The Spectacle of Interruption: Toward an Interruption Theory of Narrative according to Hölderlin’s Theory of Tragedy
University of Tübingen
The tacitly understood assumption that the truth about and in narrative as a whole is the Hegelian or Idealist aspect of narratology, of which numerous semiotic, structuralist, system-theoretical stances take heed. They all conceive of the narrative utterance as a sort of organic whole. This organic approach to narrative derives, among other sources of inspiration, in a straight line from Vladimir Propp’s Morphology of the Folktale ( 1958) that drew heavily from Goethe’s study of nature and Goethe’s interpretation of Aristotle’s Poetics, which stressed the organic unity of the tragic story and its rendering in a successful tragedy (cf. Goethe  1960). The organicism of this tradition was upheld and altered by structuralism, whose legacy serves to this day as a basis for many current approaches in narratology. This legacy continues to make itself felt in recent narratological studies, of which a volume edited by Greta Olson entitled Current Trends in Narratology (2011) is a good illustration. The volume divides present-day developments of the discipline into three areas of interest:
1. cognitive narratology;
2. transgeneric and intermedial approaches that take into consideration above all drama; and finally
3. contextualist narratology which deals with specific cultural, historical, thematic and ideological contexts of narrative and its study.
Although the authors of this volume adhere to various methodological positions, they strive unanimously to surpass structuralism. Narratologists concerned with the representation of mind in narrative (and the representation of mind in narrative), or those who study the traces of narrative in drama or other media implicitly postulate, in line with the German morphological tradition, that the whole truth be told in and about (a) narrative. Their attempts to go beyond structuralism only reinforce the holistic pathos characteristic of structuralism. The task set by contextualist narratologists to overcome oppression, exclusion or exclusivism results in the restitution wholeness, the return of the subaltern. Technically speaking, newer approaches in narratology seek to supplement or enrich classical narratology with features of contextuality and diachronicity so as to capture narrative in its wholeness, both the narrator and the reader/listener in their full biological, cognitive, historical and medial makeup.
The thesis of this paper is that the holistic character of narrative as a structure becomes apparent and is, so to speak, attached to its temporality and contextuality by a special kind of interruption: the breaking up of the whole. The most interesting and prototypical theory of such interruption is the notion of caesura, whose most perfect theoretical description is found in Friedrich Hölderlin’s notes to his 1804 translations of Sophocles’ Oedipus and Antigone (Hölderlin  2001 and 1988). According to Hölderlin, caesura joins the whole man with time as the transcendental possibility of representation while also revealing to him the meaning and shape of the time he lives in. The contextual, temporal and holistic character of his cognition, dependent on the temporal medium of narrative, comes to the fore because of the caesura, provided it is experienced by the reader. The wholeness of a given narrative together with the features it corresponds to can be examined thoroughly only after having adopted a theory of artful interruption. The reflexive qualities of caesura which, for Hölderlin, seems to be a special device that lays bare the nature of temporal representation itself, are the reason why the very general notion of an interruption theory of narrative is set forth in this article. The caesura is characterized by its dual nature. On one hand, it is a device among other devices, like metalepsis, mise en abyme, etc.; on the other hand, it plays the most central role in Hölderlin’s philosophy of art, becoming the keystone and the touchstone of this theoretical edifice. The reader must experience caesura if he is to be able to experience the temporality of time, the dimensions of space and the medial representation of representation.
The present attempt to specify a position for such an interruption theory within the framework of current narratology will begin with an explanation of Hölderlin’s definition of caesura. During the second step, I will expand on the relevance of Hölderlin’s theory for present-day narrative studies. His theory shares some of the same assumptions as present-day narratology in its opposition to reductionist approaches. This development can be seen particularly in the expansion of narratological research into the field of drama studies. In the following step, the paper will take up two competing paradigms of narrativity within the framework of the early nineteenth-century theory of drama: one that concentrates on endings will be opposed to one, represented by Hölderlin, that treats interruption as a central component. The latter paradigm, whose renewal is the task of the present paper, was to a great degree responsible for the rise of modern narratology in Eastern and Central Europe before World War II, although since then this has remained unrecognized. As a last step, I propose to compare the theory of caesura with that of metalepsis, for each serves to throw light on the other.
