A Poetics of Free Indirect Discourse in Narrative Film

Mohammad Ghaffary, Shiraz University, Shiraz, Iran

Amir Ali Nojoumian, Shahid Beheshti University, Tehran, Iran

Download PDF Version


This essay provides, for the first time, a model for identifying and analyzing “free indirect discourse” (FID) in narrative film, the most problematic mode of representing characters’ discourse which has received little attention from film theorists and critics. According to the established “dual-voice” hypothesis, FID is an ambiguous merger of the narrator’s voice and the character-focalizer’s, without one predominating over the other. The basic argument of the essay, then, is that FID occurs in a film at the moment when the spectator is not able to distinguish narratorial objectivity from characterological subjectivity. This characterizes the narrative text as polyvocal / polyphonic, leading to artistic ambiguity and such processes as “différance” and “deterritorialization.” Based on this theory, the researchers offer a detailed analysis of the textual markers and major functions of FID in filmic narratives. The model provided can be adopted for analyzing any narrative film.

[Keywords: free indirect discourse (FID), subjectivity, objectivity, focalization, artistic ambiguity]


The same approach to the inner worlds of characters developed by modernist writers can be observed in what is generally referred to as “art cinema” or what Pier Paolo Pasolini terms the “cinema of poetry,” for example, those mid-century movies associated with neorealism. From this perspective, these films are comparable to modernist fiction. David Bordwell suggests that such techniques as subjectivization (= internal focalization) came into prominence in cinema with the advent of art films because it is in such films that individual subjectivities acquire significance. One of the major discursive devices adopted by modernist writers to depict the mental processes of alienated or self-conscious characters was “free indirect discourse” (FID). It is the most problematic mode among the four major modes of representing characters’ discourses, the others being direct discourse (DD), indirect discourse (ID), and free direct discourse (FDD). FID is generally defined as a “dual voice,” i.e., an ambiguous, polyphonic mixing of the narrator’s and the character-focalizer’s voices. Based on this established theory, in this essay, we offer a model for identifying and analyzing FID in narrative film, including a detailed survey of its textual markers and major functions.

Free Indirect Discourse (FID) in Cinema

So far, several theorists and critics have, directly or indirectly, dealt with the concept of FID in film. What is shared by almost all these studies is the “ambiguous” character of FID, i.e., the intermixing of narrative voice and internal focalization, or, put differently, when the spectator is not able to distinguish whether a particular part of the narrative text is the FCD’s account of or judgment about the character in question (ID) or the character’s own unfiltered voice (FDD / DD). Of course, different researchers have employed different terminology for pointing out this feature. (“Filmic composition device” [FCD] is a term proposed by Manfred Jahn [A Guide F4.1.2.] to refer to the narrator in the cinema. The reason for adopting such a term is that the nature of narration in cinema is far different from that in prose fiction because, while the medium of prose fiction consists solely of words, that of film is composed of images, words, and sounds, among other elements, each of which provided by a separate agent or group of agents. Thus, the neutral term FCD is employed to indicate that the cinematic narrator is not a homogeneous, monolithic agent with a humanlike voice.)

The first film theorist who has focused on FID in the cinema is Pier Paolo Pasolini (1922-75), the Italian filmmaker, poet, and writer. In his classic essay “The Cinema of Poetry” (1965), Pasolini claims that the cinema is in essence “poetic” or metaphoric because the primary objective of the cinematic language is not communicating but embodying man’s dreams with its expressive power, whereas literary works are created by means of a language (words), the primary function of which is communication (543 & 547). The truth of this claim is, of course, open to doubt; for instance, what about a televisual news report or a documentary? Does such a work not primarily communicate something to the audience? Afterwards, Pasolini states that the language of cinema, which he limits to images, however, can be communicatory only in the sense that the units of this language, i.e., “im-signs,” also compose the world of man’s memory and dreams. That is, human beings perceive many things around them or inside themselves through series of images or im-signs. Nevertheless, the language of im-signs differs from verbal language (linguistic signs) in that the former is instinctive, pre-grammatical, pr-morphological, irrational, and profoundly oneiric.

