer Propp PDF

Functions after Propp: Words to Talk about How We Read Narrative
Author(s): Emma Kafalenos

Source: Poetics Today, Vol. 18, No. 4 (Winter, 1997), pp. 469-494
Published by: Duke University Press
Stable URL: http://www.jstor.org/stable/1773183
Accessed: 12-02-2017 22:17 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
http://about.jstor.org/terms
Duke University Press is collaborating with JSTOR to digitize, preserve and extend access to Poetics
Today
This content downloaded from 148.214.155.142 on Sun, 12 Feb 2017 22:17:20 UTC
All use subject to http://about.jstor.org/terms
Functions after Propp:
Words to Talk about How We Read Narrative
Emma Kafalenos
Comparative Literature, Washington University in St. Louis
Abstract A set of eleven functions, selected from the thirty-one functions Vladimir
Propp discerned, names interpretations of primary events in sjuzhet as well asfabula.
Assigning the act of interpretation to the reader, I use function analysis to trace
readers' shifting interpretations of events during the process of reading Poe's "The
Assignation," Browning's "My Last Duchess," and James's "The Story of a Master-
piece."
A recurrent criticism of structuralist narratology is the synchronic nature

of the object that its procedures seem designed to address. For Thomas G.
Pavel, for example, in an essay in the special issue of Poetics Today en-
titled 1Narratology Revisited I, "stressing the primacy of literary grammars
and conventions involved neglecting textual energy and movement ....
The interplay of structural principles has too often excluded the reader's
activity, and with it, a variety of temporal games that cross and flex the
text" (1990: 350-51). An approach developed from the work of Vladimir
Propp, however, permits analysis of readers' shifting interpretations of the
consequences of events in the continually expanding context of the other
events that the process of reading a narrative reveals.
In his Morphology of the Folktale, published in Russian in 1928, Propp re-
ports his discovery that events are functionally polyvalent. A function, in
Propp's words, is "an act of a character, defined from the point of view
of its significance for the course of the action" (1968 [1928]: 21), or, more
Poetics Today 18:4 (Winter 1997). Copyright ? 1997 by The Porter Institute for Poetics and
Semiotics.
470 Poetics Today 18:4
specifically, defined "according to its consequences" (ibid.: 67). To define

an event according to its significance, or its consequences, requires in-
terpreting its significance or consequences. Accepting Propp's definition,
but shifting his emphasis to reflect my own concerns, I define the term
function as an interpreted event, an event interpreted according to its con-
sequences, and I assign the act of interpretation, for the purposes of the
present essay, to the reader.' Because the consequences of an event depend
on the other events that are included in the sequence in which it occurs,
and on its position in that sequence, a given event can express one func-
tion in one narrative and a different function in another narrative. This
attribute of the event-that it is subject to interpretations that may shift
according to the context in which it is perceived-isfunctional polyvalence,
Lubomir Dolezel's term (1990: 144)2 for the phenomenon Propp discerned.
In addition to these two interrelated ideas, the principle of functional
polyvalence and the definition of the function, I take from Propp the
concept that an ordered set of functions can represent stages in the nar-
rative process, and I abstract from the thirty-one functions in his model
a smaller and more serviceable model that contains eleven functions. My
eleven-function model, I argue, represents the fundamental stages in the
narrative sequence.
An ordered set of functions is a semiotic system of interdependent con-
cepts and terms: a vocabulary to talk about how we read narratives. The
act of reading a narrative includes, in addition to identifying events as
they are revealed, a process of creating hypotheses about the causal rela-
tions among the revealed events and other events that may be revealed as
one continues to read. My eleven-function model provides a vocabulary
to analyze these provisionalfabulas and parts of fabulas readers construct
from the set of events discernible at any moment in the process of reading,
and to document the shifting interpretations of the consequences of events
that readers formulate as they move through a sjuzhet, page by page.
The term sjuzhet, as I use it, denotes a manifestation of fabula in any
medium and incorporates, in addition to the components of the medium
itself, all aspects of character beyond agency, the temporal factors of
sequence and duration and frequency in the form in which they are mani-
1. In an essay written after this one, I analyze protagonists' interpretations--and misinter-

pretations-of events as functions, as well as the effects of protagonists' interpretations on
readers' interpretations, in two narratives in which readers are permitted to perceive events
only through a protagonist's focalization (Kafalenos 1995).
2. Conversely, Dolezel posits the principle offunctional equivalence: one function can be ex-
pressed by different events (199o: 144).
Kafalenos * Functions after Propp 471
fested, and focalization.3 Fabula, conceived as unexpressed in any medium,

is a chronologically ordered abstraction of the events and agencies a
sjuzhet reveals to a perceiver. Although the two trajectories, as I conceive
them, are as interdependent as the signifier and the signified, the distinc-
tion between them is essential to my enterprise.
Jonathan Culler has argued that the relationship between fabula and
sjuzhet is hierarchical and thus subject to deconstruction through a dem-
onstration that the hierarchy can be reversed (1981: 183). Postulating that
narratology gives priority to fabula, he contends that the priority implies
"a hierarchy [of fabula over sjuzhet] which the functioning of narratives
often subverts by presenting events not as givens but as the products of [the
sjuzhet's] discursive forces or requirements" (ibid.: 172). As an example of
what he sees as an event produced by "discursive forces," Culler points to
the gap that exists in the fabula of Sophocles' play because Oedipus does
not ask the surviving witness whether Laius was murdered by one or many.
Culler concludes that although readers are convinced of Oedipus's guilt,
"our conviction does not come from the revelation of the deed. Instead of
the revelation of a prior deed determining meaning, we could say that it is
meaning, the convergence of meaning in the narrative discourse [sjuzhet],
that leads us to posit this deed as its appropriate manifestation" (ibid.: 174).
First, Culler's postulate that narratology gives priority to fabula is ques-
tionable. Ontologically, sjuzhet has a material form that fabula lacks.
Temporally, sjuzhet is prior for readers to the fabula it gradually reveals.4
Second, the "convergence of meaning" that implies that Oedipus mur-
dered Laius does not make the deed occur; it makes it significant. An
effect does not cause a prior event. An effect indicates the significance of
a prior event by placing it in a sequence in which the prior event can be
3. Following the distinction initially introduced by Gerard Genette (1980 [1972], primarily
chaps. 4 and 5) between the narrative voice that speaks and thefocalization that perceives and
conceives, I see focalization as an element of sjuzhet in any medium, and voice as a com-
ponent of those media that incorporate language. Although I choose the terms sjuzhet and
fabula to avoid the confusion created in French by the shifting signification of recit (which is
comparable to sjuzhet when paired with histoire, and to fabula when paired with discours),
the definitions I am using, which incorporate elements introduced by a number of post-
Formalist theorists, are strongly influenced by Shlomith Rimmon-Kenan's precise analyses
[esp. 1976: 35-36 and 36 n. 2] of the increasingly explicit definitions of French narratolo-
gists, including Claude Bremond's distinction [1964: 4] that what I termfabula is pretextual
and medium-free.
4. Seymour Chatman, writing in response to Culler's argument, emphasizes that "from the
theoretical point of view, narratology is resolutely synchronic. It does not assume that either
telling or told 'precede' each other: they are coexistent, cotemporal parts of the model"
(1988: 14).
interpreted according to its consequences. The "discursive forces" of the

