The term “mediacy” was coined by Stanzel ( 1971: 6) and describes the fact that the story is mediated by the narrator’s discourse in one of two ways. Either the story is openly transmitted through a narrator who functions as a teller of the tale (“teller mode”) or the mediation is apparently occluded by a direct,im-mediate presentation of the story through the consciousness of a reflector (character). In the reflector mode, we seem to see the storyworld through the eyes of a character and there seems to be no narratoroperating as a mediator. Since the introduction of Stanzel’s term, the fact of a mediate presentation of the story has become a general foundation in structuralist narratology. In Genette, mediation is two-fold on the levels of the discourse (récit) and the narrator’s act of telling (narration) ( 1980: 27,  1988:13); Prince ( 2003: 58) defines narrative as always having a mediating narratorial level; and Chatman, who looks at film and non-verbal narratives like ballet, speaks of “narrative transmission” (1978: 22). In recent years, the emphasis on different media using narrative has resulted in the term mediation being applied to the way in which a story is told in film, drama, cartoons, ballet, music, pictures, hypertext narratives, and other genres and forms of narrative.
Narratives can be mediated by narrators who tell and comment on the story or through agents who merely think, feel, or perceive. Stanzel discriminates between teller- and reflector-characters, arguing that they are “mediators of [...] fictional events” ( 1984: 150). However, they mediate story material, i.e. event sequences, in different ways. Teller-characters narrate, inform, and comment as if they were transmitting a piece of news or a message. Reflector-characters, on the other hand, do not narrate or transmit. Rather, the reader perceives the action through the eyes of the reflector character, and this veiled mediacy produces what Stanzel calls “the illusion of immediacy” (141). For Genette, the so-called “narrating instance” ( 1980: 212) is the communicative act that initiates both the story and the narrative discourse that produces the story. More specifically, the narrating instance represents events and existents (story), and they are thereby mediated in a particular (verbal, visual, or audio-visual) sign system (narrative) ( 1988: 13). Chatman speaks of the process of “narrative transmission” as “the source or authority for the story” (1978: 22). For him, the process of narrative transmission centrally concerns the relationship between story time and discourse time as well as issues of voice and point of view. Chatman discriminates between “overt narrators,” who communicate directly to the reader, and “covert narrators,” who remain more or less hidden in the narrative’s discursive shadows (1990: 115). Fludernik argues that all narrative is built on the mediating function of consciousness, a complex “natural” category with several available cognitive frames to choose from. She integrates Stanzel’s mediacy into a more general cognitive model of narrative transmission based on “real-life” schemata. Teller-mode narratives are mediated by the consciousness of a narrator; reflector-mode narratives by the consciousness of a protagonist; and neutral narratives by the reader who “views” and constructs narrative experience (1996: 50).
Underlying the question of what constitutes narrative is the concept of mediacy. While most narrative theorists define narrative in terms of event sequences, Stanzel and Genette reject blanket uses of the term “narrative,” the latter defining narrative stricto sensu as a “verbal transmission” ( 1988: 16). In Stanzel’s account, drama and film are im-mediate renderings of story, while (verbal) narrative is a mediated representation—mediated by the discourse of a narrator (openly mediated) or a reflector (obliquely mediated by presenting an illusion of im-mediacy). In contrast, Chatman also considers plays, movies, and cartoons to be narrative because they present stories (1990: 117). For him, there are “diegetic” and “mimetic” forms of narrative; narratives can be told or shown. Finally, Fludernik’s redefinition of narrativity on the basis of experientiality, i.e. “the quasi-mimetic evocation of ‘real-life experience’” (1996: 12), and its mediation through consciousness allows her to open up the field of narrative inquiry not only to drama and film, but also to oral storytelling and some kinds of poetry.