For Pasolini, true im-signs have a double character, i.e., they are at once objective and subjective: im-signs are by nature objective and conventional, but when a filmmaker employs specific im-signs to create an artistic work and convey his/her intended meaning, they take on a subjective aspect (548). Pasolini sees FID as the constituent element of cinema’s poeticality, on the grounds that this technique has the ability to represent the double nature of cinematic images. As one can see, by “subjectivity,” which is mixed with objectivity in the cinematic FID, Pasolini means the author’s / FCD’s subjectivity rather than the characters’.

Therefore, Pasolini believes that FID is possible in the cinema, but he emphasizes that it is different from FID in fiction. In Pasolini’s opinion, “cinema does not have the faculty of interiorization and abstraction” (551), because the film medium, i.e., the image, is concrete, contrarily to the abstract medium of a writer (words). This is another way to say that full internal focalization is impossible in the cinema. However, he mentions that a filmmaker can represent a small part of characters’ internal world through the cinematic FID, which, because of the difference between the two media, does not perfectly correspond to the literary FID. To show the difference, he coins the term “free indirect subjectivity” as substitute for FID in film (ibid.), but using this new term does not make any real difference in the theory or practice of FID. We prefer to adopt “FID” because, having this term, there is no need to coin a new one: for sure, “discourse” implies in it “subjectivity,” whether we have in mind the narrator’s subjectivity or the character’s. Furthermore, as is discussed below, there is no essential difference between the cinematic FID and the literary FID; thus, it serves us better to adopt a single term for both.

Pasolini believes that FID is textualized in film whenever the FCD, the “filmmaker” in his own terminology, reveals the “original oneiric, barbaric, irregular, aggressive, visionary qualities” of the cinematic medium (= images / im-signs) (552). Nevertheless, Pasolini does not put forward any concrete or particular textual marker for FID in film, and contents himself with the technique of representing the story world from the subjective viewpoint of characters who stick to or are obsessed with details. This is simply a generalization that, in practice, does not help the film analyst to discern FID. Another problem with his account of FID is that by the cinema of poetry, the essential feature of which is FID, he refers to those films in which the director (= FCD) deals with the dominant mental state of a neurotic or abnormal character (555), but the point is that FID cannot necessarily be used to represent mentally sick characters, but also a character who simply reflects upon or is conscious of his/her existence or surroundings.

After Pasolini, the most prominent figure who has directly commented upon FID in film is Gilles Deleuze (1925-95), the Poststructuralist French thinker. He has pointed to and discussed FID both in his works on language and literature and in those on cinema. In his seminal books Cinema 1: The Movement-Image (1983) (chap. 5) and Cinema 2: The Time-Image (1985) (chaps. 6 & 7), Deleuze discusses the notion of FID in cinema. He builds his idea of cinematic FID upon V. N. Vološinov’s and Pasolini’s concepts. For Deleuze, FID is a key component of what he terms “perception image,” a subtype of the movement-image, the dominant mode in classical cinema, as opposed to modern cinema, characterized by the time-image. Perception-image is a “semi-subjective” image (see below), for which there is no equivalent in human’s natural perception: it represents a perception that is neither subjective (DD) nor objective (ID) (Deleuze, Cinema 1 72). The perception image finds a special status in FID (74 & 76); put another way, FID best manifests the semi-subjectivity of the perception-image. For this reason, Deleuze names this special case of the perception-image “dicisign” (76), a term originally coined by Charles Sanders Peirce (1839-1914), the U.S. philosopher and semiotician, but appropriated by Deleuze to mean the particular sign of the composition of the perception-image. In cinema, FID develops the self-consciousness of the camera, thus leading to a correlation between two heterogeneous subjects, namely, the FCD (represented by the camera consciousness) and the character (represented through the perception-image) (Marks 239; see also Deleuze, Cinema 1 74). Thus, for Deleuze, FID in the cinema, or the dicisign, is the representation of a perception of another perception through the camera-consciousness (Cinema 1 217).

It is worth mentioning that Deleuze bases this idea of camera-consciousness upon Jean Mitry’s concept of the “semi-subjectivity” of the cinematic image, by which he means that the camera is neither totally inside the cinematic setting nor entirely outside it. Put simply, the camera-eye is, thus, neither purely objective nor purely subjective but semi-subjective— particularly, when its role is foregrounded, for example, in shots / sequences that seem as if the camera represented the eyes of an impersonal observer who is with the filmic characters or follows them. Using a Heideggerian term, Deleuze calls this the “‘being-with’ of the camera” (Cinema 1 72).