sjuzhet that Culler discerns, I argue, do not alter the principles of cau-
sality; rather, they guide perceivers to establish the function of the event
in question, to interpret it according to its consequences.
Given that events are functionally polyvalent, an event's meaning-
at the fundamental level of interpretation in narrative-is defined by its
function. The function of an event is dependent on its position in a se-
quence. In Roland Barthes's terms, the writer, like the bricoleur, sees the
meaning (le sens) of events by relating them: "by trying fragments of events
together . . . by tirelessly transforming these events into functions" (1972
[1962]: 182).5 Similarly, the act of reading requires readers to interpret
events as functions. For readers as for writers, narrative is a procedure for
interpreting events by perceiving them in a sequence. Functions name the
positions in the narrative sequence that give meaning to the events that
represent them.
Tzvetan Todorov outlines the large-scale pattern of the narrative se-
quence in defining the minimal complete plot as the movement from one
equilibrium, through a period of imbalance, to an equilibrium that is simi-
lar but never identical to the first (Todorov 1968: 96).6 I adopt Todorov's
definition as a definition of the narrative sequence, and I assume that one or
more or a part of a narrative sequence is embodied in every narrative, that
is, that the trajectory of every narrative traces one or more or a part of
a move from equilibrium to disruption to equilibrium. But as valuable as
Todorov's perception of an underlying periodic alternation between bal-
ance and imbalance has proven, it provides no means to identify stages
within the process.
An ordered set of functions represents stages in the narrative sequence.
Although Propp makes no claim that the results of his analysis of a spe-
cific corpus of tales will reveal information about narratives belonging to
other genres,7 the ordered set of functions he discerned, and chose letters
5. See Hayden White's analysis of the functional polyvalence of historical events, e.g., the
"death of the king" (1973: 7), and of the effect of the position of an event in determining its
function in historical narrative.
6. Todorov specifies that an individual narrative sometimes presents no more than a seg-
ment of the trajectory: the movement from an equilibrium to a disequilibrium, or the
reverse (1968: 102).
7. Propp's purpose is generally understood to be to demonstrate the generic identity of a
category of narrative he terms thefairy tale, as represented by the corpus of tales he studied.
But Propp does not claim that the one hundred Russian tales he analyzed are identical in
form. He recognizes, as I do, that not every function is embodied in every sequence, and
that narratives differ according to which functions are represented in each sequence, and
according to the number of sequences they contain and the combinatory pattern in which
to name, includes functions that recur in narratives of many genres and

periods. These are the functions I retain in my eleven-function model,
which represents the fundamental stages in the narrative sequence. Named
and defined, the functions in my model provide a vocabulary to docu-
ment readers' interpretations of the primary events in any narrative they
read.
In addition, although Propp does not distinguish between sjuzhet and

fabula, nor does he need to in the context of his study of a set of tales in
which sjuzhet and fabula are isomorphic, the shape of the narrative se-
quence, from equilibrium to disruption to equilibrium, in its entirety or in
part, can be perceived in sjuzhets as well as fabulas, and the same set of
functions can name the stages of the narrative sequence in sjuzhets and in
fabulas. The possibility of using one set of terms to analyze both trajecto-
ries is the direct result of the principle that underlies the definition of the
function. Because a function expresses an interpretation of an event in the
sequence in which it is perceived, a given event can be interpreted as a
function in any sequence in which the event occurs. This principle permits
comparisons not only between fabulas, but also between sjuzhets, between
a sjuzhet and the fabula it reveals, and between a segment of a sjuzhet and
the segment of fabula it reveals.
To avoid unnecessary confusion, I have kept Propp's names for the
eleven functions I retain.8 The definitions I provide for each function are
developed from Propp's, but are more abstract than his, and are designed
to reveal the general situations that underlie the specific conditions of the
stories he studied. The eleven functions represent the fundamental stages
multiple sequences are arranged. The question is whether Propp's purpose -"a description
of the tale according to its component parts and the relationship of these components to
each other and to the whole" (1968 [1928]: 19)-is to define a genre by revealing a pattern
to which a set of narratives conforms, or to reveal the differences that characterize an indi-
vidual narrative. The latter is my project.
According to Claude Bremond and Jean Verrier, Propp's project was the former and he
failed to accomplish it. After analyzing the eight stories for which Propp provides a formal
representation, they conclude that four of the eight "cannot be reduced to the Proppian se-
quence without severe mutilations which destroy essential aspects of the plot" (1984 [1982]:
192-93), and report what they perceive as "Propp's blindness or indifference before the re-
sistance his corpus offered to the analytical framework he attempted to impose on it" (ibid.:
193). Because I have found Propp's work so fertile a field, I urge that we gauge his accom-
plishment without limiting the scope of our analysis to whether he achieves a purpose that
we must remember he formulated in the absence of a precedent, which, for later theorists,
his pioneering study provides.
8. With one exception. I substitute C' (C prime) for Propp's arrow signifying departure,
to reinforce the close relationship between the decision to act (function C) and the initial
action (function C'), and to avoid a recurring symbol to which no ASCII code is assigned.
of the narrative sequence, from the disruption of an equilibrium to the

establishment of a new equilibrium.9
Initial equilibrium (not a function)

A (or a) disruptive event (or reevaluation of a situation)
B request that someone alleviate A (or a)
C decision by C-actant10 to attempt to alleviate A (or a)
C' C-actant's initial act to alleviate A (or a)
D C-actant is tested
E C-actant responds to test

F C-actant acquires empowerment
G C-actant arrives at the place, or time, for H
H C-actant's primary action to alleviate A (or a)
I (or Ineg) success (or failure) of H
K equilibrium
Like Propp's, my model is conceived as an ordered set; the order in which
functions occur is inalterable. Like Propp's, my model contains more fun
tions than a narrative sequence requires to be complete; the functions tha
are indented can be omitted without fragmenting the sequence.
The six functions in the left margin delineate the minimal complete se
quence: the move from an initial equilibrium to A (the disruption of the
equilibrium) and then, through the four primary functions of the C-actan
to K (a new equilibrium).
The two types of A functions mark the distinction between the tw
potential sources of disruptive change: (1) an event that alters the exter-
nal world (function A), and (2) a psychological reevaluation that alters the
perception of an otherwise unchanged situation (function a).
The primary functions of the C-actant, the agent for change, measure
9. Although I have derived my eleven-function model by analyzing (with the assistan

of students in several classes) the incidence of Propp's thirty-one functions in a num
of stories and novels, I draw attention to the similarity between the functions my mod
contains and the functions in two of the three five-function, logically ordered sequence
A.-J. Greimas uncovered in his analysis of Propp's morphology (Greimas 1983 [1966]: 228
and the functions in the three core sequences (omitting the preparatory and concluding s
quences) in Paul Larivaille's extension of Greimas's model (Larivaille 1974: 376, 383). M
model is even closer in spirit to the concise five-category plot-grammar devised by Pavel
depict, in convincing representations of the causality of fabula, causally related sequenc
as interconnected trees (Pavel 1985), although I adopt a sequential model to trace the d
chronic process of reading.
o1. By definition, function C is the act of the C-actant: the actant who decides to amelior
A (or a). For the duration of any one sequence, the role of the C-actant is played by a sin
character (or a single group of characters acting together). In narratives containing mor
than one sequence, the role of the C-actant in one sequence may be played by a differen
character from the one playing the C-actant role in a second sequence.
stages in the process of change from A to K: C (the decision to amelio-

rate A), C' (which marks the moment the C-actant begins to act), H (the
primary action), and I (the success or failure of the action).
In narratives in which indented functions are included, functions D, E,
and F (Propp's donor section) sketch the process whereby the C-actant
achieves empowerment to accomplish function H; function G marks a
shift in time or location to the scene of function H. Function B occurs in
narratives in which a potential C-actant who is unaware that a motivating
function A (or a) has occurred, or who is not immediately affected by its
disruption, is asked to act to alleviate the situation.
Since events are functionally polyvalent, and their function depends
on their consequences, readers often change their interpretation of the
function of an event as they read, page by page, and acquire further in-
formation. For example, when we read the beginning of a narrative, we
recognize that an absence or lack (of a parent, spouse, position, money,
etc.) to which a character or a narrator draws attention may indicate that a
recent reevaluation of the seriousness of the lack will motivate a sequence.
In my terms, an absence or lack that readers interpret as a motivating
situation is labeled function a. But as we read on, we may discover that the
absence or lack, rather than serving as the primary motivation, may do no
more than enable the occurrence of a disruptive event that will motivate a
sequence. In this case, we label the disruptive event function A and shift
our interpretation of the prior situation from primary motivation (func-
tion a) to a merely preparatory state of affairs.
In narratives in which sjuzhet sequence varies from the chronology of
fabula, the interpretation of the function of individual events shifts ac-
cording to the sequence in which the events are arranged. Thus the event
that seems the primary motivation and is interpreted as function A (or a)
in sjuzhet may be entirely different from the one that is interpreted as
function A (or a) in fabula. This shift in interpretation provides theoretical
support for a recognized phenomenon: the same fabula can be shared by
narratives that create very different effects. To explore this phenomenon,
and to demonstrate the plasticity of function analysis as a tool to docu-
ment and differentiate interpretations, I have chosen for analysis, first, a
story in which the functions of events in fabula and sjuzhet are dramati-
cally different, Edgar Allan Poe's "The Assignation," and, then, a pair of
narratives that share segments of their fabulas, Robert Browning's "My
Last Duchess" and Henry James's "The Story of a Masterpiece."
As we read Poe's "The Assignation," we follow the focalization of the