Stanzel’s notion of mediacy has roots in the distinction between mimesis and haple diegesis in Plato’s Republic (cf. also Lubbock [1921: 62], Blackmur [1934: xvii–xviii], and Friedman [1955: 1161–65]). In Plato’s diegetic or “pure” mode, the poet “himself is the speaker and does not even attempt to suggest to us that anyone but himself is speaking.” In the mimetic mode, however, the poet “delivers a speech as if he were someone else.” According to Plato, the poet may also combine these two modes and use the mixed mode, as in epic poetry (Plato 1937: 392c–95; cf. also Schaeffer & Vultur 2005: 309). Although Plato talks about speech representation (“pure” narrative and poetry vs. “pure” drama vs. narrative including dialogue insets), the Platonic mimesis/diegesis distinction as a dichotomy (rather than a triad) has been used to support both models of speech and thought representation (direct vs. free indirect speech) and the generic distinction between narrative and drama. Stanzel’s assignment of drama to the pole of immediacy (i.e. unmediated representation of story) therefore aligns immediacy with mimesis and mediacy with diegesis in the Platonic sense (McHale → Speech Representation).
While for Plato (and later Stanzel) the term “diegetic” refers to narratorial discourse (i.e. the act of telling), Genette uses the term diégèse (adopted from Souriau 1951) to denote the fictional world of the characters ( 1980: 27 n. 2,  1988: 17–8). Genette’s term diégèse has many affinities with Aristotle’s notion of mimesis. For Aristotle, “pure” narratives and direct representations are two varieties of what he calls mimesis because both represent a world (2002: 1448a). Similarly, Genette’s notion of diégèse refers to the primary story level, specifically excluding the narratorial discourse which is constitutive of both Plato’s and (in his wake) Stanzel’s understandings of diegesis. For Genette, “the diégèse is [...] the universe in which the story takes place” ( 1988: 17). Despite this terminological disparity, however, Genette and Stanzel agree with regard to the constitutive narratorial mediation of narrative, even though for Genette this is achieved through the narrating instance. For him, the narrator’s speech act produces the story through the narrative discourse.
Stanzel’s concept of mediacy is directed against Spielhagen’s prescriptive demand for “objectivity,” i.e. immediacy of presentation ( 1967: 220). Stanzel seeks to counter the excessive demands of “neutralists” like Spielhagen, who argued that the narrator should remain completely invisible throughout the narrative and thus wished to see every trace of a narrator erased. Stanzel’s proposal is closely related to Friedemann’s argument that the presence of a narrator in prose writings is in no way inferior to immediacy in drama, since the narrator is evocative of actual experience of the world. According to Friedemann, it is the narrator “who evaluates, who is sensitively aware, who observes” ( 1965: 26), thus conveying an image of the world as s/he sees it, not as it is in a depersonalized objectivity.
From the beginning, Stanzel presents the concept of mediacy as the linchpin for a definition of the term “narrative,” and he puts forth a sophisticated argument for mediacy as a gradable concept ( 1971: 6). More specifically, he points out that mediacy is more or less foregrounded (as revealed by the presence or absence of comments by an authorial narrator), but its absence in the figural narrative situation is merely apparent. In the final version of his model, Stanzel revises the figural narrative situation by integrating it into the illusion of immediacy in order to constitute the reflector mode of narration, which is responsible for producing this illusion. In opposing the teller mode and the reflector mode, he significantly reformulates his original typology, dating from 1955, by instituting two basic types of mediacy: teller-mode and reflector-mode mediacy.
In this discussion, Stanzel proceeds from three pairs of oppositions arranged as scaled categories of person, perspective, and mode (mediacy). The first element of the narrative situation, person, is based on the relations between the narrator and the characters, and it ranges from identity (first-person reference) to non-identity (third-person reference) of the realms of existence of the narrator (Margolin → Narrator) and the characters (Jannidis → Character). Perspective directs the reader’s attention to the way in which s/he perceives the fictional world, extending from internal (perception located in the main character or within the events) to external (perception located at the periphery of the events) (Niederhoff → Perspective – Point of View). Finally, mode breaks down into “overt mediacy of narration [teller mode, J.A./M.F.]” and “covert [...] mediacy which produces the illusion of immediacy in the reader [reflector mode, J.A./M.F.]” (Stanzel  1984: 141).