Like Pasolini, Deleuze does not mention any exact textual marker for FID in films, and instead goes to great lengths to talk about the meta-functions of this device (see below). Nevertheless, he seems to have his own reasons for not providing any description of the exact textual markers of FID, whether in literature or in cinema: he regards FID as a machine for producing subjects (“dual subjectivation”) rather than as a textual figure or a stylistic device; he believes that FID cannot be described or studied in linguistic terms or categories, because such terms are always homogeneous, whereas FID, indeed, suggests the heterogeneity of the system in which it is used (language / film) (Schwartz 125).

Deleuze’s notion of the “dicisign” (= FID) and Pasolini’s concept of “free indirect subjectivity” have a lot in common. However, one major point of difference between them is that Pasolini considers for FID in cinema a function similar to that of metaphor, resulting in a poetic cinema, whereas Deleuze treats FID as the exact opposite of metaphor because, to him, contrarily to FID which marks the heterogeneity of the system, metaphor “homogenizes” the system (Deleuze, Cinema 1 73). Thus, to Deleuze, the fundamental act of language / cinema is FID not metaphor (see also Deleuze & Guattari 85).

According to Deleuze, FID, resulting in a “dividing-in-two” of the subject, characterizes not only cinema but also thought and art (Cinema 1 73). However, this is not a functional or practical thesis. It is totally at odds with the objectives of an analytic or Structuralist study, because such theses blur the boundaries and categorizations provided by descriptive studies. What Deleuze does is only to take the established notion of FID as a dual voice and reformulate it in his own Poststructuralist discourse, linking it to such notions as split subjectivity or the system without a center / unity. In effect, he does not add anything to the theory of FID that could be adopted in a narratological study.

This considered, we turn to what film narratologists have proposed about FID in film. Inez Hedges, who has conducted one of the earliest works in this field, argues that, like verbal narratives, in films, too, there can be a situation like “dual perspective” or “substitutionary narration,” i.e., FID. This situation is a temporary fusion of the extradiegetic narrator and the intradiegetic character; in other words, the “narrator tells what the character subjectively feels, perceives, or thinks” (289-90). Nevertheless, seemingly, for Hedges, FID is ultimately the FCD’s voice, but imitating or colored with the character’s discourse. This can reduce the function of FID to discourse report or even ID, because in this case ambiguity and its corollaries would be completely irrelevant to the issue of FID.

Nick Browne believes that “a combination of narrative authority and spectatorial identification produces a perspective which overrides the represented, optical point of view” (Burgoyne 86). What Browne postulates as the fusion of the representing (= FCD) and the represented (= character) resembles the fusional nature of FID: the spectator is placed in two or more positions, identifying with the character but influenced by the FCD’s authority at the same time.

According to François Jost, in a narrative, sometimes, we may not know whether the FCD or the character-focalizer is telling the story, i.e., internal focalization and narration are confused (Burgoyne 93). Likewise, Mieke Bal argues that “the identification between the external focalizer [from within] in visual images and an internal focalizer represented in the image can . . . give rise to” a discursive conflation called FID (164).

Edward Branigan—like Gérard Genette (Narrative Discourse) and, in a different way, like Vološinov and Deleuze (Cinema 1)—treats FID, named by him “free indirect style,” as a subcategory of ID and suggests that its function in film is close to that of the subjective shot, i.e., the representation of a character’s inner world / mental processes (= “character projection”). For him, FID can be realized when a particular neutral or objective shot / sequence is metaphorically loaded by a character’s subjectivity, as if the character were telling the story with a third-person voice. Then, it is unclear whether we observe what the character perceives / thinks or the FCD’s report of it (Point of View 224-25 & 241). Furthermore, Branigan claims that identifying and understanding FID in film—which can be considered as a development of POV and perception shots—is often difficult because it has no specific textual feature and the viewer must infer it from various filmic signs (226).

In his Narrative Comprehension and Film, Branigan points to FID in film once again, but this time using the term “free indirect discourse.” Here, he does not discuss the techniques or devices by which FID can be textualized in a filmic narrative and contents himself with a brief note about its nature. According to him, in FID, which is a version of internal focalization, portions of the character’s actual thoughts are mixed with the FCD’s interpretation of them in a way that the spectator “cannot separate the two nor evaluate the accuracy of the interpretation”; the spectator “knows only that there exists a mixture of internally focalized and narrated descriptions” (168-69).