narrator as he perceives events, interprets events as functions, and reinter-
prets them as the sequence of events continues. When we reach the end of
the story we construct its fabula and again reinterpret events as functions,
this time in the sequence of fabula. Menakhem Perry proposes that when
sjuzhet sequences depart from "the 'natural' sequence of an 'external'
occurrence," they generally conform to one of two models: "the 'natural'
sequence of a character's consciousness, the sequence within a block of in-
formation transmitted from one character to another" (1979: 39-40). Poe's
story illustrates both models. Structured according to the sequence of the
narrator's perceptions, the sjuzhet incorporates-when the narrator reads
it-a text written by the protagonist that offers information essential to
the construction of fabula.
In the second paragraph of the story, the narration moves from a gen-
eralized invocation to a fully depicted scene that takes place at a specific
time -the "third or fourth" meeting between the first-person narrator and
an unnamed man who has "fallen in the flames of ... youth" (Poe 1977
[1834]: 193)-and provides indications of an initial situation: Venice, the
Grand Canal, midnight, the narrator in a gondola. The equilibrium of
the peaceful evening is disrupted by the scream of a woman's voice, which
the narrator (together with the reader) immediately interprets as func-
tion A. The narrator leaps to his feet-we assume in order to rescue the
woman who screamed. The narrator's and many readers' first interpreta-
tion of the initial events can be documented as follows:
Initial equilibrium: the calm of the canal at midnight

A woman screams (disruptive event)
[C] narrator decides to alleviate A (narrator is C-actant)
C' narrator leaps to his feet (first act to alleviate A)
This is the reading my function analysis documents: The woman's scream

is a disruptive event that will motivate a sequence (A); the narrator decides
to undertake the role of C-actant and alleviate the disruption (C; bracketed
because the event is revealed only by its effect: the narrator's movement);
the narrator leaps to his feet as his initial act to alleviate the disruption (C').
I pause immediately to specify that the interpretation my function
analysis proposes is not intended to be prescriptive, nor to represent an
only possible reading of a narrative or a segment of a narrative. Although
Poe's story offers an opening set of events for which I can conceive only
one reading (which, as we shall see, we are immediately forced to reject),
later in this essay I offer an example of two readings of a segment of text
that can be held simultaneously, at least for a period of time during the
process of reading. I am arguing, however, not for the validity of a specific
reading, but for the usefulness of functions as a vocabulary to talk about
how readers respond to the fundamental question narratives pose: how to
Kafalenos ? Functions after Propp 477
interpret a specific event in the sequence in which it occurs. My purpose

is to illustrate the effectiveness of function analysis to represent a read-
ing. If my argument is convincing, readers who disagree with my readings
will formulate their own readings in terms of functions-and discover the
value of function analysis to talk about differences at the very basic level
of interpretation this essay addresses.
For readers who have interpreted the narrator's leap to his feet as the
first sign that he is to rescue the woman (function C'), the next event in
the sjuzhet blocks our interpretive trajectory. When (or because) the nar-
rator leaps to his feet, the gondolier drops his oar into the canal and in
the darkness cannot recover it. Instead of gaining power (function F) to
alleviate A, the narrator loses even the ability to steer the gondola (Fneg).
The narrator and the reader recognize simultaneously that the narrator's
plan to rescue the woman (to perform the role of the C-actant and allevi-
ate the function-A situation indicated by her scream) cannot succeed, and
that their interpretation of events as functions has created the beginning
of a sequence that has reached an impasse and cannot continue to a suc-
cessful conclusion: A [C] C' Fneg. Readers who perceive the futility of the
narrator's initial act may expect the entrance of an additional character,
one who is better qualified for the role of the C-actant.
As the gondola drifts in the canal, the narrator discovers the reason for
the woman's scream; she is the Marchesa Aphrodite, and her child has
fallen into the canal. As a result, the narrator and the reader shift their
interpretation of the initial events. The opening scream is not function A
but function B. The child's fall is the disruptive event that will motivate
a sequence; I bracket it to indicate that it precedes the first event of the
sjuzhet (the scream):
Initial equilibrium
[A] child falls into canal
B Aphrodite screams for help
After a number of swimmers search for the child in vain, and while
Aphrodite's husband, the Marchese Mentoni, strums his guitar, a man the
narrator calls "the stranger" dives into the canal and emerges, next to
Aphrodite, carrying the still-breathing child. The narrator and the reader
immediately conclude that the stranger is the C-actant, who has just per-
formed the four primary C-actant functions. We assume that, since the
child has been saved, equilibrium is restored. This is the interpretation
that the following sequence of functions documents:
Initial equilibrium

[C] stranger decides to save the child (bracketed because revealed by
C')
C' stranger dives into canal
H stranger struggles to locate child
I stranger saves child
K equilibrium; A is alleviated
For both the reader and the narrator, this interpretation of events as func-
tions seems correct. The narrator's conviction that it is valid is revealed,
in fact, by the surprise he expresses as he narrates the next several events:
the child is not taken by its mother but by someone else who carries it off;
Aphrodite blushes; her words to the stranger are unexpected: "Thou hast
conquered-one hour after sunrise-we shall meet-so let it be!" (Poe
1977 [1834]: 197).
Although the reader, like the narrator, may have expected a scene in
which function K is reinforced-Aphrodite, for example, could have ca-
ressed the child with delight and profusely thanked the stranger-neither
the reader nor the narrator anticipates further events that would alter their
present interpretation as functions of the events they have perceived. When
the narrator arrives at the stranger's apartment early the next morning,
at the stranger's request, the narrator seems to expect-and thus we ex-
pect-to receive information that will fill out and elucidate the pattern of
functions already revealed. In the room to which the narrator is shown in
the stranger's Palazzo, along with the narrator we examine the art objects,
find an underlined passage in a book, and read a poem written in English
in the stranger's hand. When the stranger proposes a toast, drinks, and
collapses, and when a servant enters to announce that Aphrodite has died
of poison, we discover with the narrator that the stranger, too, is dead.
At that moment, for the narrator, in the words with which the story con-
cludes, "a consciousness of the entire and terrible truth flashed suddenly
over my soul." For the reader, equivalent illumination can occur only after
extensive analysis.
When the effect of Poe's ending dissipates sufficiently to permit reflec-
tion, we understand the double signification of the word assignation in the
story's title: the agreement to meet (Aphrodite's cryptic "one hour after
sunrise" [ibid.: 197]), and the meeting itself (to which the participants of
the double suicide expect it to lead). Knowing that we have now been
given the concluding events in a finite sequence, we do what we always do
when we finish reading a narrative; we begin to reevaluate our interpreta-
tions of events as functions. A probable first reading of the sjuzhet in its
entirety can be documented by this sequence of functions:
Initial equilibrium
[C] stranger decides to respond to Aphrodite's need (bracketed
because revealed by C')
C' stranger saves child
F stranger acquires empowerment (when Aphrodite agrees to the
assignation)
G stranger arrives at the place where H will occur
H stranger and Aphrodite poison themselves
I success of H: both die
[K] equilibrium: stranger and Aphrodite are united in death
Because the child's fall is positioned at the beginning of the n

