buesch-2000FocusSample-001.tif


48 Focus on Lileratur 

Hooks, 1991. 
Jones, Jarnes. "We oJlhe '(hird Sex": Literary Reresenlalions of Homosexuality 

ill Wilhelmine Gem/any. German Life and Civilizatio n 7. N ew 
York: Peler Lang, 1990. 

Kenm.-dy, Illlbcrt C. "T he 'Third Sex' T heory of Karl i Icimich Ul n chs." 
jOllmal o!Homosexualiry 6. 1-2 (1980/81) : 10)· 11 . 

Mosse, George L. Nat iollalism and Sexllality: Respectabi/ityand Abnormal 
Sexuality ill Modem Etlrope. New York: Howard Fertig, 1985. 

Ooslcrhu is, Harry. ~ H O!Tlosexual Emancipation in Germany Before 19)): 
Two Traditions." jOllnla! 0/ Homosexuality 22. 1-2 (199 1) : 1-21. 

• __ . [ L O pposing the Doctors; int roduct ion. JOllmal 0/ J-/omosexlltI {ity 22 . \-2 
(199 1) : 29-33 . 

._, VI. Male Bonding and H omosexuality in German Nationalism. jOlin/a/ 
of Homosexuality 22. 1-2 ( 1991) : 241-247. 

Wolff, Charlotte. MagmlS Hirschfeld: A Portrait of a Pioneer in Sexology. 
wndon: Quartet Books. 1986. 

Codes of Die Marquise von 0. .. : Modes of 

Signification in Kleist' s N ovella and 
Rohmer's Film 

Joanna Kedzierska Stimmel 

"\VThilc most of the scholarl y research 0 11 Heinrich vo n Kleist's 
W Die Marquise von 0. .. seems to agree that the novella presents 

"one of the most bizarre and intriguing stories,"1 very few critics at­
tempt to deci pher the pro nounced "bizarreness" of the text in st ruc­
tural terms. Most of the research foclIses on thematic, socio-cultural, 
or psyclloallalyt ical readings of tl lC story and tncs to givc a conclusive 
meaning to t he dcscribcd cllamctcrs and evcnt~, and, consequcnLly, LO 

til e text it~el r. Ultimately, I believe that the attempts to attach one 
meaning to this multi-facetcd !lovell::! prove to be a futile undertaking. 
By posing insightful questions about lilllits of knowled ge and inter­
Ilretation, the text itself refuses such one-dimensional classifications. 

·fhe pluralism of Kleist's text best demonstrates the juxtaposi­
t ion of the cinematic interpretation of the novel la. My intention is not 
to dctermi ne whct her Eric Rohmer's Die Marquise ooll 0.. "does jus­
ti ce" to the literary sourcc. Such a comparison would be very limiting 
and indeed misleading, since both med ia, literJ.lure and the cinelll<l , 
use different practices of significiltion. Instead, 1 eX:llnine not only 
how llleaning is invested in both Lexts, but also to what degree the 
varying modes of signification and the assignment of meaning depeud 
on the specificity of the different media. Furthermore, I want to dis· 
cern if (and in what way.~) the narrative changes that Rohmer made in 
his reading of Kleist impact the plurality of the text.1 

Fiinl and I iterature are I in ked by a f ul1dament:li structural simi­
b ri lY - lhey are bOlh communication acts in which information is 
encoded by the a.ddresser (writer or film-maker) and decoded by the 
addressee (reader or SpeCl~1l0r) of the coded message. Christian Metz 
notes that bOlh "Iiter-Hure and the cinema arc by their nature con· 
demned to cOllllotation, since dellotation always prccedes their arllS-

ruC//s on Lilcm(lIr Volume 7 (2000) 



50 Focus on Lileralur 

tic enterprise. "J Both the language of literature and the language of 
film are secondary systems, since prior to literary form there is idiom­
atic and genre convention, and prior to filIn, there is perceptual anal­
ogy. There is, however, a major difference between the two communi­
cation acts. Metz goes on to argue that "[i]n the cinema, aesthetic ex­
pressiveness is grafted onto natural expressiveness - that of the land­
scape or face the film shows us. In the verbal arts, it is grafted, not onto 
any genuine prior expressiveness, but onto a conventional significa­
tion, that of language, which is generally inexpressive."4 Metz thus 
asserts that in literature we are dealing with "heterogeneous connota­
tion," which is composed of nonexpressive denotation (first order: 
linguistic signification) and expressive connotation (second order: sym­
bolic codes of literature). In contrast we are confronted with "homoge­
neous connotation" in the cinema, since both codes (that of denota­
tion and that of connotation) are expressive because of visual imagery 
involved in both codes. 

The ultimate difficulty in transmitting a written text into a 
screen version stems from the fact that the representation of the "real" 
can be executed in film (because of the expressive denotation). Even in 
the most "realistic" novel, however, the referent has no reality (be­
cause of the inexpressive denotation). On the other hand, some fea­
tures of the literary text which can be rendered by the use of written 
language (irony, metaphor, self-reflexive play with the text on the level 
of syntax and semantics) constitute a substantial challenge for the trans­
lation into the more representational and expressive medium of the 
cinema. Benjamin Rifkin defines these ambiguous features of literary 
texts which cannot be transmitted into film without some kind of 
distortion as textual indeterminacies (Unbestimmtheiten).5 These, I 
would argue, constitute the most interesting aspect of the comparison 
between the two "versions" of The Marquise von 0 .... The inability of 
the film medium to represent some of the textual indeterminacies which 
are driving the textual discourse in Kleist's text present a fundamental 
problem for the translation of this text onto the screen. 