1. Hölderlin’s definition of caesura
Much as the classical narratologists Tzvetan Todorov or Algirdas Greimas, Hölderlin assumes that the beginning and the end of a narrative is equilibrium: narrative aims toward a state of equilibrium after passing through a momentary disequilibrium that sets the story going (cf. Todorov  1973: 82). Sophoclean tragedy is, according to Hölderlin, a model for all narrative forms for the reason that, as a genre, it presents nothing in particular except its own form of development in time, a development toward equilibrium. According to Hölderlin, the tragic form was designed to abolish every particular piece of content (or to present the content as abolished; Hölderlin  2001: 67; 1988: 257) so that the pure form of temporality will be highlighted. This occurs as follows:
the transport in tragedy is of itself empty and the most unbounded.
For this reason in the rhythmical succession of scenes in which the transport is made manifest, it becomes necessary to have what in the study of verse is known as caesura: the pure word, the counter-rhythmical interruption, is needed, so as to confront the pull of the succession of representations at its summum and in such a fashion that instead of change of the representation there appears representation itself.
In a narratological reading of the quoted passage from Hölderlin's commentary on Sophocles, caesura is a moment of division that places narrative representation as such on display. It may be a lexeme, like some “empty words” in Hölderlin’s hymns, an image, or a scene in a drama such as the example provided by Hölderlin: the entries of the priest Tiresias in Oedipus and in Antigone. Tiresias “enters the process of fate as an overseer of the natural order which, in tragedy, removes man from his own zone of life, from the midpoint of his own inner life, and carries him into the eccentric zone of the dead.” (Hölderlin  2001: 64).
Also, the tragic festival in Athens itself has been more often than not described as a caesura: a leisurely moment, an interruption that restores a cosmic and social balance so that things in the cosmos could go smoothly. Hölderlin put it in his typically cryptic way: “tragedy consists […] in the form that reason takes in the terrible interlude of a tragic age, which, having presented itself then, in its wild genesis, in antitheses, later, in a humane age, will count as an established opinion, born of a divine fate.” (Hölderlin  2001: 11). Tragedy that contains an interruption revealing the genre’s compositional pattern locates, in addition, the action at a time when chaos breaks into the world so that a new order is born out of a cosmic catastrophe. Sometimes the public interrupts such an interruptive fest in order to stress its purging effect, the cathartic character that the authorities seek to suppress for the benefit of the tragedy’s restorative and conservative functions. Thus the revolutionaries in Berlin in 1849 interrupted the performance of Hölderlin’s Antigone, exactly at the moment when, according to Hölderlin, caesura occurs in the drama, i.e. during Tiresias’s speech, just as the revolutionaries in Warsaw in 1968 did during the performance of Adam Mickiewicz’s tragedy Forefathers’ Eve. In both cases, the revolutionaries wanted to extract from the tragic plays what they were really about: freedom and balance in the shape of fairness and justice (Flashar 1991: 75; Ostrowska 59–66).
According to some authors, every narrative should, for structural reasons, be perceived as an interruption of some kind: “traditional narrative is a quest after that which will end questing; [...] it is an interruption of what will be resumed,” writes J. Hillis Miller (1981: 272).
2. The Topicality of Hölderlin’s Theory
Turning to Hölderlin means returning to sources of modern narratology that abound in the German theory of drama during the Goethe period ( Goethezeit). The problems and goals of present-day narratology echo the questions raised by the aesthetics of literature at the turn of the nineteenth century. This can be seen in current attempts, for instance, to extend narratological research methods to drama. In fact, narratology’s progress toward drama should be even more resolute, since this represents a clear route to revealing its now forgotten conceptual foundations, which were originally elaborated drawing on the example of ancient Greek tragedy.