Stanzel regards the three narrative situations (first-person, authorial, and figural) as descriptions of basic possibilities of theorizing narration as mediacy. He also introduces a dynamic analysis into narrative transmission by demonstrating that narrative situations do not span entire novels uniformly. In his remarks on narrative dynamization, he discusses narrative profile and narrative rhythm. Although this dynamization is defined as a dynamization “of the narrative situation,” i.e. a study of “the variations of the narrative situation during the course of the narrative process,” the subsequent analysis actually focuses on the “relation of the narrative parts, that is, to dialogue and dramatized scene; specifically [on] their purely quantitative ratio and their distribution” ( 1984: 63–7). Besides these proportions, the incidence of direct speech vs. indirect and free indirect speech and thought representation is also taken into account. The second term, narrative rhythm, concerns the distribution of narratorial emphasis in a specific novel and refers to the fact that in most novels, the narrator figure manifests him- or herself prominently at the beginning of the text and sometimes at the end, but then lapses into inactivity when the plot becomes exciting, resurfacing only at moments of narrative report, commentary, or description. The result of this configuration is a simultaneous “decrease in these authorial intrusions [which] parallels the increase of the hero’s ‘perspective solipsism’” ( 1984: 69).
Nevertheless, it must be noted that the introduction of the three axes (identity vs. non-identity of realms of existence; external vs. internal perspective, teller vs. reflector modes) and emphasis on the dynamization of the narrative situation tend to foreground “mode” (i.e. the distinction between tellers and reflectors) and to background “person” (Cohn 1981: 168). Cohn additionally points out that Stanzel’s category of perspective merges the “presentation of space (the visible outer world)” into the “presentation of consciousness (the invisible inner world)” (175). And since perspectives on fictional space and fictional minds do not always coincide (Uspenskij 1973: 105–07), Cohn considers this axis to be less unified than the other two (cf. also Cohn 1990). She therefore proposes to simplify Stanzel’s typological circle by subsuming the category of perspective under the heading of mode (1981: 179).
Genette considers Stanzel’s category of mode to be superfluous, as he finds it “easily reducible to our common category of perspective” ( 1988: 116). In his view, Stanzel’s distinction between teller- and reflector-characters confuses the question of voice, or, more precisely, person (“who speaks?”) with that of mood or, more precisely, perspective (“who sees?”). He thus revises Cohn’s amendment of Stanzel by proposing a different taxonomy which “diversifies an initial typology that was [...] altogether too limited to the most frequent situations” (119). Genette’s model is based on the cross-tabulation of heterodiegetic and homodiegetic forms of narrating (“who speaks?”) and the three types of focalization (zero, internal, external) (“who sees?”) (21;  1980: 189–94, 245). Genette considers this taxonomy to be an improvement because it is more systematic and includes less common narrative forms such as Hemingway’s “The Killers,” a form of heterodiegetic narration with external focalization (the neutral subtype in Stanzel ( 1971:93), and Camus’s L’Étranger, a form of homodiegetic narration with external focalization.
Stanzel’s mediacy is equivalent to what Genette calls “narrating act” and “narrative.” More specifically, Genette discriminates between “story (the totality of the narrated events), narrative (the discourse, oral or written, that narrates them), and narrating (the real or fictive act that produces that discourse—in other words, the very fact of recounting)” ( 1988: 13). In this model, the narrating act shapes and transforms the story through the narrative discourse. Similarly, Rimmon-Kenan uses the terms story, text, and narration ( 2002: 3), while Bal modifies Genette’s terminology by arguing that it is by way of the text that the reader has access to the story, of which the fabula is a memorial trace that remains with the reader after the reading ( 1997: 5).