Kleist's Die Marquise vonO ... is structured like a detective story, 
with an enigmatic "crime" at the beginning and a plot which strives to 
solve the mystery. The mysterious set of circumstances (the marquise's 
unexplained pregnancy and her search for the child's father) is pre­
sented at the outset of the novella in the form of a newspaper an-

Codes of Die Marquise von 0. .. 51 

nouncement, and the resolution of the mystery is deferred until the 
very end of the narrative. However, the genre of detective story is ironi­
cally also undermined from the beginning of the narrative. Under­
neath the mystery, there is an answer lurking behind various state­
ments made by the novella's characters and narrator. Armine Kat ill 
Mortimer writes of a double story within Die Marquise von 0. .. (, Prom 
the beginning to the very end, Mortimer suggests that the juxtaposi­
tion of the two stories drives the narrative by placing the reader be­
tween the two poles of the "apparent" (or "na·ive") and the "real" (or 
"devious"). According to Mortimer, the second story unfolds at the 
moment when the reader learns of the marquise's "reines Bewufhsein" 
and thus concludes that the marquise has (consciously or uncon­
sciously) taken part in a sexual act. 7 I want to argue that the second 
story begins to evolve at the outset of the novella, with the very an­
n~uncement in whi~h the marquise's "vortreflicher RuE" is juxtaposed 
With her unexpIa1l1ed condition. From that moment on the 
Unbestimmtheit of the story has been thcmatised in the text and' COI1-
tinues to. b~ ref~rred to by a hermeneutic code "by which an enigllla 
can be dIstmgUlshed, suggested, formulated, held in suspense, and fi­
nally disclosed."8 Because of the paradoxical nature of the novella's 
narrative (for example, to disclose and to obscure at the same time), 
the story progresses with (to use Roland Barthes' categories): snares 
(deliberate evasion of the truth, as in the mother's invented story about 
the servant, Leopardo, admitting to have raped the marquise); partial 
answers (which exacerbate the expectation of the truth, as in the count's 
swan story); jammed answers (as in the count's exclamation in the gar­
den scene "ein einziges, heimliches, geflustertes _!"9 where the solu­
tion remains 'jammed' at the tip of the tongue); and suspended answers 
(numerous dashes which stop the disclosure on the one hand, but re­
fer back to the initial dash, on the other hand). Through these strate­
gies, the enigma of the text is kept open until the narrative is ready for 
a disclosure at the end of the story. And yet, even here, as the second, 
enigmatic, story seems to meet the first, obvious, one, the text leaves 
something untold. The enigma appears to be resolved, but the force­
fully conventional happy-ending leaves the reader wondering what 
else is being kept undisclosed. 

By posing such a question, the novella's plot and its conclu­
sion constitute an ironic critique of the prevalent literary cOllvellliolls. 



52 Focus on Literatur 

it h:1s been suggested th:1t the marquise's revolt at the textual level 
appears as a protest of "a fictional character against the narrative con­
vention that apparently has her trapped."lo Moreover, I believe, the 
reader herself feels entrapped by the textual convention and, through­
out the reading process, is constantly being challenged by the discourse 
to shake them off. Hence, the deferral of the happy-ending unlil the 
second, "happier" wedding and many little "Russen, die dem ersten 
folgen"" on one hand ironically play into her expectations, but, on 
the other hand, do not allow her to accept them. The ambiguous text 
invites the reader to assume a more active role in the process of read­
ing, or, to go one step further, the reader is challenged to "write" the 
second story as it discloses, by responding to the textual indicators 
that point to the hidden text behind the upset literary convention. 

Probably the most celebrated motif of Kleist's hermeneutic 
writing is that of the "pregnant hyphen" that hides the enigma of the 
conception of the marquise's child. Although the event itself does not 
errupt into speech, the text gives many clues as to what is being hid­
den by the ellipsis. For example, the attack by the Russian troops, 
which precedes the celebrated dash, is introduced with a "Hier,"12 af­
ter which a description of the soldiers' behavior follows, which does 
not leave much room for doubt about their intentions regarding the 
marquise. The dash which hides the moment of the marquise'S sexual 
encounter with the count also follows a "Hier."1l This illustrates a 
paradigmatic relationship between this event and the previous one. 
Nonetheless, the absence of the enigmatic event places both the narra­
tor and the reader in a position of creative interpreter of the text. Even 
if the reader does not recognize the hyphen as the place where the 
ellipsis occurs, the reader has to question other indices, such as the 
count's clinical reassurance that the marquise should be fine after what 
had happened to her, or the fact that he leaves the citadel "sehr erhitzt 
im Gesicht."14 As Mehigan observes, in Kleist's text, "the expected 
celebration of the detective's - that is, the narrator's - rationality is 
forestalled so that the role of the detective, and the need to supply an 
interpretation of events, now devolves upon the reader."" Since the 
narrator does not supply the reader with more information regarding 
the textual clues, the reader must take up the role of an inquisitive 
interpreter. Shifting the responsibility to resolve the question that the 
text poses to the reader, the novella allows a plurality of meanings and 

Codes of Die Marquise von 0. .. 53 

a potentially endless number of interpretations. 