As for the common ground shared by the classical German theory of drama and modern narratology, it suffices to put a number of principles characteristic of current developments in narratology alongside programmatic statements by famous German philosophical voices from the turn of the nineteenth century. David Herman, in a series of recent studies, considers that classical narratology, which strictly separated the living man from the purely textual and conventional presentation of the human in text, can be enriched by taking account of the full range of faculties of the living human being:
narrative analysts can move from classical theories of narrative perspective toward a unified account of construal or conceptualization processes and their reflexes in narrative. Such construal operations, which underlie the organization of narrative discourse, are shaped not just by factors bearing on perspective or viewpoint, but also by temporal, spatial, affective, and other factors associated with embodied human experience. (Herman 2009: 103; see also Herman 2007: 245)
In Herman’s arguments against textualist reductionism and in favor of “embodied human experience,” there resonates an echo of old complaints against Kant and Enlightenment intellectualism, first raised by Herder and his teacher Hamann, whose approach was rendered by Goethe in the following way:
The principle to which all Hamann’s expressions may be referred is this: “All that man undertakes, whether by deed, by word, or otherwise, must proceed from all his powers united; everything isolated is worthless.” A noble maxim, but hard to follow. (Goethe  2013: 446–447)
The common opinion of the time, shared also by Hölderlin when he wrote Hyperion and by the authors of the so-called Earliest System Program of German Idealism, was that forms of the beautiful serve to ward off or prevent the feared isolation of human faculties. Today’s theories, drawing from the newest research, are driven by the same pathos that inspired the philosophers and poets at the turn of the nineteenth century. In present day narratology, the artful interplay of forces tending toward a fragile equilibrium restores Schiller’s well-known ideal of ganzer Mensch, the whole man. The following formulation by Schiller, expressing the uniting power of the beautiful form, must have influenced the Russian formalists, whether they were conscious of it or not, and thus modern narratology:
In a really beautiful work of art, the content ought to be inoperative, the form should do everything; for by the form the whole man is acted on; the content acts on nothing but isolated forces. Thus, however vast and sublime it may be, the content always exercises a restrictive action on the mind, and true aesthetic liberty can only be expected from the form. (Schiller  1990: 70)
Hölderlin ascribed the uniting force to the rhythm of poetry, which is more formal than anything else in poetry, and whose development in time is ruled by what he calls “the calculable law”:
The rule is one of various sequences in which imagination and feeling and reasoning develop according to poetic logic. For whereas philosophy only ever treats one of the soul’s capacities, so that the presentation of this one capacity makes up a whole and the mere hanging together of the parts of this one capacity is called logic, poetry treats the various capacities of the human being so that the presentation of these various capacities make up a whole, and the hanging together of the – more autonomous – parts of these different capacities may be called rhythm (in a higher sense) or the calculable law. (Hölderlin  2001: 113)
The main part of this calculable law, the one owing to which calculus balances out and embraces the whole man, is caesura, a counter-rhythmical interruption that may operate on all levels of literary work of art. As already indicated above, caesura may occur in a lexeme, an image or a scene in a drama, as in the entries of the priest Tiresias in Oedipus and in Antigone. Paradoxical as it may sound, interruption works against the isolation of the faculties of man and his alienation from temporal and spatial specificity. For Hölderlin, the whole man, with all his powers, should feel and understand the temporal nature of his world and his understanding: that is what the narrative form of tragedy, sealed with caesura, dictates. Thanks to the tragic form of development in time, history regains its Greek meaning of “experience” (Koselleck 2003: 20). The same should be said about the histoire that is mediated by the performative discours in narrative: it becomes a vital part of experience, in which the whole historical and sensual man takes part within the context of the time and space in which he lives and which, thanks to narrative, he feels.