In FID, since the exact point where narration and focalization overlap is not clear, we cannot be sure to what extent the information provided by the narrator is shared by the character (Forceville 121). The spectator, thus, shares with the narrator a knowledge of the narrative which the character may lack (ibid.: 125). Therefore, one might redouble Babak Ahmadi’s statement that FID in film is neither purely subjective nor purely objective but perhaps “semi-subjective” (262)—a term Jean Mitry had used to characterize the film medium or cinematic images in general.

Finally yet importantly, FID in film seems to be mainly, not solely, FIP (free indirect perception) or FIT (free indirect thought). It is less probable that the FID shot / sequence is a character’s free indirect “speech” (FIS), partly because most of the subjectivizing or semi-subjectivizing cinematic devices are concerned with the character’s perception or at best his/her mental processes, though, in the final analysis, there is no clear-cut distinction among these categories.

The Textual Markers of FID

Unlike literature, in narrative film, there are no textual elements peculiar to FID. Therefore, to identify FID in film, we must examine those points in the narrative where objectivity and subjectivity or narration / external focalization and internal focalization merge together and the narrative becomes ambiguous from this viewpoint. Almost all devices and techniques used for subjectivization, stream of consciousness, internal focalization, or even external focalization from within may be employed to represent FID, as well, for instance, dialogue, voice-over narration, sound perspective, sound bridge, music, POV structure, eye-line match, shot / reverse shot, flash frames, freeze or still frames, slow motion, repetition of particular images, flashback, camera movement, superimposition, matte shot, snorri-cam, and double exposure, among others. The important point is that these devices can be considered to signal FID whenever they create ambiguity, i.e., when we are not able to decide whether they represent the character’s discourse or the FCD’s / external focalizer’s discourse.

Apart from the above-mentioned techniques, some others might be utilized to represent FID specifically. One of them is “subjective voice.” Subjective voice, or “internal sound,” can signify FID when the voice is the internal voice of the character but talking about him-/herself as the “other” (Ahmadi 262), as in retrospective commentary. Then, although the voice is the character’s, the scene is not purely subjective, so it can be considered FID. More than that, an extradiegetic voice-over who retells the focal character’s inner thoughts, while the character is shown to be reflecting, can be regarded as FID, too.

Hedges suggests that FID in film can be represented by splitting the narrative point of view (= focalization) between the narrator and the character, for example, the information on the image track can be attached to or controlled by one of them and that on the sound track to the other (290). The extradiegetic voice-over can be an example of this technique. Another example could be an open POV shot (i.e., a POV structure in which a point-glance shot is defined and promises a point-object shot, but we never see the promised object; see Branigan, Point of View 203-11) accompanied by an extradiegetic voice-over.

According to Julian Murphet, two kinds of focalization are possible in film, namely, associative and affective, between which there is sometimes no sharp distinction. He, furthermore, believes that if the two kinds are cleverly combined together, the produced effects will be close to those of FID in fiction (91). The eye-line match seems to be the best means for this end, particularly, when the point-object shot, to use Branigan’s term, does not show exactly what the character could have seen; for instance, the distance between the object and the subject might become shorter via the use of zoom lens, or the angle of vision from which the object is shown might be slightly different with that of the character. Then, the shot in question is neither totally subjective nor totally objective because the FDC’s trace is clearly observable in the shot. Thus, this type of eye-line match can be treated as FID. Of course, this device appears similar to Branigan’s cheated or forged POV structure (see Point of View 203-11); however, one can claim that, in the case of the latter, the character’s discourse is so foregrounded that we do not hesitate to treat it as DD.

Another situation in film narrative that can be interpreted as FID happens when there are some contradictions in the relationship among the elements of a POV structure, as conceived of by Branigan, which could lead to a kind of ambiguity associated with FID. For example, consider a character who, in a point-glance shot, looks at a scene but, just before the point-object shot begins, s/he turns his/her face away from that scene (Branigan, Point of View 162). Is the following point-object shot his/hers or the FCD’s? Ambiguity in a POV structure may arise where more than one point, whether of glance or of object, are shown; these points can even be contradictory. Some other times, the transition between the two shots may not be very clear, thus leading to ambiguity as to whether the object has been seen by the character shown or not (FID). A similar point is implied by Charles Forceville (127-28). We can draw this conclusion from his argument that, throughout a film, some shots can appear to be (open) POV shots, but logically, at that moment, the character at issue could not have been in the place from which the camera frames the images. Therefore, finally, it is unclear whether these shots should be attributed to the character or to the FCD; therefore, they are FID shots.