. T~1e role of the film's spectator in solving the text's mystery is 
entirely chfferent. Here, although the initial enigma of the text is not 
explicitly revealed, it is implied so strongly via cinematic conventions 
that it does not leave any doubt about the nature of the act or who is 
responsible for the marquise's pregnancy. Ironically, that which is not 
represented in the non-expressive medium of literature seems to Con­
stitute an insurmountable difficulty for the more representational and 
expressive medium of film. The absence of representation of the event 
in the Kleisti.an text- i~1deed, the gap that drives the whole story - is 
transformed 111 the [llmlllto a conventional representation of the sexual 
act which plays heavily into the spectator's expectations. 
. In the rape-sequence, the count returns to the marquise's lodg-
1I1g after her rescue. In a shot-reverse shot, the camera captures the 
count at the door, cuts to the view of the marquise on her bed in an 
unquestionably erotic position, and cuts again to the count who is 
looking at her. The sequence ends in a fade-out. Such editing does not 
leave much doubt about what happens between the two characters. 
We know that after her liberation from the Russian troops the mar­
quise has been administered a sleeping potion. In the novella, the mar­
quise is rescued by the count and then supposedly loses conscious. 
ness. "Supposedly," because the reader must question the narrator's 
emphasis OIl "vollig bewuBtlos," as though she could have been only 
partially unconscious. 16 The film, on the contrary, denies such specu­
lations. The spectator is given a more pronounced answer to the ques­
tion of what had happened between the count and the marquise after 
the fade-out. First, the procedure of shot-reverse shot establishes the 
count as the guilty party, and secondly, the narrative change (the trope 
of the sleeping potion) renders the marquise unconscious and hence 
truly "not knowing" what had happened to her. Consequentl~, in th~ 
novella, the Unbestimmtheit that is created by the lack of the rape 
initiates a crisis of knowledge for all: the characters (all but the count), 
the narrator, and the reader. The marquise's lack of knowledge and 
the issue of her conscious or unconscious involvement in the forbid­
den sexual act remain an unresolved question and point to the prob­
lem of "knowing" in general. However, in the film representation, 
such a problem is not even addressed. Whereas in the novella, the 
story behind the ellipsis has yet to be "written" by the reader (and the 



54 Focus on Literatur 

narrator, for whom the events are also just unfolding), the spectator of 
the film finds him/herself in a privileged position of knowledge from 
the beginning to the conclusion. 

The remainder of the novella's narrative also demonstrates the 
workings of the hermeneutic code. Subsequent hints that follow the 
"pregnant hyphen" appear as indices pointing back to the original 
ellipsis. When the count returns to ask the marquise to marry him, 
both lead a "secret" conversation beneath the overt one that is directed 
at the rest of the family. Throughout the conversation, numerous hints 
from the count serve to complicate the mystery by outlining it through 
half-truths, or, to borrow Barthes' terms, snares and suspended answers. 
For example, the count appears to "know" more than the marquise 
herself knows about her physical well-being. He is happy to hear that 
she "fiirchte inzwischen nicht, da~ diese (Kranklichkeit) weiter von 
Folgen sein wiirde."17 After his hasty marriage proposal, the count 
asks the marquise, "ob sie ihn verstanden hatte?", 18 and his subsequent 
statements still try to get through to the marquise. He claims that he is 
''clurch die Umstande gezwungen"19 to speak and act quickly, and as 
he mentions "dringende Verhaltnisse jedoch, iiber welche er sich naher 
auszuiassen nicht imstande sei ( ... ),"20 he looks at the marquise as if 
directing the coded message toward her, and, consequently, toward the 
inquisitive reader. Finally, in another attempt to hint at the "truth," he 
says "da~ die einzige nichtswiirdige Handlung, die er in seinem Leben 
begangen hatte, der Welt unbekannt, und er schon im Begriff sei, sie 
wieder gut zu machen ( .. . )"Y All these ambiguous pieces of informa­
tion are indices of the count's guilt and feelings of remorse. Yet, while 
pointing to the solution of the enigma of rape and the marquise's preg­
nancy, they come short of revealing everything. The mystery comes 
close to being solved, but remains jammed, as it were, at the tip of the 
tongue. The text blocks the truth from being disclosed, although it is 
deeply desired and awaited: "Man erwartete nur, nach den ersten 
Hoflichkeitsbezeugungen, da~ dieser Gegenstand zur Sprache kommen 
wiirde, um ihn mit vereinter Kraft zu bestiirmen, den Schritt, den er 
gewagt hatte, wenn es noch moglich sei, wieder zuriickzunehmen. 
Doch vergebens ( ... )"22 Here, the discourse once again gives ~ broad 
hint at what had happened between the count and the marqmse, the 
"Schritt, den er gewagt hatte" in the "apparent" story refers to the 
Count's impulsive decision to take the family's invitation to stay as 

Codes of Die Marquise von 0. .. 55 

their guest, forfeiting thus his military duties. The discourse points 
however to the second meaning of the phrase, the step that he had 
taken during his first encounter with the marquise - the rape, which 
is, here as elsewhere, not named, but, instead, referred to as 
"Gegenstand" of the conversation. "Doch vergebens ... ," despite the 
ambiguous signs, the message is not yet fully decoded, and the truth 
about the count and the marquise not disclosed. 

Most of the marriage proposal scene in the film is shot in such 
a way as to give an impression that the spectator is placed in the middle 
of the room. Thus the spectator can assume a role of an uninvolved 
and objective onlooker. Only the secret dialogue between the count 
and the marquise is constructed via point of view and alternating re­
verse shots. The marquise is shot here from a higher point of view of 
the standing count, while the latter is shot from the lower angle which 
corresponds to the perspective of the marquise sitting on the divan. As 
Mary Rhiel observes, such editing strategy places the count in the 
position of superior knowledge about the events to which the secret 
conversation refersY Here, the film is consistent with the representa­
tion of the sexual act during which the marquise was clearly repre­
sented as unconscious and unknowing. At this point, it becomes obvi­
ous that the film's attempts to recreate the secrecy of the dialogue 
through editing procedure does not make much sense since the film 
had already established that only one of the interlocutors is in full 
possession of the knowledge about the facts to which such secret dia­
logue would refer. Moreover, the spectator, also in the privileged posi­
tion of knowing more than the marquise knows, remains passive and 
merely consumes what is shown on the screen. 