The recent expansion of narratology into the field of drama (cf. Hühn and Sommer 2012) can be seen as something of an oedipal pathos when it is realized that the conquest is actually an unconscious homecoming. In Hölderlin’s time, the problems of histoire were elaborated within the framework of the theory of drama, for the theory of prosaic discours was elaborated in the scope of the rhetorical elocutio whose relationship with fictive discourse has always been problematic. When Manfred Pfister ( 1988), Ansgar Nünning and Roy Sommer (2002, 2008), Monika Fludernik (2008) and Brian Richardson (2007) ponder whether the tools of narratology can be applied to dramatic works in the face of their apparent lack of an extradiegetic narrative level, they tend to overlook not only the historical origin of most of their tools but also the simple truth that, although perhaps most plays are devoid of a primary narrator who relates the story, narrative discourse as such is to a degree a drama.
The dramatic aspects of narrative were explicitly stated by the Russian scholars who laid the foundations for modern narratology. Boris Ėjchenbaum ( 1989), for example, stressed the theatricality of the skaz narrator who concentrates on his gestures and his play rather than on the plot. Viktor Šklovskij ( 1966: 114) noted with amusement that he begins to imitate the devices of Cervantes’s narrator instead of describing them. Yurij Tynjanov ( 1957: 418, 428) describes how Gogol uses “the device of the mask” in his descriptions of people and demonstrates how Dostoevskij uses Gogol’s “verbal” and “material” masks in order to cover characters that are opposite to those of Gogol. Viktor Vinogradov (1936: 131) spoke about the faces and antics that make up the author’s image throughout the narrative (cf. Do Hay Fong 2012). Mikhail Bakhtin ( 2010: 12–21) began his (re)construction of the carnivalesque that stands for the specificity of the novel as a genre with the distinguishing of its ritual-scenic components.
The recognition of the dramatic in narrative can also be found in Greimas’s structural semantics where he adopts Lucien Tesnière’s model of the sentence as a spectacle homo locuens plays for himself (cf. Ricœur 1984: 72) as well as in Paul Ricœur’s description of emplotment (mise en intrigue) as a reflexive judgment in the course of which the subject stands back in order to observe his own plotting (Ricœur 1984: 92–94). Thus the subject of mise en intrigue becomes a mask, and his activity a spectacle for him and for his readers. In narrative, the author (or his stand-in), present and absent at the same time as though he were both dead and alive, impersonates a narrator who speaks out or writes down characters’ discourse, playing the parts of the protagonists who, in turn, may become narrating actors themselves. As early as the 1970s, Kazimierz Bartoszyński (1971), a representative of the Polish School of Literary Communication, called narrative a “spektakl sytuacji komunikacyjnej,” that is “a (dramatic/theatric) performance of the communicative situation” that actualizes the “roles of emitters and receivers” inscribed in narrative. The teller, however, is not the only quasi-theatrical phenomenon inscribed within the narrative text understood as a script for the participant in the spectacle of literary communication. Bartorszyński’s colleagues, Michał Głowiński (1968) and Edward Balcerzan (1971), stress that every empirical reader assumes “this or that role” such as it is pre-programmed by an author. Some fifty years earlier, in 1924, Jurij Tynjanov offered a vivid description of a reader before a skaz: he enters the narrative, begins to repeat the intonation and to gesticulate, to smile. He does not read the skaz, he plays it (Schmid 2010: 131). In view of these and analogous factors, Polish scholars have generalized the theatrical to cover narrative. Narrative, like theater, must partake of both roles, sender and receiver, in order for literary communication to take place, regardless of the type of narrative.
Bartoszyński claims the performance inscribed in narrative becomes perceptible and communicable (contagious for the reader) against the backdrop of stereotypes with which every narrative has to measure itself in order to become intelligible and, in some cases, aesthetically appealing. For him, stereotypical frames are pretty much synonymous with social frames of memory, as they were described by Maurice Halbwachs ( 1952). The theatric performance of narration is a negotiation with collective memory, and in Hölderlin it is precisely the theatrical character of narrating and the mediating role of memory (and temporality) that the interruption triggered by caesura exhibits and stages. This kind of interruption reveals how forms of time and forms of memory are interdependent with narrating performances. This Polish tradition turns out to be especially relevant for the conceptualization of caesura in relation to other similar devices such as mise en abyme and metalepsis. Before I expand on the subject, a short review of the suppressed tradition of theory of narrative based on interruption is in order.