What Branigan calls an “implicit and indefinite” POV structure can also be possibly an FID structure, for example, an open retrospective POV structure in which “we [only] infer that someone sees” the object (Narrative Comprehension 163) (according to Branigan, a “retrospective” or “discovered” POV structure is one in which the point-object shot comes before the point-glance shot). In such a POV structure, the character-subject’s perception of the object is only implied, and the viewer does not know exactly how, when, and from where the character looks at the object. We may even seriously doubt if it is a POV shot because it is more like an objective shot. Thus, it can be a perfect example of FID in film.

The Major Functions of FID

In his study of FID in prose fiction, Michal Peled Ginsburg argues that FID is not a mere stylistic device for implying variety or impersonality; rather, it reflects on the microcosmical level, i.e., single sentences, the total thematico-rhetorical concerns of the text (53-54). Brian McHale offers a similar idea, proposing that the number of possible functions of FID is infinite: each FID sentence can have its own unique textual or thematic function, such that one cannot generalize about them (“Free Indirect Discourse” 207). In order to determine such functions, the reader must analyze FID by situating it in the overall structure of the text. However, he goes on to state that such functions are “second-order” functions and that there are first-order functions that have the capacity for generalization. Here, we explore the possible first-order or immediate functions and effects of FID in film.

We can consider three types of function for FID in fiction, namely, authorial functions, textual functions, and what McHale labels “meta-functions” (ibid.). Theoretically, all these functions can be enumerated for FID in film, too, bearing in mind the differences between the two media. Then, in summary, the authorial functions of FID in film would be as follows: 1) it is a labor-saving, economical narrative instrument for characterization or focalization, helpful in avoiding monotony in discourse representation (Fowler 174; Palmer 605); and 2) it is a useful device for reflecting “situations of tension, crisis, upheaval, turmoil, spiritual searching, and inward struggle” in film narratives (Palmer 605).

The textual functions of FID in film include: 1) it is a device for controlling the degree of distance between the FCD and the character: it can either raise empathy in the reader / FCD for the character or lead to the ironic repudiation of the character by the FCD (McHale, “Speech Representation” 437); 2) it may entail dramatic irony; 3) it can cause “irony of register” (McHale, “Free Indirect Discourse” 208) (the association of the objective or formal style of the total narrative and the semi-subjective style of the FID sequences); 4) it represents internal focalization or the presentation of the focalized from within; 5) it is a good technique for representating stream of consciousness (Banfield 29; McHale, “Free Indirect Discourse” 209 & “Speech Representation” 437; Rimmon-Kenan 115; Jahn, Narratology N8.9.) (Pasolini sees FID as a useful technique for rendering a character’s stream of consciousness in film, though he does not mention the term “stream of consciousness” [554]); 6) it suggests polyvocality or polyphony (the FCD’s and the character’s voices directly interact with each other, without either of them being dominant) (McHale, “Free Indirect Discourse” 212; Rimmon-Kenan 117); 7) it enhances a film’s power of defamiliarization; and 8) FID adds to “the semantic density” of the cinematic narrative (Rimmon-Kenan 115).

Then, the meta-functions of FID in film are as follows: 1) it is an index of the filmic (its bi- or polyvocality reflects the most characteristic feature of the fiction film as a genre, a feature which this type of film has in common with the novel) (cf. McHale, “Free Indirect Discourse” 215); 2) it is both a representation of an object and in itself an object of representation (through FID, subject and object dissolve into each other) (Ginsburg 29); 3) it reinforces the process of différance in the film text (Jacques Derrida’s notion); 4) it is a miniature reflection of the nature of all filmic texts, including such qualities as iterability and intertextuality (cf. Rimmon-Kenan 116); 5) it is the driving force of film language in the course of evolution since it leads to the interplay of contradictory cinematic signs used by or attributed to different consciousnesses (cf. Schwartz 113); 6) it marks the filmmaker’s class consciousness as, through FID, s/he imitates the socially different language of characters who belong to a usually lower class (Pasolini 550); and 7) FID is a central feature of all great films because it “deterritorializes” the cinematic and creates “difference” (Deleuze’s idea) (Colebrook 113-14); in other words, it reveals the heterogeneity of cinema and shows that cinematic signs do not derive from coherent subjects, but it is the cinema that produces subjects.