It is important to note that in the novella, the scene of the 
marriage proposal is written in indirect speech with the help of mount­
ing daft-sentences. The repetition of the same grammatical construc­
tion (23 dependent clauses with daft- conjunction) for almost two full 
pages gives an additional impression that the text engages the reader in 
a linguistic game that consists of multiple use of the same form of 
language. This gives the impression of having exhausted the poten­
tially infinite number of signifiers of one and the same signified ("the 
count's guilt"'). Such repetition and variation of the signifier affirms 
then, to use Barthes' notion, "the plural existence of the text."14 In the 
film, on the other hand, as the role of the narrator is assumed by the 



56 Focus on Literatur 

camera, the whole conversation is transcoded into indirect speech, 
which robs the filmic text of the playfulness that the literary source 
has to offer. 

Finally, the Count's feverish day-dream of Thinka, the swan, 
appears as a tale of purity, desire, and rape, which might be read as a 
symbolic condensation of the whole story. Because of its importance 
for the analysis, the passage deserves to be cited in its entirety. The 
count describes: 

[W]ie er die Vorstellung von ihr, in der Hitze des 
Wundfiebers, immer mit der Vorstellung eines Schwans 
verwechselt hatte, den er, als Knabe, auf seines Onkels 
Guter gesehen; da~ ihm besonders eine Erinnerung 
riihrend gewesen ware, da er diesen Schwan einst mit 
Kot beworfen, worauf dieser still untergetaucht, und 
rein aus der Flut wieder emporgekommen sei; daB sie 
immer auf feurigen Fluten umhergeschwommen ware, 
und er Thinka gerufen hatte, welcher der Name jenes 
Schwans gewesen, daB er aber nicht imstande ware, 
sie an sich zu locken, indem sie ihre Freude gehabt 
hatte, bloB am Rudern und In-die-Brust-sich-werfen; 
versicherte plotzlich, blutrot im Gesicht, daB er sie 
au~erordentlich liebe ( .. ys 

The shifting between grammatical and biological gender creates un­
doubtedly a powerful index to the person of the marquise. Further­
more, the play with different textual levels demonstrates another at­
tempt to involve the reader in a creative collaboration with the dis­
course by filling the gap created at the outset of the text. Interestingly, 
Rohmer's film shifts the most significant part of the dream to the end 
of the narrative when the question of fatherhood has been resolved 
and followed by the reconciliation between the marquise and the count. 
Before .that has been accomplished, the count abruptly ends the retell­
ing of the dream, a break which in my view attempts to recreate the 
"pregnant hyphen" from the novella. The difference is that at this 
point within the film's narrative, this break does not point to any­
thing secret, since the act of the rape has been already established in 
visual terms at the beginning of the film. Moreover, the pronoun sie 

Codes of Die Marquise von 0. .. 57 

employed in the novella, which referred only implicitly to the mar­
quis: (c?m~licating the enigma of the text, rather than outright dis­
closmg It), IS changed at the erid of the film into du, thus leaving no 
doubt as t~ whom t~e swan represents. Hence, the plural meaning of 
the dream IS lost, as It does not move the reader to question the events 
in the same way as it does in the novella. When the dream is retold in 
the film, it serves to illustrate the already-disclosed truth. 

. An imp?~a~lt aspect that points to the plurality of Kleist's 
text ~s the mul~lpltcIt!. of narrative perspectives. Transcoding of the 
multiple narratIve pOSItIOn of the literary text creates one of the hurdles 
for the cinematic adaptation which, in my view, proved too difficult 
for Rohmer to overcome. While in the novella, the narrator assumes 
various positions, in the film, where the camera appropriates the role 
of the narrator, such plurality of positions is practically denied. Thus, 
whereas the novella in some way disconnects the authorial voice from 
its origin Oanguage, not the "author" speaking) and thus allows a mul­
tiplicity of readings, the film reinstalls the authorial voice and en-
forces one conclusive reading. ' 

There are two broader types of a narrative position that can 
be discerned in Kleist's Die Mar-quise von 0. ... In the first one, the 
narrator app~ars a~ an o~nisciel1t consciousness that tells the story 
from a supenor pomt of VIew. This type of narrator narrates either as 
a disembodied voice or from within the characters, sometimes also 
judging and interpreting their thoughts and actions. Such narrator, 
Klaus Schwind points out, "interpretiert selbst in schein bar neutraler 
Aussensicht die erzahlten Vorgange, und der Autor manipuliert durch 
die Sprachverwendung den Leser, der nUr als Eingeweihterdas ironische 
Spiel mitspielen kann."26 Mehigan proposes that the writer projects 
into the novella a narrator who em the one hand "seeks to interpret 
the events he describes and on the other wishes to remain passive and 
impartial."27 For example, in the passage mentioned above which de­
scribes the count's return and hid unexpected marriage proposal, the 
whole conversation is rendered mainly from the perspective of an 
omniscient narrator, yet, at times, the narrator's discourse slips into 
the perspective of the marquise. Although she is denied speech, the 
narrative voice refers to her perceptions and interpretations of the 
unfolding events. Narrative references such as "schon wie ein junger 
Gott,"28 "die Marquise, durch die Herzlichkeit wOluit er dies , 



58 Focus on Literatur 

vorbrachte, gut gestimmt,"29 "die Marquise wufhe nicht, was sie von 
dieser Auffilhrung den ken solle"30 represent the narrator's wavering 
between disembodied voice and interpretation. 