and is the only clearly disruptive event until the final moment
story, we retain that event as function A. Our reading of the stran
the C-actant remains unchanged as well. But since we now know th
stranger's primary undertaking is a joint suicide to achieve eternal
with Aphrodite, we interpret the suicide as function H. Retrospect
we understand that Aphrodite's agreement to the assignation is
powerment (function F) that permits the double suicide to occur, an
retain the stranger's retrieval of the child from the water as his in
(function C') because it is the first act we see him undertake. Funct
and K are bracketed because the events to which they refer do not
within the temporal limits demarcated by the sjuzhet.
We respond to the concluding events of the sjuzhet with surp
part because it is only at the end of the story that we are forced t
that the narrator's focalization is located at the earlier time in which he
watched the events occur, while his voice is located at the later time of his
retrospective narration." The temporal gap between focalization and voice
permits the narrator to express amazement, which enhances our own.
11. Both focalization and voice are retrospective in the opening paragraph, but split in the
second paragraph: the source of the words remains at the later time of retrospection; the
focalization shifts (through a very heightened form of memory) from "confused recollection"
(Poe 1977 [1834]: 193) to the perceptions/conceptions of the time of the events. Nonetheless,
the initial indications of retrospection are so clearly established and the split is so subtle that
readers cannot be aware of it on first reading until the final paragraph.
I am indebted to James Phelan, who accepted an earlier version of this analysis of "The
Assignation" for a session sponsored by the Society for the Study of Narrative Literature at
the 1991 MLA Convention, for leading me to clarify this strategy by pointing out the "para-
dox in the retrospective nature of the telling. The knowledge that the narrator achieves at
the end of the time of the action, which is of course before the time of the narration, doesn't
inform that narration" (Phelan 1991).
But the fact that we have not anticipated the double suicide is only one
element in the effect Poe's story creates. The sequence of functions that
documents our interpretation of the sjuzhet reveals a logical gap. The sui-
cide does not save the child; function H does not alleviate function A. The
shock with which we respond to the final event is created by this gap. We
cannot understand how the child's fall into the canal can have caused the
double suicide. The placement of the child's fall at the beginning of the
sjuzhet is what leads us to interpret the event as the function A that will
motivate the rest of the sequence. The effect of Poe's story is in large part
created by sjuzhet sequence.
Once we recognize the inadequacy of the child's fall to motivate the
double suicide, we rethink (and reread) the sjuzhet to find the fabula it re-
veals. Within the sjuzhet, in which the order of events is determined by
the sequence of the narrator's experience, we find an embedded sjuzhet
that conforms to what Perry describes as "the sequence within a block
of information transmitted from one character to another" (1979: 39-
40): the poem the narrator finds in the stranger's apartment, handwritten
in English, with the word "London" inscribed on the page and crossed
out (sous rature). A focalization framed by a focalization, the poem pro-
vides readers with both a momentary glimpse of the situation through the
stranger's focalization and an opportunity to discover elements of fabula
otherwise unrevealed. The first four stanzas of the poem, which Poe pub-
lished separately under the title "To One in Paradise" (Stern 1977 [1945]:
204), invoke a beloved and describe a "dream too bright to last." The final
fifth stanza, which appears only in the story, reads:
Alas! for that accursed time
They bore thee o'er the billow,

From Love to titled age and crime,
And an unholy pillow-
From me, and from our misty clime,
Where weeps the silver willow!
(Poe 1977 [1834]: 204)
Reading the poem as the primary source of information about events that
precede the child's fall, we construct a fabula, which we interpret accord-
ing to the following sequence of functions:
Initial equilibrium: the lovers together in London

A Aphrodite carried over the waves to Venice to marry the Marchese
Mentoni
C stranger decides to attempt reunion with Aphrodite

C' stranger follows Aphrodite to Venice
Kafalenos ? Functions after Propp 481
D stranger is tested: Aphrodite screams for help

E stranger passes test: saves child
F stranger receives empowerment: Aphrodite agrees to assignation
G stranger arrives at place where H will occur
H double suicide
I success of suicide: both die
[K] equilibrium: the potential reunion in death
The last five functions of this fabula analysis (from F to K) are filled b
the same events as the last five functions of the sjuzhet analysis (from F
to K). In both analyses, the stranger is the C-actant. But whereas, as w
read the sjuzhet, the event we interpret as the motivating function A is th
child's fall, in the chronological sequence of fabula the event we interpret
as function A is the forced separation of Aphrodite and the stranger. As
a result, in fabula, the stranger's decision to alleviate the separation,
effect reunion with Aphrodite, is the event that fills the C function. In th
context, we shift our interpretation of Aphrodite's scream, which we ini
tially read as A (the disruptive event), and then as B (a call for help), to D
the first function of the donor sequence. Her scream functions as a test,
to which the stranger responds successfully (E), thereby gaining the pow
(F) to accomplish the primary conflict (H, the double suicide). Functio
provide words to talk about the shifting interpretations readers give
events as they read.
In addition, a function analysis of an individual narrative permits us to
determine the number of narrative sequences in each of its trajectorie
its fabula and its sjuzhet, and to describe the sequence or sequences o
part of a sequence in each of the two trajectories according to which func
tions it includes. Looking again at the fabula analysis of Poe's story, w
see that the fabula contains one complete narrative sequence; its traje
tory includes every function, with the exception of function B, as it move
through an entire narrative sequence from an initial equilibrium to a dis-
ruptive function A to the new equilibrium of function K. Looking still at
the fabula analysis, however, we can see that the segment of fabula that
the trajectory of the sjuzhet traces--from Aphrodite's scream in the open
ing scene to the double suicide of the conclusion-is less than a comple
narrative sequence; the sjuzhet traces the part of the fabula that mov
through the six functions from D to I.
The exposition in a narrative, as Meir Sternberg has shown, extend
from the beginning of the fabula (1978: 13) to "that point in time which
marks the beginning of the fictive present in the sujet" (ibid.: 21). For th
scene in the sjuzhet that marks the conclusion of the exposition, Sternber
perceives, "The author's finding it to be the first time-section that is 'of