In addition to the above functions, which were adapted from different studies of FID in fiction, several others have been proposed by various researchers specifically for FID in film. Another textual function of FID in filmic narratives is that the character is “neither totally aware nor totally unaware of” his/her significance (Forceville 124). This function resembles the one mentioned above concerning the creation of dramatic irony in the narrative. Sometimes, by reference to the context of the sequence under consideration, the viewer will probably be able to decide whether dramatic irony occurs or not.

There are also some other meta-functions served by FID in film, as well. For Pasolini, FID is a stylistic device for foregrounding narrative discourse (551-52): since in the cinema the medium is composed primarily of concrete images, the filmmaker (= FCD), in order to differentiate his/her “look” from that of the character-focalizer, must have recourse to such stylistic operations as FID. One can conclude that it has the same nature and function as the literary FID, as postulated by Ann Banfield and her followers, which is not used in casual conversation or everyday texts. Therefore, FID is not part of our natural perception but a complex artistic technique.

Filmic narrative, in general terms, is a mixture of mimesis and diegesis, mimesis being the larger proportion, i.e., unlike the literary narrative, it mostly “shows” through images, as the main narrative components, rather than “telling.” However, sometimes, it may not be possible to distinguish mimesis (showing) from diegesis (telling), or the direct presentation of the story world and the characters’ physical / mental states—roughly equal to what André Gaudreault calls monstration (Burgoyne 117)—from the FCD’s manipulations, comments, evaluations, or interpretations. This happens exactly in the case of FID. Thus, one might confirm Genette’s idea that, in the last analysis, all mimesis merges into diegesis and vice versa (“Boundaries” 5). Genette draws this conclusion after having analyzed some examples of the literary narrative. However, to the present researchers, not only it also holds true for all other forms of narrative and, in particular, film narrative, but in film narratives this merging is most manifest.

According to Bordwell, “ambiguity” is a characteristic feature of art cinema. He believes that “a realistic aesthetics [= objectivity] and an expressionist aesthetics [= subjectivity] are hard to merge. [However,] the art cinema seeks to solve the problem in a sophisticated way: through ambiguity” (212). By “ambiguity,” he generally means those moments in a film when the spectator finds contrary clues for whether to attribute a specific shot or sequence to characters or to consider it as the FCD’s commentary or interpretation. Thus, his notion of ambiguity is the same as our concept of FID. He further explains that in such films “uncertainties persist but are understood as such, as obvious uncertainties” (ibid.). Put simply, such indeterminacies results in various and, sometimes, contradictory interpretations on the viewers’ parts, and this maximizes the film’s value, making the text “writerly,” to use Roland Barthes’s term. Nevertheless, the present researchers do not concur with Bordwell’s view that realistic objectivity and expressionistic subjectivity lead to artistic ambiguity because they are “mutually exclusive” (ibid.); rather, the researchers prefer to treat these two categories in Derridean terms, i.e., his paradoxical logic. Then, objectivity and subjectivity, or the narrator’s and the character’s voices, supplement rather than oust or negate each other. It should also be mentioned that Bordwell’s “ambiguity,” in what we called FID in film, is close to Branigan’s idea of “indeterminacy,” which is a more neutral term. Branigan believes that FID is not ambiguous or vague in the negative sense of the word but only “implicit and indefinite” or “indeterminate”: it is “merely composite and open in a specific way,” and the viewer does not have to choose between the two possibilities (objectivity and subjectivity), nor is s/he perplexed by such a narrative (Narrative Comprehension 170). The result is that a competent spectator will have no considerable difficulty in comprehending narratives containing FID.

Following Pasolini, Deleuze believes that modern cinema, what Pasolini names “cinema of poetry,” is mainly composed of FID images, the free indirect vision or thinking of a “spiritual automaton” (Bogue 177-78). By this term, he roughly means what Walter Benjamin names “flâneur,” while talking about modernist literature: distracted, mummified, passive, wandering, reflective characters commonly observed in modernist fiction or art cinema. Moreover, Deleuze claims that these spiritual automata represented in modern films are only indirect signs of the FID of a much more general spiritual automaton which is the film itself (Bogue 178). Thus, FID, in this sense, is an index of modern cinema.