In the second type of narration employed in the novella, the 
perspective of the narrator seems to be limited to only that what the 
characters can see and know. At times, it seems that the events unravel 
before the eyes of the narrator in the same way as before the reader. 
One of the examples of such limited narrator's perspective is the scene 
of the reconciliation between the marquise and her father. The scene 
is presented from the voyeuristic perspective of marquise's mother, 
which limits the reader's perception to that what this particular char­
acter can see: 

Sie vernahm, da sie mit sanft an die Tilr gelegtem Ohr 
horchte, ein leises, eben verhallendes Gelispel, das, 
wie es ihr schien, von der Marquise kam; und, wie sie 
durchs Schli.isselloch bemerkte, sa~ sie auch auf des 
Commendanten Scho~, was er sonst in seinem Leben 
nicht zugegeben hatte. Darauf endlich offnete sie die 
Tilr, und sah nun - und das Herz quoll ihr vor Freuden 
empor: die Tochter still, mit zurilckgebeugtem 
Nacken, die Augen fest geschlossen, in des Vaters 
Armen liegend; indessen dieser, auf dem Lehnstuhl 
sitzend, lange, hei~e und lechzende Kilsse, das gro~e 
Auge voll glanzender Tranen, auf ihren Mund driickte: 
gerade wie ein Verliebterpt 

The mother's perspective is introduced by the dash following "und 
sah nun." If the dash, as numerous other dashes within the text, stands 
for that which is not representable, it would appear that the text 
struggles with the representation of that what the mother actually 
sees. Moreover, the reader's perspective is limited even further since 
she is first given the information about the mother's reaction ( "und 
das Herz quoll ihr vor Freuden empor") to what she perceives. Only 
here, the colon, a textual keyhole, admits the reader as the spectator of 
the events. But even then, the picture that the reader is confronted 
with is tainted by the mother's subjective perception. Interpretive de­
scriptions like "lange, hei~e und lechzende Kilsse" and "gerade wie ein 

Codes of Die Marquise von 0. .. 
59 

Verliebter!" belong to an ironic discourse, perhaps mixed with the 
implied marital jealousy. The mother's perspective appears to slip out 
of the grasp of the objective narrator who, like the reader, is trying to 

make sense of the story. 
In this and other examples of a limited narrative perspective, 

the narrator's apparent lack of control over the characters projects onto 
the reader who, throughout the reading, tries to assume control over 
the text by interpreting it. One of the signs that the text slips out of 
the narrator's grip is the disquietude of the syntax (multiplicity of 
dependent clauses with conjunctions wdhrend, indem, daft, extensive 
use of non-verbal signs such as colons, exclamation marks and dashes, 
and the predominance of indirect speech, again with mounting depen­
dent clauses, like in the marriage proposal scene). This tense syntax 
leads to frustration of the fluidity of narration - the text does not 
"read" easily, but instead requires continuous work on the part of the 
reader. 

In the cinema, the narrative position is being constituted 
mainly through the camera work. Generally, only the camera can as­
sume the role of an unseen narrator. This is due to the lack of verbal 
cues which render the subjective narrative perspective in a literary 
text. The spectator in the cinema might also look for a point of view 
shot or change in camera position and focus (blurry images for dreamy 
sequences, for example), textual voice-overs, or editing procedures (flash­
backs and flash-forwards). In Rohmer's Die Marquise von 0. .. , such 
available cinematic maneuvers are used very sparingly. Instead, the 
norm of the camera position - long takes and medium shots - is 
already established at the outset of the film. The shot of the men in an 
inn talking and laughing about the marquise's newspaper announce­
ment refuses the spectator any kind of identification with the charac­
ters. Instead, the camera demonstrates the gaze of an uninvolved but 
omnipresent observer - the men are shot from far away, there is not 
shot-reverse shot procedure. Here, and in most of the film's sequences, 
the camera is placed in front of the profilmic event, thus creating a 
theatrical atmosphere where the camera, and, by implication, the 
viewer, assumes the role of a "fourth wall." 

In the few exceptions from this norm, the camera assumes the 
point of view of one of the figures. For example, in the scene when the 
marquise first encounters the count, the latter is shot from below, where 



60 Focus on Literatur 

the marquise is supposedly lying on the ground, undeniably from her 
point of view. The count is clad in white and back-lighted in such a 
way that he appears angelic against the background of the night air. At 
this point, the recreation of the marquise's perspective demonstrates a 
successful transcoding of the same position established by the novella's 
narration, where the narrator comments on the marquise's perception 
of her rescuer: "Der Marquise schien er ein Engel des Himmels zu 
sein."32 

In very few instances, the editing procedure of shot-reverse 
shot is employed, like in the implied rape scene mentioned above. 
There, the (otherwise very rare) use of alternating point-of-view shots 
emphasizes the importance of the encounter between the two charac­
ters. However, as analyzed above, by robbing the marquise of con­
sciousness (even if only speculated consciousness), Rohmer portrays 
her as a passive victim of the male predator. More importantly for my 
purposes, the change of the narrative instance from the unknowing 
narrator (who strives to understand the events as they unfold) to the 
omniscient camera (who sees enough of the scene to come to conclu­
sions very early on within the narrative) has a far-reaching impact on 
the role of the spectator in the process of the active deciphering of the 
story. 