consequence enough' to deserve full scenic treatment turns it, implicitly
but clearly, into a conspicuous signpost, signifying that this is precisely the
point in time that the author has decided, for whatever reason, to make
the reader regard as the beginning of the action proper" (ibid.: 20).
What Poe has done in "The Assignation," by selecting the scene that
begins with Aphrodite's scream as the first to receive full scenic treatment,
is to guide readers to perceive everything that is chronologically prior to
Aphrodite's scream as exposition--even though the chronologically prior
events include the events that in the fabula constitute the primary moti-
vating event (function A) and the initial actions of the C-actant (functions
C and C'). Poe's placement of the scene in which the child nearly drowns
as the first fully depicted scene in the sjuzhet is, of course, the reason that,
as we read, we interpret the child's falling into the canal as a function-A
event that motivates everything that follows.
For a writer to choose to deviate from "the straight chronological order
of presentation," is, as Sternberg observes, "clearly an indication of artistic
purpose" (ibid.: 33). The differences between sjuzhets and fabulas, in in-
stances in which a scene that occurs late in the chronology of fabula is the
first to receive full scenic treatment in the sjuzhet, have been described,
since Aristotle's Poetics, as a proportion between temporal durations: the
less than twenty-four hours that the trajectory of the sjuzhet of, for ex-
ample, Oedipus Rex traces, to the years required for the events of the fabula
to unroll. Sternberg's recognition that the exposition of a narrative can be
delimited only with reference to both fabula and sjuzhet enables his anal-
yses of the aesthetic effects of the ways in which elements of fabula are
distributed in a sjuzhet, in his Expositional Modes and Temporal Ordering in
Fiction (1978), but also permits a comparison of the two trajectories that re-
veals the specific events in a fabula that a sjuzhet relegates to exposition. A
function analysis permits us to perceive the causal effects -the function -
of the events in fabula that a sjuzhet consigns to exposition and to specify
the instances in which the placement of an event in a sjuzhet leads us to
interpret it as one function in the sjuzhet and a different function in the
fabula. These instances of functional polyvalence are among the artistic
effects that a compressed sjuzhet can produce.
Moreover, I argue, for readers who interpret a given event as one func-
tion in sjuzhet, and then as a different function in fabula, the event tends to
remain functionally polyvalent. Both Sternberg and Perry (e.g., in English,
Sternberg 1978: 93ff.; Perry 1979: 53ff.) have summarized and analyzed
studies reported by psychologists who designed experiments to determine
the effect on interpretation of the sequence in which information is per-
ceived. These studies appear to document the primacy effect: our tendency
to accept as valid the information we are initially given, even when that
information is contradicted later in the same message. When we analyze
the consequences of a given event in a narrative, first as we read a sjuzhet
and then as we construct its fabula, our interpretation of the function of
the event that is formed by its position in the sjuzhet is not immediately
forgotten if later we reinterpret the function of the event in fabula.
In reading "The Assignation," even after we have determined that the
child's fall into the water is not the function-A event that motivates all that
follows, the scene in which the child is in danger and then rescued retains
a heightened quality in our memories that reflects the causal implications
of our initial interpretation. In considering the story retrospectively, we re-
main able to envision this scene in detail--in a way that we cannot envision
the chronologically prior events, including the enforced separation in Lon-
don of Aphrodite and the stranger-and the initial scene remains tinged
in our minds with the strong emotions with which we initially perceived it.
The effect of the functional polyvalence of an event, in instances in
which we interpret a given event as one function in sjuzhet and a different
function in fabula, is perhaps strongest immediately after we have con-
cluded our reading and completed the process of establishing a fabula and
gradually fades as we forget, probably first, the effects of the sjuzhet, and
then even the chronological sequence of fabula. But the artistic effect of
the sjuzhet endures, I argue, until the sjuzhet is forgotten, and it can be
reactivated by rereading. The sjuzhet shapes readers' interpretations. Two
narratives with similar fabulas provide a situation to test the power of the
sjuzhet to shape interpretations of events.
The fabulas of Robert Browning's "My Last Duchess" and Henry James's
"The Story of a Masterpiece" share a motivating event and an identical
set of actants: an older man, the woman he marries, and the male art-
ist who paints the woman's portrait. In one of the two sequences in each
fabula, the portrait the artist paints reveals and confirms, for the older
man (Browning's Duke, James's John Lennox), the insufficiencies of the
woman it depicts (the Duke's "last Duchess," Lennox's fiancee Marian).
Explicitly in the story and implicitly in the poem, the fixed image of
the portraits intensifies, for the Duke and for Lennox, their earlier and
perhaps only half-conscious misgivings about the living model, thereby
provoking a reevaluation of wife and fiancee that disturbs both men's pre-
vious complacency. Because the action is motivated by the Duke's and
Lennox's reevaluation of the women, I interpret the motivating event in
both sequences as the second type of A function: the lowercase function
a that represents a psychological reevaluation of an otherwise unchanged

situation. It is not the women who change, but the Duke's and Lennox's
perception of the women, and of their relationship with wife and fiancee.
In both sequences, the reevaluation of the woman (function a) leads the
Duke and Lennox to assume the C-actant role, first by determining (func-
tion C) and then by initiating (function C') a course of action designed to
ease the discomfort the relationship now causes them.
The fabulas of the two narratives are not identical, and, in fact, in sev-
eral aspects, James's fabula reverses Browning's. In the conclusions of the
sequences that are similarly motivated, whereas Browning's Duke keeps
the portrait but gives commands that condemn his wife to death, James's
Lennox destroys the portrait but marries Marian. In the other sequence in
the two fabulas, whereas in Browning's poem a marriage contract is being
negotiated, in James's story an engagement is broken off. Moreover, in
the chronology of fabula, the sequence that is similarly motivated in both
narratives precedes in Browning's poem-but follows in James's story-
the other sequence. Nonetheless, the differences in the effects of the two
narratives are in part the result of the strategies of the two sjuzhets.
Browning's poem opens, as I read it, by suggesting a scene of stability
and equilibrium; the Duke is ensconced at his family seat, entertaining his
guest, the narratee,'2 by showing him the Duchess's portrait. The locus of
the event places the Duke against the backdrop of his inherited wealth,
and in a social situation in which his confidence bespeaks his entrenched
power in the represented world. Browning creates this equilibrium through
the selection of the time and place - the segment of fabula -that the open-
ing lines of the sjuzhet depict.
The sjuzhet begins during the meeting between the Duke and the nar-
ratee that is the final event in the fabula. Moreover, the entire sjuzhet is
set within the final event in the fabula, which permits the Duke to narrate
a retrospective account of events he has witnessed. The juncture of voice
and focalization, which is assigned to the Duke and-unlike Poe's narra-
tor- at the time at which he speaks, lends credence to the Duke's words.'3
12. The one whom the narrator addresses, overtly (as here) or otherwise, conceived as in
the text and at the ontological level of the narrator. The term was introduced by Gerald
Prince (1980 [1973]).
13. In a dramatic monologue, the restriction of voice and focalization to one character re-
quires readers to "adopt [the speaker's] viewpoint as our entry into the poem," as Robert
Langbaum notes as early as 1957 (1963 [1957]: 78). Langbaum's recognition of a causal re-
lationship between, on the one hand, one of the defining characteristics of the genre, and,
on the other, the experience of readers as they read the opening lines, is not invalidated
by arguments against his further claim that the genre incites sympathy for the speaker to
the degree that readers "suspend moral judgement" [ibid.: 83], which, as Wayne C. Booth
But the Duke's focalization, in two passages early in the sjuzhet, con-
tains other focalizations: the unvoiced response of the narratee (and pre-
vious viewers) to the Duchess's portrait, and the portrait itself, which is
the visual representation of the painter's perception of the Duchess. Be-
cause a contained focalization can be colored by the focalization through
which it is perceived, a contained focalization cannot confirm-but may
subvert-the focalization that contains it. In Browning's poem, the first
of the contained focalizations may lead readers to wonder what possible
expression anyone could read on the faces of viewers of the painting as
seeming, in the Duke's words, "as they would ask me, if they durst, / How
such a glance came there" (Browning 1988 [1842]: 11. 11-12). Readers who
question, in this passage, the validity of the Duke's interpretation of the
represented world are prepared, when the Duke's account turns from re-
sponses to the painting to the painting itself, to doubt the accuracy of the
Duke's ascription of the depicted lady's painted charms to her pleasure
in the painter's compliments, and to distrust his judgment of his wife as
having had "a heart ... too soon made glad" (ibid.: 1. 22).
At this point in the sjuzhet, as I read the poem, the equilibrium of
the opening moments has been undermined sufficiently that most readers,
during the following section in which the Duke continues to speak dis-
approvingly of the Duchess's behavior (ibid.: 11. 23-34), will decide that
the motivating situation (function a) for the Duke is his perception that his
wife smiles too readily at other men-a disruption that is intensified for
readers who suspect that the Duke's interpretation of his wife's behavior
may not be accurate. For readers, the knowledge of already having read
more than half the poem (34 of 56 lines) reinforces an interpretation that
the depicted situation, rather than an event yet to be revealed, is function a.
It is just at this moment that the sjuzhet turns to the Duke's rumina-
tions on whether it is appropriate to "stoop ... to make your will / Quite
clear to such an one" (ibid.: 11. 34, 36-37), a passage that represents in the
present the Duke's mental process of deciding (function C) whether and
how to respond to the disruption of function a. As shocking as the resul-
tant events remain, even after repeated readings, the sequence of functions
they fulfill is absolutely logical. Eight words, "I gave commands; / Then
all smiles stopped together" (ibid.: 11. 45-46), reveal the events that fill
three functions: function C' (the commands), function H (the murder, sup-
pressed but understood because of its result), and function I (the successful
alleviation of the function-a situation: She no longer smiles at other men).
perceives, "seriously underplays the extent to which moral judgment remains even after
psychological vividness has done its work" [1961: 250 n. 6].
The Duchess's no longer being alive, however, is doubly interpreted by