Another meta-function of FID proposed by Deleuze is that FID indicates the essence of “the Cogito” and “the Cogito of art,” in particular; that is, FID shows that in general “an empirical subject cannot be born into the world without simultaneously being reflected in a transcendental subject which thinks it and in which it thinks itself,” and that, accordingly, in art, “there is no subject which acts without another which watches it act, and which grasps it as acted, itself assuming the freedom of which it deprives the former” (Cinema 1 73). As Ahmadi succinctly explains, in the case of FID, the FCD does not follow the Cartesian rule of subjectivity (262): a human subject is not able to think about him-/herself and at the same time think in someone else’s place, but the cinematic narrator can simultaneously see what it is supposed to see itself and see the story world through a character’s eyes—a character who is him-/herself seen by the FCD.

Still another meta-function of FID put forward by Deleuze is what he states about the nature of narrative truth. According to Deleuze, in classical cinema, the objective images provided by the FCD and the subjective images of characters are clearly distinguishable. Conventionally, the FCD’s discourse is thought of as authentic, reliable, and truthful, while the character’s is conceived of as an unreliable discourse that may not necessarily be true. By contrast, in modern cinema, where FID images are dominant, it is not possible to separate subjectivity from objectivity, and consequently the story does not any longer represent “an ideal of the true” but turns into a “pseudo-story,” in which seeking truth or discerning subjectivity and objectivity is minimized (Cinema 2 147-49).


This essay provided for the first time a model for identifying, analyzing, and interpreting free indirect discourse (FID) in the cinema, considered in terms of the established dual-voice hypothesis. This model was based on the same discourse representation logic that has been applied to prose narratives. We discussed the possible textual markers of FID and its major functions and effects. The most important function of FID can be considered its leading to artistic ambiguity and increasing the artistic value of the “text.” It was also argued that there are no textual markers peculiar to FID in film, but all the devices and techniques employed to manage internal focalization, subjectivization, and stream of consciousness in the cinema can be taken to represent FID, too, whenever they create vocal ambiguity in the text. Moreover, almost all the authorial functions, textual functions, and meta-functions counted for FID in prose fiction could be similarly attributed to FID in film.

The general hypothesis of this research was that there seems to be no clear-cut distinction between the different modes of representing characters’ discourses in the cinema and that one can usually make a demarcation between DD and ID, but, most of the times, it is very difficult to distinguish FID from ID or FDD. Nevertheless, in the course of the study, it was revealed that the cinematic medium does have the full potential for establishing almost all modes of discourse representation, as well as internal focalization and, most importantly, such a semi-subjective mode as FID, although the language of cinema is substantially different from that of prose fiction.


Ahmadi, Babak. From Pictorial Signs to the Text: Toward a Semiotics of Visual Communication. Tehran: Markaz Publishers, 2007.

Bal, Mieke. Narratology: Introduction to the Theory of Narrative. 2nd ed. Toronto: U of Toronto P Incorporated, 1999.

Banfield, Ann. “Narrative Style and the Grammar of Direct and Indirect Speech.” Foundations of Language 10.1 (1973): 1-39.

Bogue, Ronald. Deleuze on Cinema. London: Routledge, 2003.

Bordwell, David. Narration in the Fiction Film. London: Routledge, 1997.

Branigan, Edward. Narrative Comprehension and Film. London: Routledge, 1998.

Branigan, Edward. Point of View in the Cinema: A Theory of Narration and Subjectivity in Classical Film. Translated into Persian by Majid Mohammadi. Tehran: Fârâbi Cinematic Foundation, 1997.

Burgoyne, Robert. “Film-Narratology.” In Stam, Robert, Robert Burgoyne, and Sandy Flitterman-Lewis. New Vocabularies in Film Semiotics: Structuralism, Post-structuralism and Beyond. London: Routledge, 2005. 70-124.

Colebrook, Claire. Gilles Deleuze. London: Routledge, 2002.

Deleuze, Gilles. Cinema 1: The Movement-Image. Trans. Hugh Tomlinson and Barbara Habberjam. Minneapolis: U of Minnesota P, 2003.