3. Endings vs. interruptions
Paul Ricœur once stated that in the western tradition the paradigms of composition are at once the paradigms of terminating the narrative work (1984: 35). In accordance with the prevailing tradition of literary studies since Aristotle (1449b; cf. Aristotle 1922: 22–23), who claims that the praxis imitated by the plot must be uniform and finished (completed, perfect, τέλειος), present-day narratology works rather on endings than on interruptions, providing even more evidence that the theory of narrative stems from the theory of drama. For the sake of a happy or an unhappy ending, the greatest generic division of all time was made, between the comic and the tragic.
Modern narratology does not really distinguish between good and bad endings, but rather between endings that fall on a stable position in the development of the story (D.A. Miller 1989; Smith 1968; Reisling  2002; Kermode 1967) or on endings that fall on an unstable position (the Russian formalists; Vygotskij  1986: 201; Kucich 1978). Those scholars who are as generous and erudite as Ricœur take into account both types of endings (1984: 38–39); those who abide by deconstructionist methods, like J. Hillis Miller (1978), consider an ending both impossible and possible. Among the advocates of stability as a goal of every narrative, who presently outnumber their opponents, Smith claims that a stable state is attainable for a narrative, whereas D.A. Miller, Reisling and Kermode question the possibility of a narrative achieving a stable state, contending that there is no proper ending to a story. Miller attributes the impossibility of a narrative ending to the internal structure of narrative; Reisling inclines toward cultural and historical circumstances to be perceived as the agents of decomposition; Kermode, echoing Kenneth Burke and Northrop Frye, refers to transcendent considerations and preaches the permanence of the Apocalypse. Thus it is not surprising that when, in the present interest in endings that do away with interruption, a narratologist like Brian Richardson takes on drama, he focuses precisely on theorizing “Endings in Drama and Performance” (Richardson 2011).
It has not always been like this, however. At the turn of the nineteenth century, two competing paradigms of narrativity were negotiated within the framework of the theory of drama: the finalistic one we all adhere to because we do not know any better; and the one that treats interruption as a central component. The finalistic approaches were set forth both by the early German Romantics (above all by Friedrich Schlegel) and by the German Idealists, Schelling and Hegel. Both the Romantics and the Idealists concentrated on conflict and its tragic or comic solution as the essential components and conditions of narrative development. The focus of this line of inquiry was on the conflict, contest and the catastrophe. Schiller and Hölderlin argued in another way: Hölderlin, who thought he understood Schiller better than Schiller understood himself, stipulated interruption not only as a condition of the tragic but also as a condition of all narrative progress of representation. It is good to refresh our memory of the interruptive theory of narrative and recall that the origins of modern narratology in the writings of the Russian formalists go back to the interruptive paradigm of narrative studies. Viktor Šklovskij analyzed the disruptive devices in Tristram Shandy and Ėjzenštejn’s montage techniques to show how interruption lays bare the devices that make up the continuity of narration. Similarly, Puškin formed the plot of his novel in verse Yevgeniy Onegin in such a way that he plugged interrupting digressions into the narrative material of the novel (Šklovskij  1966: 135–140). The plotting of this novel boils down, in other words, to a series of interruptions.