the Duke: as the successful resolution of the initial disruption (function I in
the first sequence), and as motivation to remarry (function a in the second
sequence, which the sjuzhet reveals only through the marriage negotia-
tions). The two chained sequences of Browning's sjuzhet can be depicted
as follows:
Initial equilibrium (not a function)

a Duchess's (painted and real) smiles disturb Duke
C Duke decides not to be disturbed
C' Duke gives commands
[H] Duke's henchmen murder Duchess (revealed by function I)
I Duchess is dead: she no longer smiles
[a] Duchess is dead: Duke lacks a wife (interpretation revealed by
function H)
[C] Duke decides to marry (revealed by function H)
[C'] Duke approaches Count (revealed by function H)
H Duke negotiates marriage contract with narratee
If we remove the brackets that indicate the suppression of an even

in the sjuzhet, these two chained sequences also represent the poem
fabula. This congruence of functions in sjuzhet and fabula occurs because
the causal events are revealed in the sjuzhet in chronological sequenc
Browning accomplishes this-while placing the Duke's retrospective ac
count within the concluding event of fabula-by moving directly fro
the final moments of the fabula to the first event of the fabula, throug
the metonymical shift from the painted Duchess (11. 1-12) to the process
of painting her ("How such a glance came there" [1. 13]). As a result
this strategy, only the situation of the initial equilibrium differs in sjuzh
and fabula. In the fabula it represents a state of affairs prior to the Duk
concern that his wife smiles too freely; in the sjuzhet it is a product
the Duke's confidence, during his meeting with the narratee, who is the
Count's emissary, that his negotiations to marry the Count's daughter ar
about to reach a successful conclusion.
But the initial equilibrium as Browning constructs it is highly effective
because the Duke's supreme confidence that the opening of the sjuzhet
portrays increases, in retrospect, the abhorrence with which readers regard
the Duke at the poem's conclusion. In contrast, imagine the sympathy for
the Duke one might feel if the poem began with a scene in which the
Duchess smiled at the Duke, and he was pleased. But Browning portrays
the Duke in the opening scene as coldly confident, rather than loving, and
further intensifies readers' dislike for the Duke by delaying the revelation
that the Duke is responsible for the Duchess's death, adding the element of
surprise to enhance our disapproval.'4 Browning postpones this informa-
tion by extending the events of the first two functions (functions a and C
of the first sequence) for forty-four lines (more than 75 percent of the fifty-
six-line poem). The Duke's commands occur on line 45, and the events
of the seven functions that complete the first sequence and represent the
second are compressed within the final twelve lines.
Although the second sequence is technically left open, stopping at func-
tion H, my response to the poem's conclusion is that the Duke's power and
confidence in the marriage negotiations the sjuzhet portrays (the scene
that in fabula fills function H) are such that the sequence must reach suc-
cessful completion (function I and function K): the Duke will marry the
Count's daughter; he will again have a Duchess. If other readers respond
as I do, then by suppressing an account of the marriage, Browning in-
creases his readers' dislike of the Duke by making us write for ourselves
the abhorrent but inevitable conclusion. In addition, although the restric-
tion of focalization to the Duke permits the suppression of some data (for
example, how the Duchess dies), the Duke's own proud acknowledgment
of his crime and its motivation leaves readers in a position to determine
that, even if his interpretation of the Duchess's behavior is accurate, his
motivation is insufficient for his crime. As a result, if the conclusion of
the fabula is only technically open, and if the Duke's actions and charac-
ter are unambiguous, readers - including James, his story suggests - must
turn their conjectures instead toward the events that precede the depicted
scene, to the Duchess herself, and to the painted image of her smiling face.
Another My Last Duchess, named after Browning's poem and painted by

James's painter Baxter, is, Baxter explains as he shows the painting to
Lennox, "an attempt to embody my own private impression of the poem,
which has always had a strong hold on my fancy" (James 1962 [1868]: 264).
Viewing the painting, Lennox perceives "the correspondence between the
lady's expression and that with which he had invested the heroine of
Browning's lines," (ibid.) as well as a correspondence between the lady in
the painting and Marian, his fiancee. Admiring both the painting and the
14. Tamar Yacobi draws attention to the effect of the even longer delay before readers are
told the identity of the narratee. In response to Langbaum's assertion that "the Duke de-
termines the arrangement and relative subordination of the parts" (1963 [1957]: 83) of the
poem, Yacobi argues convincingly that "the Duke deserves credit for engineering the grad-
ual and indirect emergence of the message, but surely not for (say) the shock given us by the
last-minute discovery of the messenger's identity [since the Duke] of course knows all along
whom he is addressing" (1987: 344).
depiction of Marian he sees in it, Lennox commissions Baxter to paint

Marian's portrait. This second painting, Baxter's portrait of Marian, is
the "masterpiece" of James's "The Story of a Masterpiece." When Lennox
sees the completed painting, it induces a change in his perception of the
woman he is to marry.
In addition to the two paintings, the fabula of James's story includes,
in the sequence in which Lennox is the C-actant, two C-decisions and the
events they initiate, as well as a new second sequence in which Baxter is
the C-actant. In the chronology of fabula, the sequence in which Bax-
ter is the C-actant begins some two years before the sequence in which
Lennox is the C-actant and concludes just as Lennox's sequence begins.
Traveling in Europe, Baxter and Marian meet and become engaged. Bax-
ter leaves for several months to paint, and returns to discover that Marian
has "trifled with the affections" (ibid.: 275) of two young men. The fabula
of Baxter's sequence can be represented by the following functions:
Initial equilibrium (Baxter engaged to Marian)

A Marian "trifles"; Baxter hurt
C Baxter decides to break engagement and leave
C' Baxter breaks engagement
G Baxter travels to Spain
H time, Velazquez, and a new love heal Baxter
I Baxter is engaged to another woman
K Baxter to marry soon, returns to New York
In Lennox's sequence, at summer's end (when Baxter returns to New

York), Lennox and Marian are recently engaged and their marriage is
scheduled for late October. Lennox is very much in love, but his "dreams
and fancies" (ibid.: 294) come to an end when his perception of Marian's
character changes in response to the portrait Baxter paints. Reasoning,
with "grim logic," that "if he had mistaken her, the fault was his own, and it
was a hard thing that she should pay the penalty" (ibid.: 295), Lennox de-
cides not to cancel the wedding. But he also decides, in a second C-decision
that initiates its own set of events, to destroy the portrait. The fabula of
Lennox's sequence, as I interpret it, can be represented by these functions:
Initial equilibrium (Lennox engaged to Marian and in love)

a Marian's painted expression alters Lennox's love
C1 Lennox decides not to cancel the marriage
C2 Lennox decides to destroy the portrait
C '2 Lennox stabs the painted face
H2 Lennox shreds the canvas
12 the portrait is ruined

K2 Lennox feels "immense relief" (ibid.: 296)
C' Lennox marries Marian
Hi the marriage
The sjuzhet of James's story begins shortly after Lennox's engagement