When Chatman introduced the principle of “narrative transmission,” he discriminated between “overt narrators,” “covert narrators,” and forms of “non-narration” for neutral narratives (1978: 22). Later, Chatman rejects the idea of non-narration by arguing that “every narrative is by definition narrated—that is, narratively presented” (1990: 115), but he maintains the distinction between overt and covert narrators, equivalent to Stanzel’s mediacy. His model is in close agreement with Stanzel’s, except that he includes drama and film among the narrative genres and therefore does not reduce narrative transmission or mediacy to the discourse of a narrative voice. Chatman provides a sliding scale from overt to covert narrators based on the linguistic markers of subjectivity, the presence of narratorial comments, and the use of evaluative phrases. Like Stanzel and Genette, he argues that all narratives have a narrator, so that all three theorists clearly oppose the Banfieldian “no-narrator” theory (1982), according to which certain sentences of fiction cannot possibly be enunciated by a narrator. Chatman argues that “narrative presentation entails an agent,” even when “the agent bears no signs of human personality” (1990: 115). The three authors agree that narratives always present a story which is mediated by a narrator’s discourse. Furthermore, Chatman stresses the conjunction of story and mediatory discourse by pointing out that “narrative entails movement through time not only ‘externally’ (the duration of the presentation of the novel, film, play), but also ‘internally’ (the duration of the sequence of events that constitute the plot)” (9).
It is quite apparent that Stanzel’s teller mode corresponds to Chatman’s scale which ranges from overt to covert narration (i.e. from subjective and foregrounded tellers to “objective,” neutral, and backgrounded narration). By contrast, with regard to Stanzel’s reflector-mode narrative, in which an illusion of immediacy is projected, Chatman (1978: 198) argues that a covert narrator expresses the thoughts of a character, while Genette ( 1988: 115) describes such a scenario as heterodiegetic narration with internal focalization. What the two terminologies fail to take into account, however, is the prototypical absence of a foregrounded narrator in reflector-mode narratives or, to put it differently, the fact that in order to read an extended passage as internal focalization, a pronounced teller must not interfere because such a foregrounded narrative voice would impede a reading of the text from the character’s perspective. Stanzel shows that Modernist novels (e.g. Joyce’s A Portrait of the Artist as a Young Man) establish a representation of the narrative world which is (or seems to be) filtered through the consciousness of the protagonist (cf. also James  1934: 322–25). This effect can only be achieved by completely backgrounding the narrative voice reporting on external events (for a critique of this claim, see Schmid 1968). By distinguishing between a teller and a reflector mode, however, the mere reduction of the narratorial voice to a default existence is not sufficient to characterize the reflector mode, since it is equally necessary to have a predominant internal perspective to produce the relevant effect. The reflector mode as mode only makes sense theoretically when one conceives of a different type of transmission through the character’s perspective or consciousness in contrast to the prominent (first- or third-person) teller-mode narrative which is mediated by an explicit transmitter.
Schmid (1982) puts forth an alternative model of narrative mediation by breaking down the story vs. discourse dichotomy into four terms: Geschehen (events); Geschichte (fabula or story); Erzählung (plot); Präsentation der Erzählung (narrative discourse). He goes on to posit three processes of transformation between these levels, all of which are accomplished by the narrator. According to Schmid, the mediating narrator first selects particular situations, characters, events, and qualities from the invented story material and transforms them into a story. The narrator then transforms the story into a narrative plot, going through a process that correlates with the linearization of simultaneous event sequences and the permutation of chronological story segments. And finally, the narrator presents the narrative by verbalizing it in a particular style. However, as Cohn argues, fictional narratives do not typically transform something pre-existent into a narrative, and they are thus plotted rather than emplotted (1990: 781). It is therefore worth noting that Schmid assumes an ideal-genetic perspective: the invented story material logically precedes the presentation of the narrative.
Fludernik (1996) takes Stanzel’s concept of mediacy further by locating all mediation in narrative transmission through consciousness (which can surface on several levels and in different shapes). For her, all narratives operate through the projection of consciousness—the character’s, that of the narrative voice, or the reader’s. She also departs from the general tendency to identify narrativity (Abbott → Narrativity) with the presence of a story/plot transmitted in narrative discourse. While most narrative theorists define narrative through sequentiality or progression, Fludernik argues that there can be narratives without plot, but there cannot be narratives without a human experiencer of some sort at some narrative level. She redefines narrativity in terms of experientiality, with embodiment constituting the most basic feature of experientiality: embodiment evokes all the parameters of a real-life schema of existence which has to be situated in a specific time and space frame. In addition, she broadens the analysis to include a wide variety of narratives, following on from Chatman (1978: 96, 1990: 115) and Bal ( 1997: 5).