As mentioned earlier, cinematic message is composed of ele­
ments of two signifying systems. According to Roland Barthes, in the 
denoted system, photography or film strives to appear as a perfect 
"analogon of reality" (due to the visual nature of images.)33 The code 
of the connoted system, on the contrary, is constituted either by a 
universal symbolic order or by a period rhetoric, in other words, by 
cultural stereotypes (schemes, imagery, colors, gestures, expressions, 
and spatial arrangements). Barthes argues that every photographic image 
strives to present itself as lacking this connoted code, which is seen as 
"tainted" by cultural influences, and professes instead "to be a me­
chanical analogue of reality, (as though) its first-order message in some 
sort completely fills its substance and leaves no place for the develop­
ment of a second-order message."34 Naturally, it is important to note 
that even the visual codes of denotation that are employed in the art of 
photography are so enshrined in the web of signification we call cul­
ture that the viewers consider them to be "natural" representations of 
the "real," while actually they are nothing more but cultural constructs. 

Codes of Die Marquise von 0. .. 61 

. Rohmer's film clearly expresses a desire to pass for the analogon 
of the Ilterary text - a purely denoted translation of the novella rather 
than an adaptation which employs a variety of cultural and symbolic 
codes. Such an aspiration is discernible from Rohmer's comments about 
his film. In an interview for the New York Times, the director stated: "] 
wanted to use the text as if Kleist himself had put it directly on the 
s~reen, as if he were making the movie."35 The suggestion that the 
Viewers should "read" the film as a collaborative effort of Kleist and 
Rohmer is already evident in the opening titles. The names of both 
"co-authors" constitute here a bracket for the name of the product of 
their colla~oration. Die Marquise von 0 .... The purely denotative sta­
tus, to whi~h the film appears to aspire, would give it an objectivity 
and a~thon? through the trope of "Kleist" and the place of the no­
vella 111 the hterary canon. 

Yet the film text is definitely far removed from the literary 
original. In general, the hermeneutic code which is prominent in the 
~ovella, as the analysis above has shown, is employed minimally in the 
ftln:. Instead, th.ere is an abundance of cultural and symbolic codes, 
which are used 111 order to invest the cinematic text with ideological 
meaning. Importantly, whereas in the case of textual indeterminacies, 
the medium specificity might have been "guilty'; of the distortion in 
the transcoding procedure, here the choice of cultural codes points 
clearlJ:' to the film-maker's intentionality. For example, io the inter­
pretatiOn of the count, the film subscribes to a naive romaOtic ideal of 
courtly love and heroic knighthood. The idealistic readiog neglects 
the variety of clues that indicate the count's guilt and remorse. In 
Mortimer's view, "Rohmer's remarkably close reading is :llso a thor­
oughly romantic, angelic, and idealistic interpretation."36 The count 
is portrayed as a romantic lover and a perfect gentleman, heroic and 
k~i~htly throughout the film. In one of the few departures from the 
ong111al text of the novella, Rohmer inserts the marquise's explanation 
for why she canno: conceive of the count being the sought-after father 
of her unborn cluld. During a family conversation the marquise's 
brother jests that the tone of the newspaper response;o the marquise's 
announ~ement points to someone of aristocratic origin, possibly the 
count hunself. The marquise impulsively rejects such accusations and 
explains that the count is the only person who cannot be under suspi­
cion because of his courtly behavior on the night of the b:lttle. In her 



62 Focus on Literatur 

naivete, she insists on seeing the count as an honorable knight, whereas 
the rapist who had answered to her advertisement can be only a vil­
lain. In the novella, by contrast, even after the count returns to claim 
his paternity, the marquise rejects him as "Teufel!" and refuses to pro­
vide the family, and the reader, with reasons for this rejection. Thus, 
although the mystery of paternity has been resolved, the reader is still 
left with unanswered questions about the marquise's involvement in 
the unrepresented sexual act as well as about the reasons for her refus­
als to know. Clearly, the conventions and stereotypes against which 
Kleist's text struggles are reconstructed with full force in its film inter­
pretation. Moreover, such conventions play into the desired interpre­
tation of the film as a tale of romantic love being born out of the ashes 
of the forgiven act of rape. The narrative closure that is denied by the 
novella, transformed instead into an unresolved question, is achieved 
in the film thanks to the director's changes to the narrative and the 

narrative stance. 
To conclude, I want to return to Kleist's text. When, on the 

agreed day, the count appears before the family, the mother asks the 
hysterical marquise matter-of-factly: "Was fehlt dir? Was ist geschehen, 
worauf du nicht vorbereitet warst?"37 This question is directed also to 
the reader who must be well prepared for this disclosure since the 
discourse of the second story has led her to it throughout the narra­
tive. The same question is featured in Rohmer's film, but here, it does 
not have the same implications. The spectator of the film text is ulti­
mately denied any kind of A hal-Erlebn is, since his/her privileged posi­
tion of knowledge has been established when the rape act had been 
implied by the use of cinematic conventions. Hence, the reader is in­
volved in an entirely different interpretive process than the spectator. 
Whereas the written text involves the reader from beginning to end in 
the "writing" of the story and in questioning persisting literary con­
ventions (thus allowing multiple interpretations stemming from each 
reader separately), the film offers answers to the viewer and imposes 
upon his/her the closure of the film text. Borrowing Barthes' notions, 
the novella can be termed a "text of joiussance," which, by resisting 
one-dimensional interpretation and escaping closure, presents the cri­
sis of the writer's (and the reader's) relationship with language and 
knowledge. Comparatively, Rohmer's film as a "text of pleasure" un­
folds accordingly to the spectator's expectations, and thus makes him/ 

Codes of Die Marquise von 0 ... 
63 

her happy (pleasured) by confirming his/her relationship to language 
and knowledge as stable and firm. Undeniably, some of the changes in 
interpretation are closely bound to the medium specificity as the in­
formation that could have been only hinted at in the written text must 
be more or less explicitly shown in the film. But in the majority of the 
instances when the message delivered by the novella has been distorted 
by the film, not only medium specific factors can be blamed. Rohmer's 
use of cultural codes and his narrating procedures testify to conscious 
choices made by the film-maker aiming to deliver a very specific mes­
sage. Barthes' distinction between the "classic" and the "modern" novel 
captures well the difference between the film and novella, respectively: 
"[I]n the former we have a view of History which is harsh but coher­
ent and certain of its principles, the triumph of an order; in the latter, 
an art which in order to escape its pangs of conscience either exagger­
ates conventions or frantically attempts to destroy them."38 The no­
vella, although chronologically preceding the film by almost two cen­
turies, thus appears more "modern" in its attempts to play with the 
straight-jacket of literary conventions, and in its questioning of our 
relationship to knowledge and the limits of interpretation. 