and traces the trajectory of Lennox's sequence in chronological order (with
exceptions noted below) to the wedding and the months thereafter. Like
Browning, James postpones the first C-decision until late in the sjuzhet.
Dividing the story into two parts, the first somewhat longer than the sec
ond, James places the completion of the portrait and Lennox's first view of
it in part 2, effectively extending throughout the first part Lennox's sense
of happy equilibrium. The duration in the sjuzhet of Lennox's period o
happiness, far beyond the initial moments of equilibrium (and no happi-
ness) in Browning's poem, is one of the ways James makes readers car
when Lennox's happiness is disturbed. In part 2 of the story, the even
of function a are extended in the sjuzhet, which lingers over Lennox
"earnest efforts [in his response to Marian] to repress an impulse of con-
stant mistrust and aversion" (ibid.: 291), shifts the focalization temporaril
to Marian (ibid.: 291-92), then follows Lennox when he views My La
Duchess again and this time finds "the likeness [to Marian] as utterly absent
as if it had never existed" (ibid.: 294). By these strategies, James postpone
Lennox's Cl-decision that the marriage "should stand" (ibid.) to the thirty-
sixth page of a thirty-eight-page story.
Just as Browning compresses the events of the seven final functions of
his poem's fabula, James postpones the events of the seven functions that
conclude the fabula of Lennox's sequence to the last five paragraphs o
his story. Four paragraphs are devoted to the destruction of the portrait.
The final paragraph includes the wedding, the discovery of the fate of th
portrait, and the first months of the marriage. Like Browning's sjuzhet,
James's ends at function H. Unlike Browning's poem, which, as I argu
above, guides readers to accept the Duke's marriage as abhorrent but in
evitable, James's story seems effectively left open. The story concludes,
"How [Lennox] has fared-how he is destined to fare-in matrimony, it
rather too early to determine. He has been married scarcely three months
(ibid.: 296).
Gerald Prince (1992: 28-38) establishes three categories of negation in
narrative, including two categories -the unnarratable and the unnarrated- in
which elements of the represented world are suppressed in the narration.
The "unnarratable" includes elements that, in a given instance, cannot
be narrated or are not sufficiently interesting to narrate. In the function
analysis of Poe's sjuzhet, the stranger's decision to respond to Aphrodite's

need (function C) is placed in brackets because it is revealed only by its
effect, the rescue of the child. The decision cannot be narrated because
it is not in the purview of the focalizer; although it is suppressed in the
sjuzhet, its effect reveals it, to the narrator and the reader, as an event in
fabula. The Duchess's murder is suppressed in Browning's sjuzhet because
the Duke chooses not to describe it, but it is revealed as an event in fabula
through the cessation of her smiles. Similarly, Oedipus's murder of Laius
is "unnarratable" except by Tiresias, whose vision exceeds the limits of
human sight, because the surviving witness is too frightened to speak, or,
if he should speak, to be believed.15
James's conclusion, on the other hand, falls into the category of the
"unnarrated": ellipses or omissions, whether explicitly stated-as in this
instance -or not. The strategy permits James's narrator to reveal the exis-
tence of information while simultaneously withholding it. Its effect, in this
instance, is to establish empty functions-functions that are empty because
the events to fill them are suppressed. A function is a position in the nar-
rative process, a placeholder that can hold-and by holding, interpret-
events. In the story, without the concluding sentences James provides, we
might interpret the wedding not as the initial act (function C'), but as itself
the primary endeavor (function H), which, once accomplished, would
create a successful conclusion (function I) and establish an equilibrium
(function K). The story's "unnarrated" ending establishes the wedding as
function C' and the marriage that follows as an empty function H, thus
permitting us to speculate whether we - and Lennox - would interpret the
marriage, if we had the information to judge it, as a success (function I) or
a failure (function Ineg).
In instances other than the open ending, however, in this very early
story by James, the potential for ambiguity in interpreting events as func-
tions is ultimately removed by information placed later in the sjuzhet-
information that includes explicit statements by the narrator, whom we
are led to trust because he is not a participant in the action and because
15. If we consider, as we must, the murder of Laius and the questioning of the witness as
separate events, the questioning falls under Prince's third category, the disnarrated: elements
included in the narration- in the negative or as hypothetical - but suppressed in the repre-
sented world (Prince 1992: 28-38). Oedipus's hypothetical questioning of the witness about
Laius's murder is proposed in the sjuzhet, just after the middle of the play, at the moment
when Oedipus and the audience are so certain he killed Laius that, unless some element to
postpone closure is introduced, the play will end. By the time the witness arrives, Oedipus's
guilt in the murder of Laius is subsumed by his guilt in his relationship with Jocasta. Fur-
ther testimony about the murder would be anticlimactic, in addition to lacking credence
because of the witness's fear.
the source of his information, which includes the motivations of all the
characters, is never portrayed.16 Although the sjuzhet follows in broad out-
line the chronology of Lennox's sequence in the fabula (his engagement
and marriage), temporal linearity is broken in the sjuzhet when the events
of Baxter's sequence are inserted, when elements of an event that help to
define it are postponed, and when the sequence in which readers receive
information is complicated by the parallel but different sequence in which
information is given to or withheld from specific characters. The distribu-
tion of elements of fabula in the sjuzhet affects readers' interpretations of
events as functions. Four examples will suffice.
First, the initial equilibrium is intensified and extended by the postpone-
ment in the sjuzhet of information that reveals that Lennox's happiness
is less secure than readers are initially led to believe. Early in the sjuzhet
Lennox admits that he is "'afraid sometimes [that Marian] doesn't really
love me'" (James 1962 [1868]: 262). But the source of his fears is with-
held-and readers' interpretation of Lennox's happy state is sustained-
until a passage in part 2, which summarizes the silent occasions and de-
picts the one time Lennox asks aloud, "Marian, where is your heart?"
(ibid.: 284). "Everywhere," she says, and the narrator explains that since
she "had distributed her heart impartially throughout her whole organ-
ism ... its native seat was somewhat scantily occupied" (ibid.: 285).
The placement in the sjuzhet of Baxter's sequence aids readers in iden-
tifying Lennox's shocked response to the portrait of Marian as the primary
motivating disruption (function a). Although readers and Lennox learn
together (first from Baxter [ibid.: 265] and then from Marian [ibid.: 268-
69]) that Baxter knew Marian in Europe while he was working on My
Last Duchess, readers are informed about the pair's prior engagement-
but Lennox is not-near the end of part 1, during the painting of the
portrait. This knowledge -and the knowledge that Lennox does not share
it--increase our expectations that Lennox's response to the painting is im-
portant enough to motivate further events.
In contrast, the elongation in the sjuzhet of Lennox's desire to believe
the painting is wrong, coupled with our knowledge of Baxter's relationship
with Marian, suggests that we interpret Lennox's response to the portrait,
not as an extremely extended deliberation to perceive the portrait's truth
(leading to a function-a change in perception), but as the behavior of a
16. Yet he speaks in the first person, expresses attitudes that suggest he lives in the same
time and place and social class as Lennox, and betrays the rudiments of a personality that
in comparison to Lennox's is less discriminating in thought and language. The narrator and
his narratees, the "gentlemen" to whom he addresses an aside (James 1962 [1868]: 260),
deserve analysis beyond what is appropriate in this context.
potential C-actant confronted with a function-A event. According to this

interpretation, Baxter's portrait is the disruptive function A, and Lennox's
decision to defend Marian (function C) is revealed (function C') in his
words to Baxter, when he first sees the portrait just before it is completed:
"Your picture frightens me, and if I were Marian I should feel as if you'd
done me a certain violence" (ibid.: 282).
Further, although Lennox never learns about Marian's former engage-
ment to Baxter, he does discover, when he asks, that Baxter was once in
love with Marian. On the basis of this information, Lennox declares to
Baxter, "You loved her, she was indifferent to you, and now you take your
revenge" (ibid.: 288). Lennox's angry words suggest a very different inter-
pretation of the events of the story from the one we have been considering
throughout this analysis:
A Marian trifles, hurts Baxter

C Baxter decides to seek revenge
C' Baxter accepts commission to paint Marian
H Baxter paints Marian to reveal her faults to Lennox
I Baxter's portrait harms Marian (as Baxter intends)
A Baxter's portrait harms Marian (in Lennox's eyes)