Fludernik proposes to expand the ways in which narrative transmission occurs, arguing that all mediacy (or mediation) occurs through cognitive schemata (Emmott & Alexander → Schemata) and that what is being mediated is not primarily a story (although in the vast majority of narratives such a series of events does indeed occur), but experientiality, a conjunction of reportability and point (Baroni → Tellability). “Reportability” characterizes the interest which tellers and listeners entertain in narratives while “point” refers to the motivations for telling the story. Since experience is closely associated with actions, event sequences underlie experientiality, with suspense fulfilling a prominent role. Other emotions or thoughts may be foregrounded, however, and some narratives (though few) actually operate without plot. Beckett’s short prose work “Ping” is an example of a plotless narrative. In this text, a disembodied voice presents us with repeated descriptions of the same strange world which is somewhat reminiscent of a prison scenario. The only thing we learn is that a body is trapped in a small, white container. This prose work lacks events, but it clearly depicts consciousness and might be read as the agonized ruminations of the body’s mind struggling with some kind of traumatic experience (Alber 2002).
Mediacy is constituted by the following cognitive frames or schemata, all of which relate to our real-world knowledge (about telling, experiencing, viewing/observing, and reflecting) and provide us with access to the narrative: (a) the “telling” frame (narratives focusing on a teller figure); (b) the “experiencing” frame (narratives roughly corresponding to reflector-mode narratives); (c) the “viewing” frame (this frame occurs less frequently than (a) or (b), but relies on a basic witness position in relation to observed events); (d) the “reflecting” frame (when narratives project a ruminating consciousness). Consciousness mediates these frames in the reading process in which readers narrativize what they read as narrative, resorting to these four schemata but also to generic concepts and narratological tools as well as basic real-world knowledge (such as our understanding of intentionality as a goal-oriented process) which is also stored in scripts and frames (Fludernik 1996: 12–52). On this basis, natural narratology moves away from the idea of the narrator or the illusion of narration to a wider spectrum of cognitive frames and processes on different levels which feed into the constitution of narrative and its reception. Like all cognitive approaches, this model is grounded in the real-world frames of everyday experience and is reader- rather than production-oriented (Alber 2005).
The question of mediacy in narrative fiction has also been examined by Walsh, who argues quite provocatively that “the narrator is always either a character who narrates, or the author” (2007: 78). For him, “extradiegetic heterodiegetic narrators […], who cannot be represented without thereby being rendered homodiegetic or intradiegetic, are in no way distinguishable from authors” (84). Walsh suggests eradicating both “impersonal” and “authorial” narrators. While the first case aligns with Stanzel’s illusion of immediacy, the second differs radically from Stanzel’s distinction between authors and authorial narrators. Walsh maintains that the only way to account for the knowledge of an authorial narrator would be to take quite literally the figurative concept of omniscient narration: “in order to know rather than imagine, the (evidently superhuman) agent of narration must indeed have such power, or some lesser or intermittent version of it” (73). Thus, omniscience is not a faculty possessed by a certain class of narrators, but a quality of the author’s imagination. While some theoreticians infer from this an implied author (Schmid → Implied Author) (“an ideal, literary, created version of the real man” (Booth  1983: 75) as the mediating agent of narrative, Walsh speaks of “the author,” stating that “our idea of the author of a written narrative is no more than an interpretation” (2007: 84). Two things are worth noting here. First, the difference between Booth’s implied author and Walsh’s interpretation of the author is of course minimal or non-existent. Second, why should it be problematic to argue that third-person narrators can occasionally have “supernatural” (Ryan 1991: 67) or “unnatural” (Cohn 1999: 106) powers?