Deleuze, Gilles. Cinema 2: The Time-Image. Trans. Hugh Tomlinson and Robert Caleta. Minneapolis: U of Minnesota P, 1997.

Deleuze, Gilles, and Félix Guattari. A Thousand Plateaus: Capitalism and Schizophrenia. Trans. Brian Massumi. London: Continuum, 2004.

Forceville, Charles. “The Conspiracy in The Comfort of Strangers: Narration in the Novel and the Film.” Language and Literature 11.2 (2002): 119-35.

Fowler, Roger. Linguistic Criticism. 2nd ed. Oxford: Oxford UP, 1996.

Genette, Gérard. “Boundaries of Narrative.” Trans. Ann Levonas. New Literary History 8.1 (1976): Readers and Spectators: Some Views and Reviews: 1-13.

Genette, Gérard. Narrative Discourse: An Essay in Method. Trans. Jane E. Lewin. Ithaca: Cornell UP, 1983.

Ginsburg, Michal Peled. “Free Indirect Discourse: Theme and Narrative Voice in Flaubert, George Eliot, and Verga.” Diss. Yale University, 1977.

Hedges, Inez. “Form and Meaning in the French Film, II: Narration and Point of View.” The French Review 54.2 (1980): 288-98.

Jahn, Manfred. A Guide to Narratological Film Analysis. English Department, University of Cologne, 2003. <>

Jahn, Manfred. Narratology: A Guide to the Theory of Narrative. English Department, University of Cologne, 2005. <>

Marks, John. “Deleuze and Literature: Metaphor and Indirect Discourse.” Social Semiotics 7.2 (1997): 233-46.

McHale, Brian. “Free Indirect Discourse: A Survey of Recent Accounts.” Narrative Theory. Ed. Mieke Bal. London: Routledge, 2004. Vol. 1: 187-222.

McHale, Brian. “Speech Representation.” Handbook of Narratology. Ed. Peter Hühn, Wolf Schmid, Jörg Schönert, and John Pier. Berlin: Walter de Gruyter, 2009. 434-46.

Murphet, Julian. “Point of View.” In Fulton, Helen, et al. Narrative and Media. New York: Cambridge UP, 2005. 86-95.

Palmer, Alan. “Thought and Consciousness Representation (Literature).” The Routledge Encyclopedia of Narrative Theory. Ed. David Herman, Manfred Jahn, and Marie-Laure Ryan. London: Routledge, 2005. 602-607.

Pasolini, Pier Paolo. “The Cinema of Poetry.” Trans. Marianne de Vettimo and Jacques Bontemps. Movies and Methods. Ed. Bill Nichols. Berkeley: U of California P, 1976. Vol. 1: 542-58.

Rimmon-Kenan, Shlomith. Narrative Fiction: Contemporary Poetics. 2nd ed. London: Routledge, 2008.

Schwartz, Louis-Georges. “Typewriter: Free Indirect Discourse in Deleuze’s Cinema.” SubStance 34.3 (2005), Issue 108: French Cinema Studies, 1920s to the Present: 107-35.

Vološinov, V. N. Marxism and the Philosophy of Language. Trans. Ladislav Matejka and I. R. Titunik. Cambridge: Harvard UP, 1986.

 Mohammad Ghaffary is PhD Candidate in English Literature, Shiraz University, Shiraz, Iran.

Amir Ali Nojoumian is Associate Professor of English Literature and Literary Theory, Shahid Beheshti University, Tehran, Iran

Related Contents

Dramatizing Water: Performance, Anthropology, and the Transnational
views 2332
Kanta Kochhar-Lindgren, University of Washington, Bothell  Download PDF Version   Place: Athipatti, a fictional South Indian village Vellaisa...
What is Performance Studies?
views 4001
Richard Schechner, Tisch School of the Arts, New York University  Download PDF Version Because performance studies is so broad-ranging and open to ...
The Entangled Vocabulary of Performance
views 2250
Sruti Bala, University of Amsterdam, Netherlands  Download PDF Version This article attempts to map the concept of performance, in terms of its gene...
‘All the world’s a stage and I’m a genius in it’: Creative Benefits of Writers’ Identification with ...
views 1847
Claudia Chibici-Revneanu, ENES, UNAM León in Mexico Download PDF Version Abstract This paper focuses on the romantic notion of artistic genius and ...