Roughly contemporaneous with Šklovskij, Walter Benjamin eloquently demonstrated that if a literary work somehow lacks interruption, it is the task of the critic to rip open its beautiful surface in order to show its truth-content. Even the most basic critical forms, commentary or quotation, break up the continuity of a commented text (Benjamin  1990: 123–127;  1990: 207–209, 357–358; [1931a] 1990: 354–367). Throughout his career, from the early essay on Goethe’s Wahlverwandtschaften right up to Central Park and the theses on the notion of history, written with the premonition of his death, Benjamin kept producing new reformulations of Hölderlin’s caesura. Once it was das Ausdruckslose at the root of Goethe’s prose; another time it was an image in the sequence of Calderon’s drama, an image that “stands out, in the image of the apotheosis, as different in kind, and gives mourning at one and the same time the cue for its entry and its exit” (Benjamin  2003: 235). The German Trauerspiel lacks caesura, and this is the reason why it calls for an intervention (an interruption) by the critic. Still another time, in the late historiosophy, caesura became das dialektische Bild: a moment – Jetztzeit – knocked out of the continuum so that past and present may form a constellation in which the messianic future becomes comprehensible (Benjamin  1990: 691–705). Combined with Berthold Brecht’s theory of epic theater, according to which the tearing apart of the ideologically laden theatrical illusion has a revolutionary effect on the audience, caesura has become and remains up to the present day the agent of the truly political within art.
(Even Benjamin’s description of the ghostly body of a film actor, summoned and summed up by montage cutting, owes its emergence to Hölderlin’s description of Sophocles’ caesura: the entry of Tiresias who oversees future events and the shadows of past life.)
Writers have always been a few steps ahead of theoreticians. That interruption is an inalienable part of narration is attested by the metanovel par excellence, Tristram Shandy. In order for the story to begin in Sterne’s novel and for the hero to be conceived, his poor mother interrupts the coitus interruptus with Tristram’s father because she fears that the clock will stop.
“Pray, my dear,” quoth my mother, “have you not forgot to wind up the clock?” — “Good G—!” cried my father, making an exclamation, but taking care to moderate his voice at the same time, — “Did ever woman, since the creation of the world, interrupt a man with such a silly question?” (Sterne 1980: 2)
Through the loss of concentration on the part of the father, the most autodiegetic narrator ever slips into life. As Šklovskij wrote, foreshadowing the birth of modern narratology, the accentuation of form by means of its being incessantly destroyed makes up the significance of Sterne’s novel (Šklovskij  1966: 135).
Unlike in Benjamin, one of whose earliest essays was devoted to an interpretation of two poems by Hölderlin (Benjamin  1990: 105–126) and who belonged to the generation who venerated and rediscovered Hölderlin, tragic caesura surfaces most unexpectedly and in an unmediated fashion at the dawn of the modern novel, for which discourse is far more important than story, and, at the same time, at the outset of modern narrative enquiry into perception and mind: caesura surfaces abruptly in the introduction to The Ambassadors by Henry James. Caesura in the stream-of-consciousness novel is described in such a way that it almost seems that James copied parts of it from Hölderlin, even though he was not familiar with Hölderlin’s writings.
The situation involved is gathered up betimes, that is in the second chapter of Book Fifth, for the reader’s benefit, into as few words as possible – planted or “sunk,” stiffly and saliently, in the center of the current, almost perhaps to the obstruction of traffic. […] The whole case, in fine, is in Lambert Strether’s irrepressible outbreak to little Bilham on the Sunday afternoon in Gloriani’s garden, the candour with which he yields, for his young friend’s enlightenment, to the charming admonition of that crisis. The idea of the tale resides indeed in the very fact that an hour of such unprecedented ease should have been felt by him AS a crisis, and he is at pains to express it for us as neatly as we could desire. (James  2011: xxvii)
Meaning is revealed at a time that seems to have fallen out of time and obstructs its inescapable traffic. Hölderlin, like James, never misses a chance to stress the coincidence of leisure and crisis in tragedy that presents nothing but its own shockingly empty temporal form, a form that surpasses any particular content. “Live all you can; it’s a mistake not to,” cries Lambert Strether in The Ambassadors’ caesura. “It doesn’t so much matter what you do in particular so long as you have your life” (James  2011: xxviii).