As pointed out in Nünning & Nünning (2002) and Wolf (2002), the definition of narrativity in reference to experientiality and the extension of mediacy to include an open list of cognitive frames, scripts, and schemata lead in the direction of transmedial and transgeneric narratology, as proposed in Fludernik (1996; Hühn & Sommer → Narration in Poetry and Drama; Ryan → Narration in Various Media). Many forays have recently been made into the area of narratological approaches to film, hypertext narrative, ballet, comic strips, drama, poetry, even painting and music (Ryan 2006, ed. 2004; Wolf 2002, 2003a, 2003b, 2004; Nünning & Nünning 2002). In this area, Chatman (1978, 1990) was an important innovator, for it was he who staked out a place for film in narratology (Kuhn & Schmidt → Narration in Film) and who also confronted narrative with other text-types, putting the concept of narrative under a new light.
Chatman sees narrative transmission as media-related, and he therefore dissociates narrativity from the figure of a human narrator (1990: 116; cf. Ryan 2001, 2006). Although he reintroduces a so-called “cinematic narrator” for film, this figure is not a human or human-like narrator as in novels. Rather, the term denotes “the organizational and sending agency” (1990: 127) behind the film and fulfills a neutral or covert shower or arranger function. The notion is similar to what Jahn calls the “filmic composition device (FCD),”which refers to “the theoretical agency behind a film’s organization and arrangement” (2003: F4.1). Even so, the question of who (or what) mediates a film as a whole remains highly disputed. Bordwell, for one, argues that film has narration but no narrator, and that consequently cinematic narration is created by the viewer (1985: 61). On the other hand, Lothe (like Chatman) posits a cinematic or film narrator as “the superordinate ‘instance’ that presents all the means of communication that film has at its disposal” (2000: 30). And finally, theoreticians such as Gaut speak of an “implied filmmaker” who mediates the film (2004: 248). From the perspective of natural narratology, one can alternatively argue that film resorts more generally to the “viewing” frame than to the “telling,” “reflecting,” or “experiencing” frame.
Like experimental literary narratives (Alber 2009), new media such as hypertext narratives or computer games require the introduction of new cognitive frames into the model proposed by Fludernik. From this perspective, mediacy does not refer to mediating through a (narrator’s) discourse, but mediation through consciousness. More specifically, we can gain access to these new media through the identification of consciousness. The verbal medium of a teller/narrator is only one possibility among many others; cognitive frames such as viewing, observing, experiencing, and reflecting (and maybe others) also play an important role.
However, some of the media that have come into focus since the turn towards transmedial narratology are hard to analyze on the basis of narratological categories. As shown by Wolf (2002), paintings and music can only occasionally be narrativized. These aesthetic products lack crucial elements of experientiality in what they are able to represent (most types of music are perhaps not able to represent anything at all). With poetry, the situation is more vexed. On the one hand, there is narrative poetry (the epic, the ballad), a genre much neglected by narrative theory. On the other hand, many lyric poems exist that are also readable as narratives or contain narrative elements (Fludernik 1996: 304–10; Hühn 2002, 2005; Hühn & Schönert 2002; Müller-Zettelmann 2002, 2011; Schönert et al. 2007). All types of poetry (narrative and lyric) are mediated by a speaker. The lyric persona also clearly operates as a mediator on the “reflecting” frame. However, this does of course not turn lyric poetry into a narrative genre. Lyric poetry does not typically evoke experientiality, i.e. temporal and spatial parameters, and thus lacks the situatedness of narrative. In prototypical cases of lyric poetry, we are confronted with the musings of a disembodied voice about feelings or abstract ideas.