C Lennox decides to defend Marian
C' Lennox speaks in Marian's defense
According to this second interpretation, Baxter's sequence not only pr

cedes (as it does in fabula) but motivates Lennox's sequence, and Baxter
C-actant role is at least as important as Lennox's.
A function analysis illuminates contradictory interpretations. Shlomith
Rimmon-Kenan, in her brilliant analysis of ambiguity in James's late
works, defines narrative ambiguity as "the coexistence of mutually exclusiv
fabulas in one sjuiet" (1977: 51) and locates the first fully developed instan
(ibid.: 79) in James's "The Lesson of the Master," published in 1888. In
"The Story of a Masterpiece," written twenty years earlier, the two contra-
dictory analyses and the resulting ambiguity are not sustained throughou
the story as a whole, as they are in the later works in which Rimmon
Kenan traces the phenomenon. The function analysis above, suggested by
Lennox's reading of the painting as an act of revenge, is invalidated i
several ways, including, finally, the impossibility of interpreting Baxter's
act as the motivated and intentional act of a C-actant. Happily engage
to another woman, Baxter has no motive to prevent Marian's marriag
Nor does he recognize the opportunity for revenge that Lennox's lov
for Marian offers, since he has "taken for granted that [Lennox] had de-
signedly selected as his second wife a mere pretty woman" (James 1962
[1868]: 286).
But even this early story demonstrates the effect of sjuzhet sequence on
readers' interpretations of events as functions. The power of the sjuzhet
to govern the interpretation of functionally polyvalent events lies in large
part in the sequence in which it dispenses elements of fabula to the reader.
The set of events available to readers, at each moment as they move
through a sjuzhet page by page, determines readers' interpretations at that
moment of the discernible events as functions. Functions provide a vo-
cabulary to talk about the interpretations readers formulate as they read,
the different interpretations of readers who disagree, and the shifting in-
terpretations that are characteristic of the pattern of thought during the
process of reading.
References
Barthes, Roland
1972 [1962] "Literature and Discontinuity," in Critical Essays, translated by Richard
Howard, 171-83 (Evanston, IL: Northwestern University Press).
Booth, Wayne C.
1961 The Rhetoric of Fiction (Chicago: University of Chicago Press).
Bremond, Claude
1964 "Le Message narratif," Communications 4: 4-32.
Bremond, Claude, and Jean Verrier
1984 [1982] "Afanasiev and Propp," translated and with an introduction by Thomas G.
Pavel and Marilyn Randall, Style 18 (2): 177-95.
Browning, Robert
1988 [1842] "My Last Duchess," in The Poetical Works of Robert Browning, edited by Ian Jack
and Rowena Fowler. Vol. 3, Bells and Pomegranates, 186-88 (Oxford: Clarendon Press).
Chatman, Seymour
1988 "On Deconstructing Narratology," Style 22 (1): 9-17.
Culler, Jonathan
1981 "Story and Discourse in the Analysis of Narrative," in The Pursuit of Signs: Semiotics,
Literature, Deconstruction, 169-87 (Ithaca, NY: Cornell University Press).
Dole2el, Lubomir
1990 Occidental Poetics: Tradition and Progress (Lincoln: University of Nebraska Press).
Genette, Gerard
1980 [1972] Narrative Discourse: An Essay in Method, translated by Jane E. Lewin (Ithaca,
NY: Cornell University Press).
Greimas, A.-J.
1983 [1966] Structural Semantics: An Attempt at a Method, translated by Daniele McDowell,
Ronald Schleifer, and Alan Velie (Lincoln: University of Nebraska Press).
James, Henry
1962 [1868] "The Story of a Masterpiece," in The Complete Tales of Henry James, edited by
Leon Edel. Vol. 1, 1864-1868, 259-96 (Philadelphia: Lippincott).
Kafalenos, Emma
1995 "Lingering along the Narrative Path: Extended Functions in Kafka and Henry
James," Narrative 3 (2): 117-38.
Langbaum, Robert
1963 [1957] The Poetry of Experience: The Dramatic Monologue in Modern Literary Tradition
(New York: Norton).
Larivaille, Paul
1974 "LAnalyse (morpho) logique du recit," Poetique 19: 368-88.
Pavel, Thomas G.
1985 The Poetics of Plot: The Case of English Renaissance Drama (Minneapolis: University of
Minnesota Press).
1990 "Narrative Tectonics," Poetics Today 11 (2): 349-64.
Perry, Menakhem
1979 "Literary Dynamics: How the Order of a Text Creates Its Meanings [With an
Analysis of Faulkner's 'A Rose for Emily']," Poetics Today 1 (1-2): 35-64, 311-61.
Phelan, James
1991 Letter to the author. April 5.
Poe, Edgar Allan
1977 [1834] "The Assignation," in Stern 1977 [1945]: 192-207.
Prince, Gerald
1980 [1973] "Introduction to the Study of the Narratee," translated by Francis Mariner, in
Reader-Response Criticism, edited by Jane P. Tompkins, 7-25 (Baltimore: Johns Hopkins
University Press).
1992 Narrative as Theme: Studies in French Fiction (Lincoln: University of Nebraska Press).
Propp, V.
1968 [1928] Morphology of the Folktale, translated by Laurence Scott (Austin: University of
Texas Press).
Rimmon [-Kenan], Shlomith
1976 "A Comprehensive Theory of Narrative: Genette's Figures III and the Structuralist
Study of Fiction," PTL: A Journalfor Descriptive Poetics and Theory of Literature 1: 33-62.
1977 The Concept ofAmbiguity - The Example ofJames (Chicago: University of Chicago Press).
Stern, Philip Van Doren, ed.
1977 [1945] The Portable Poe (New York: Penguin-Viking).
Sternberg, Meir
1978 Expositional Modes and Temporal Ordering in Fiction (Baltimore, MD: Johns Hopkins
University Press).
Todorov, Tzetan [sic]
1968 "La Grammaire du recit," Langages 12: 94-102.
White, Hayden
1973 Metahistory: The Historical Imagination in Nineteenth-Century Europe (Baltimore: Johns
Hopkins University Press).
Yacobi, Tamar
1987 "Narrative Structure and Fictional Mediation," Poetics Today 8 (2): 335-72.

Funcions Anfer Propp PDF

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Funcions Anfer Propp PDF

Transféré par

Droits d'auteur :

Formats disponibles

Functions after Propp: Words to Talk about How We Read Narrative

Author(s): Emma Kafalenos

A recurrent criticism of structuralist narratology is the synchronic nature

specifically, defined "according to its consequences" (ibid.: 67). To define

1. In an essay written after this one, I analyze protagonists' interpretations--and misinter-

fested, and focalization.3 Fabula, conceived as unexpressed in any medium,

interpreted according to its consequences. The "discursive forces" of the

to name, includes functions that recur in narratives of many genres and

In addition, although Propp does not distinguish between sjuzhet and

of the narrative sequence, from the disruption of an equilibrium to the

Initial equilibrium (not a function)

E C-actant responds to test

9. Although I have derived my eleven-function model by analyzing (with the assistan

stages in the process of change from A to K: C (the decision to amelio-

As we read Poe's "The Assignation," we follow the focalization of the

Initial equilibrium: the calm of the canal at midnight

This is the reading my function analysis documents: The woman's scream

interpret a specific event in the sequence in which it occurs. My purpose

B Aphrodite screams for help

[K] equilibrium: stranger and Aphrodite are united in death

Because the child's fall is positioned at the beginning of the n

Alas! for that accursed time

They bore thee o'er the billow,

Initial equilibrium: the lovers together in London

C stranger decides to attempt reunion with Aphrodite

D stranger is tested: Aphrodite screams for help

[K] equilibrium: the potential reunion in death

perceives, "The author's finding it to be the first time-section that is 'of

a that represents a psychological reevaluation of an otherwise unchanged

The Duchess's no longer being alive, however, is doubly interpreted by

Initial equilibrium (not a function)

If we remove the brackets that indicate the suppression of an even

Another My Last Duchess, named after Browning's poem and painted by

depiction of Marian he sees in it, Lennox commissions Baxter to paint

Initial equilibrium (Baxter engaged to Marian)

In Lennox's sequence, at summer's end (when Baxter returns to New

Initial equilibrium (Lennox engaged to Marian and in love)

12 the portrait is ruined

The sjuzhet of James's story begins shortly after Lennox's engagement

analysis of Poe's sjuzhet, the stranger's decision to respond to Aphrodite's

potential C-actant confronted with a function-A event. According to this

A Marian trifles, hurts Baxter

A Baxter's portrait harms Marian (in Lennox's eyes)

C' Lennox speaks in Marian's defense

According to this second interpretation, Baxter's sequence not only pr

Vous aimerez peut-être aussi