4. Caesura vs. metalepsis
We should, for precision’s sake, set Hölderlin’s notion of caesura off from such kindred devices as narrative metalepsis andmise en abyme (cf. Genette  1980: 233; Dällenbach 1977; Cohn  2012; Pietrzak 2007; Pier 2014: par. 32–34). The task is especially difficult because, as Sonja Klimek (2010: 52–54) has clearly demonstrated, the devices of mise en abyme and metalepsis intersect and because numerous occurrences of mise en abyme lead to paradoxes concerning the distribution and division of narrative levels just as, conversely, there are instances of metalepsis that reflect a degree of similarity between the encompassing and the encompassed levels. Here, I would like to follow the classical definition proposed by Dällenbach – toute enclave entretenant une relation de similitude avec l’œuvre qui la contient (1977: 18, original emphasis) – and place mise en abyme in the category of similarity; metalepsis (defined by Genette as “any intrusion by the extradiegetic narrator or narrate into the diegetic universe […], or the inverse”;  1980: 234) will be considered within the framework of narrative levels rather than similarity.
Although mise en abyme may also, like caesura, be disruptive, it merely aims at the repetition or reflection of the elements of the embedded story in the embedding one. The experience of mise en abyme consists, then, in perceiving a similarity between narrative levels or instances whereas, as I will demonstrate in this chapter, caesura splits the emitting instances while interrupting the train of narrative and relates the instances to the very situation of communication. As such, mise en abyme does not replace but rather multiplies caesura, as in the example from James where, in an interruptive moment of leisure, the protagonist hints at the meaning of such a leisure – in life, but also, we assume, in narrative.
Caesura, on the other hand, is closely related to narrative metalepsis, since they both descend in a straight line from Fichte’s theory of knowledge, even though caesura seems to be a more profound and consequent application of transcendentalism to literature than its prominent sister device metalepsis. The term metalepsis itself comes from rhetoric, notably from Fontanier and Dumarsais (Genette  1980: 234–237; 2004: 7–20; cf. Pier 2014, § 3.1), while caesura comes from metrics. Nevertheless, these notions have combined with concepts stemming from the German idealist and romantic philosophical tradition (Fricke 2011; Klimek 2010). If metalepsis is understood as an intrusion of the world of the creator into the created world, or vice versa, then metalepsis is akin to Friedrich Schlegel’s concept of (romantic) irony. The romantics, when developing the theory of the possibility of permanent incursion of the teller into the told, took the name of Fichte’s ultimate source of all knowledge (Ich) for the grammatical personal pronoun referring to the subject of an utterance, and so transferred the almighty and absolutely free “I” to the literary field. Theirs was an anthropomimetic and language-oriented interpretation of Fichte’s theory of knowledge that is practically indiscernible from metalepsis as an intrusion of “I” into the utterance or simply the collision of the level of the narrative act – at which some “I” operates whose activity creates the presented word – with the intradiegetic level. In contrast to the egological romantics, Hölderlin focused on Fichte’s description of the imagination that mediates between the determined-determining “I” and the “not-I” in a series of sublime interruptions of the flow of time. Imagination discerns not only between the first and the third persons but also between the levels of the transcendental I or Ego. The highest one, pure reason, enters the small moments of the sublime that are torn in the flux and sees the temporality of representation which it brought into being:
This floating of imagination between irreconcilable links [the determined I and not-I within the absolute I, pure reason], this its self-contradiction is […] that which extends the condition of the Ego to a time-moment. (For pure reason everything is at once; only for imagination is there a time.) Imagination cannot stand this floating long, that is, not longer than a moment, (except in the feeling of the sublime, where astonishment, a halt of the interchange in time, arises.) Reason steps in, (and thus there arises a reflection,) determines imagination to take B up in the determined A, the subject; but as soon as this is accomplished, the determined A must again be limited by an infinite B, etc. etc., until it has arrived at a complete determination of the (here theoretical) reason, which needs no other limiting B outside of reason in imagination, that is, until it has arrived at the representation of the representing. (Fichte [1794/1795] 1889: 181–182)
The “representation of the representing” forms a common ground between metalepsis and caesura, on the backdrop of which the differences between the two figures rooted in Fichte become more evident. In contradistinction to Fichte who, as the last sentence suggests, aimed at a purely intellectual representation of the representing in the theory of knowledge, Hölderlin assumed that the truly absolute cannot be merely abstract. In line with the tenants of German Idealism, Hölderlin, one of the architects of this movement, perceived the absolute as the unity of phenomena and theory, the sensuous and the intelligible. As we have already seen, caesura was introduced by Hölderlin as a central device of his system by virtue of its ability to bring the unity of phenomena and theory to light: a poetisches Kalkül intended to bond all faculties of man and, accordingly, all the dimensions in which he lives. This unity also meant that the “fate” of the phenomenalized absolute is time, as Hegel put it at the end of Phänomenologie des Geistes ( 1979: 583–584), so that the representation of representation needs to account for this absolute temporality. The revelation of the absolute should thus be accomplished, according to Hölderlin, by imagination in the sublime moment of caesura which operates at all levels of narrative (as its mise en abyme-like instances prove) and shows the temporality of representation as such.