GeorgetlYWn University 

Notes 

1John M. Ellis reviews the state of Kleist scholarship up to 1976 in 
his book The Stories of Kleist. Studies in the Meaning and Character of His 
Writings (Chapel Hill, 1979). 

2 The existing research on the comparison of the two texts is rather 
limited, and the few scholarly undertakings to date deal with different 
questions than those I am posing. For example, in a detailed essay informed 
by semiotic theories, Thomas Bauermeister poses the important question 
of the "medienspezifische Kommunikationsebene." Bauermeister examines 
the texts very carefully by concentrating on units of language(s): signs, 
words and sentences for literature, and shots for the film, and structures of 
narration. However, he limits his inquiry to only one text passage and one 
corresponding film sequence. The result is a very thorough analysis of one 
specific moment within the narrative; however, he does not give much 
insight into the final "meanings" that are delivered by the novella and the 
film. For more see Thomas Bauermeister "Erzahlte und dargestellte 



64 Focus on Literatur 

Konversation" in Klaus Kanzog, ed., F.rzahlstrukturen - Filmstrukturen: 
Erzahlungen Heinrich von Kleists und ihre Jilmische Realisation (Berlin: 
Erich Schmidt Verlag, 1981). 

In one of the more interesting examples of this scholarship, Mary 
Rhiel comparatively analyzes the film and the novella on the level of 
authorial discourse. She claims that thanks to the spectatorial expectations 
created by the " ... pre-textual discourse (Kleist's novella), the film text is 
constructed around a phenomenological center and privileged point of 
view created through the textual figure of the author, Kleis."( Mary Rhiel, 
"The Author-Function as Security Agent in Rohmer's Die Marquise von 
0 ... " The German Quarterly 64.1 (Winter 1991: 7). Rhiel adds that Rohmer's 
rendition of the Marquise von 0 ... delivers a very "un-Kleistian" message: 
"Whereas Kleist's novella disrupts the stability of any coherent understand­
ing of reality and final resolution of the narrative crisis, the idealistic 
realism of Rohmer's mise en scene subscribes to a belief in the unique 
ability of film to capture reality." (Rhiel,13) Although I agree with Rhiel's 
conclusion, I believe that her examination of the film and the novella lacks 
a very important issue of the specificity of the two very different media of 
communication. One cannot assume that meaning can simply be shifted 
from one text to the other, but, instead, it must be seen as "transcoded in 
the process of projection of elements from one system into the 
other"(Benjamin Rifkin, Semiotics o/Narration in Film and Prose Fiction 

(New York: Peter Lang, 1994) 47. 
) Christian Metz, Film Language: A Semiotics 0/ the Cinema. (New 

York: Oxford University Press, 1974) 76. 
4 See Metz, 77. 
5 In his analysis of two Soviet filmic adaptations of literary works, 

Rifkin draws on the theories of Soviet semioticians and practitioners of the 
cinema, most importantly Eisenstein and Lotman. See Benjamin Rifkin, 
Semiotics o/Narration in Film and Prose Fiction (New York: Peter Lang, 

1994). 
6 Armine Kotin Mortimer, "The Devious Second Story in Kleist's 

Die Marquise von 0. .. " The German Quarterly 67.3 (Summer, 1994) 293. 

7 Ibid. 
8 See Barthes. 
9 H. v Kleist, Die Marquise von 0 ... , In: Samtliche Werke, Vol. 2 

:171. 
IOAnthony Stephens, Heinrich von Kleist: The Dramas and Stories. 

(Oxford/Providence: Berg, 1994) 215. 
11 H. v Kleist, Die Marquise, 186. 

Codes of Die Marquise von 0. .. 

12 H. v Kleist, Die Marquise, 144. 

13 H. v Kleist, Die Marquise, 145. 
14 Ibid. 

65 

15 Timothy J. Mehigan, 'lext as Contract: the Nature and Function 0/ 
Narrative Discourse in the Erzahlungen 0/ Heinrich von Kleist. (Frankfurt am 
Main: Peter Lang, 1988),313. 

16 H. v Kleist, Die Marquise, 145. In fact, throughout the novella, 
the text gives also additional hints to the fact that the marquise might know 
more that she would like to admit. For example, after the visit of the 
doctor and his confirmation of the marquise's pregnancy, the latter exlaims 
in despair: 

Was kann dieser Mann, der uns bis auf den heutigen Tag 
schatzenswurdig erschien, fur Griinde haben, mich auf eine so mutwillige 
und nidertrachtige Art zu kranken? Mich, die ihn nie beleidigt hatte? Die 
ihn mit Vertrauen, und dem Vorgefuhl zukunftiger Dankbarkeit, empfing? 
Bei der er, wie seine ersten Worte zeugten, mit dem reinen und 
unverfalschten Willen erschien, ZlI helfen, nicht Schmerzen, grimmigere, als 
ich empfand, zu erregen? (162) 
The reader cannot help but wonder who is meant by "dieser Mann" who 
was "empfangen" by the marquise. The double meaning is obvious: in the 
apparent reading of the story, the passage refers to the doctor, but in the 
hidden 'second' story, the text clearly points to the count, and, by implica­
tion, to some kind of knowledge on the part 'of the marquise. 