Drama has long been a neglected object of narratological analysis. Drama was the focus not only of Aristotle’s discussion of mimesis and has thus become a subtext of all narrative theory, but like epic forms it is closely bound up with sequentiality and thus invites narratological analysis. Hence, Pfister (1977) undertakes a narrative analysis of drama, studying the relationship between story time and discourse time. Since then, Richardson (1987, 1988, 1991, 2006), Fludernik (1996, 2008), Jahn (2001), and Nünning & Sommer (2002, 2008) have started to focus on drama and its relation to narrative. Much of this work analyzes elements in drama which have to do with mediacy such as the introduction of teller figures (the Stage Manager in Wilder’s Our Town), first-person narrators (Henry Carr in Stoppard’s dream play Travesties), or the fictionalizing of stage directions to include psychonarration, puns, or authorial commentary (Fludernik 2008). For the present purpose, these impositions of a teller figure on the plot level, the introduction of an extradiegetic frame into the play, or the narrativization of stage directions are not really relevant due to the fact that the mediacy of drama is constituted by other factors. Plays partake of the same stock of cognitive parameters and depend on the same reception frames as do other narratives. Since plays represent experientiality, they are narrative, irrespective of narrator figures or additional narrative techniques (such as the use of music). In other words, having a narrating character on stage, for example, is not required to bring plays within the domain of narrative.
From this perspective, a problem very similar to that of film arises: what is the discourse level of drama? Here, the dramatic performance needs to be distinguished from the dramatic text (Berns → Performativity) (cf. also Jahn 2001: 675). Does one treat only performances as drama in which performance is the discourse and the script merely the plot with instructions on how to perform? Or is performance a separate manifestation of the play and the play script the equivalent of the dramatic discourse? If one takes the text as central, it could be argued that an idealized abstract performance is sketched in it and that a unique center of origin can be posited for the performance: the text underwrites a singular “meaning” of the play that one might associate with “the implied author,” i.e. the real author’s “second self,” which, according to Booth, satisfies “the reader’s need to know where, in the world of values, he stands, that is, to know where the author wants him to stand” ( 1983: 73). By contrast, if the performance is to be taken as the only acceptable discourse, there results a collaborative venture—as in film—for which the term “dramatic composition device,” in analogy with Jahn’s “filmic composition device” (2003: F4.1), might be appropriate. Most crucially, assuming performance to be the basic medium of drama requires taking account of the acoustic, visual, kinetic, and spatial aspects of a performance within narratological description. Jahn in fact argues that plays “are structurally mediated by a first-degree narrative agency which, in a performance may either take the totally unmetaphorical shape of a vocally and bodily present narrator figure [...] or remain an anonymous and impersonal narrative function in charge of selection, arrangement, and focalization” (2001: 674). This suggestion is of course reminiscent of Chatman’s distinction between overt and covert narrators. If only the script and a possible performative realization are focused on as the relevant medium of drama, then kinesis, lighting, and sound would acquire narratological significance only if they are explicitly grounded in the script. The performance level in drama is much more complicated than in film. Filming results in one fixed copy of the narrative, whereas with plays a variety of productions and different performances within each production occur, and none of them (unless videotaped) is accessible except in a viewer’s experience of watching the performance.
It is obvious from these remarks that playscripts are much easier to handle for narratologists and that they allow a much clearer idea of how story and discourse are related to one another. Performance poses quite difficult problems for mediacy. In fact, one could enquire whether the notion of mediacy might here be an exclusively reception-oriented one. Is the story mediated to the audience through the experience of the performance? This question indicates that current research on mediacy has some distinct limits or horizons and that there are numerous matters waiting to be resolved by further research.
(a) The role of mediacy in drama and film remains open to study: does it make sense to posit a dramatic or cinematic narrator? Can one argue that they are mediated by the performance? Or should we assume that plays and films are mediated by an implied author or filmmaker? Or are all of these terms dispensable so that we can simply speak of the author or filmmaker (a larger group of professionals) as mediating instances (see also Alber 2010)? (b) One should also address the question of whether we can follow Walsh’s proposal to dispense with all extra- and heterodiegetic narrators in novels and short stories. In most cases, it certainly makes sense to discriminate between the author and the authorial or impersonal narrator. (c) It is also necessary to investigate the development of new cognitive frames of mediation in relation to experimental literary narratives and new media (hypertext narratives and computer games).