17 H. v Kleist, Die Marquise, 150. 
18 Ibid .. 
19 Ibid. 

20 H. v Kleist, Die Marquise, 151. 

21 H, v Kleist, Die Marquise, 152. 

22 H. v Kleist, Die Marquise, 156. 

23 Rhiel, 10. 
24 See Barthes, 58. 
25 H. v Kleist, Die Marquise, 156-157. 

26 Schwind, 110. 
27 Timothy J. Mehigan, Text as Contract: the Nature and Function 0/ 

Narrative Discourse in the Erzahlungen 0/ Heinrich von Kleist. (Frankfurt am 
Main: Peter Lang, 1988) 126. 

28 H. v Kleist, Die Marquise, 148. 

29 Ibid. 

30 H. v Kleist, Die Marquise, 149. 

31 H. v Kleist, Die Marquise, 180-181. 



66 Focus on Literatur 

32 H. v Kleist, Die Marquise, 144. 

33 Roland Barthes, A Barthes Reader. Ed. Susan Sontag {New York: 
Hill and Wang, 1995} 196-197. 

34 Barthes, Reader, 197. 

3S Alan Spiegel, "The Cinematic Text: Rohmer's 'The Marquise of 
0 ... '" Modern European Filmmakers and the A rt 0/ Adaptation Ed. I'lorton, 
Andrew and J. Margetta {New York: Frederick Ungar Publishing Co., 
1981} 316. 

36 Mortimer, 297. 

37 H. v Kleist, Die Marquise, 183. 

38 Barthes, Reader, 51. 

Works Cited 

Anker-Mader, Eva-Maria. Kleists Familienmodelle - 1m Spannungs/eld 
zwischen Krise und Persistenz. Munchen, 1992. 

Barthes, Roland. A Barthes Reader. Sontag, Susan, ed. New York, 1995. 
---. SIZ. New York, 1974. 
Bauermeister, Thomas. "Erzahlte und dargestellte Konversation." In: Ed. 

Kanzog, Klaus. Erzahlstrukturen-Filmstrukturen: Erzahlungen 
Heinrich von Kleists und ihre filmische Realisation. Berlin, 1981. 

Beckmann, Beat. Kleists Bewusstseinskritik: Eine Untersuchung der 
ErzahlJormen seiner Novellen. Bern, 1978. 

Ellis, John M. The Stories 0/ Kleist. Studies in the Meaning and Character 0/ 
His Writings. Chapel Hill, 1979. 

Grathoff, Dirk, ed., Heinrich von Kleist: Studien zu Werk und Wirkung. 
Opladen, 1990. 

Guenther, Beatrice M. The Poetics 0/ Death: The Short Prose 0/ Kleist and 
Balzac. New York, 1996. 

Heinritz, Reinhard. Kleists Erzahltexte: Interpretationen nach /ormalistischen 
Theorieansatzen. Erlangen Studien: Band 42, 1983. 

Kanzog, Klaus, ed., Erzahlstrukturen·Filmstrukturen: Erzahlungen Heinrich 
von Kleists und ihre filmische Realisation. Berlin, 1981. 

Kleist, Heinrich v. Samtliche Werke und Briefe in vier Banden. Frankfurt 
a.M., 1990. 

McGlathery, James M. Desire's Sway: The Plays and Stories 0/ Heinrich von 
Kleist. Detroit, 1983. 

Mehigan, Timothy J. Text as Contract: the Nature and Function o/Narrative 
Discourse in the Erzahlungen 0/ Heinrich von Kleist. Frankfurt am 
Main, 1988. 

Metz, Christian. Film Language: A Semiotics 0/ the Cinema. New York, 

Codes of Die Marquise von 0. .. 

1974. 

Mortimer, Armine Kotin. "The Devious Second Story in Kleist's Die 
Marquise von 0. .. " The German Quarterly 67.3 (Summer, 1994): 
293-303. 

Muller-Seidel, Walter. Versehen und Erkennen. Kaln, 1961. 

67 

---. "Die Struktur des Widerspruchs in Kleists 'Marquise von 0. .. '" Deutsche 
VierteljahrsschriJt 28. 28 {1954}: 497 - 515. 

Politzer, Hainz. "Der Fall der Frau Marquise: Beobachtungen zu Kleists 
'Die Marquise von 0 ... '" Deutsche VierteljahrsschriJt 51.1 (1977): 
98 - 128. 

Rhiel, Mary. "The Author-Function as Security Agent in Rohmer's Die 
Marquise von 0. ... " The German Quarterly 64.1 (Winter 1991): 6-
16. 

Rifkin, Benjamin. Semiotics o/Narration in Film and Prose Fiction. New 
York, 1994. 

Schwind, Klaus. Heinrich von Kleist: Die Marquise von 0. ... Frankfurt am 
Main, 1991. 

Spiegel, Alan. "The Cinematic Text: Rohmer's 'The Marquise of 0 .... '" 
Modem European Filmmakers and the Art 0/ Adaptation. Horton, 
Andrew and J. Margetta, eds. New York, 1981. 

Stephens, Anthony. Heinrich von Kleist: The Dramas and Stories. Oxford/ 
Providence, 1994. 

Swales, Erika. "The Beleaguered Citadel: A Study of Kleist's Die Marquise 
von 0. ... " Deutsche VierteljahrsschriJt 51.1 (1977): 129 - 147.