The most striking similarity between metalepsis and caesura is the fact that metalepsis also has, as Debra Malina put it, a “disruptive effect on the fabric of narrative” (2002: 1). Similarly, Werner Wolf described metalepsis as a “collapse of the narrative system” (1993: 356–358). Nevertheless, the differences are numerous, starting with the fact that metalepsis is always associated with humor, logical paradoxes and the fantastic (Genette  1988: 88; cf. Genette  1980: 234–237; Pier 2014: § 2, par. 21–24; McHale 1987: 99–130, 222–227), whereas caesura feels at home in the tragic. Metalepsis in its proper sense conveys a contamination between “the world of the telling and the world of the told” (Pier 2014: par. 1), whereas caesura means rather the intrusion of the silence that separates the narrator from what makes his narrating possible, namely the theatrical component described by the Polish narratologists as an element present in every narrative performance. This dramatic component which is revealed by caesura sometimes takes on the form of the difference between the narrator and the silent “implied author” who is the bearer of the possible forms of communication and whose mask the defined and limited narrator is; it may also be the difference between the frame of collective memory in the form of the uncanny all-encompassing memory of the implied author and what is actually told, i.e. chosen and actualized, the difference which endows the narrator and his discourse with a discernible shape and an ability to affect the receiver. Caesura lays bare the difference between the given narration and the temporary (in both meanings) measure of all things against which this narration appears as reliable or not, innovative or archaic, funny or boring, tragic or banal. It is a secondary issue whether we place this measure in the text, like the structuralists did (Okopień-Sławińska  1998, 1975; Schönert 2014: par. 7; Schmid 2013: par. 14), or within the cognition of a living person that processes the text (Herman 2013b). Caesura opens the gap between the possible and the actual in narrative without openly disclosing the intervention of the telling subject into the told, but rather the conditions of the possibility of dramatic performance that lie at the root of narrative. Commenting on Diderot’s intuitive description of narrative communication from Ceci n’est pas un conte – “ Lorsqu’on fait un conte, c’est à quelqu’en qui l’écoute ; et pour peu que le conte dure, il est rare que le conteur ne soit pas interrompu quelquefois par son auditeur ” – Michał Głowiński writes that it is part of the reader’s condition as the receiver of the literary work to interrupt the process of narration in order to ask questions on conventions and stereotypes, according to which the particular occurrence of communication functions (Głowiński 1992: 144, original emphasis). The extradiegetic narrator whose epiphany is metalepsis (and sometimes also mise en abyme) is but a part of a dramatic situation that caesura is supposed to reveal. He turns out to be one of many “human but not personal” (Balcerzan 1971: 82) roles to be assumed by the implied or empirical author – the roles of emitters and receivers whose repertoire is narrative fiction.
Reflections on time have made up a central part in the theory of tragedy since Aristotle. They continued in the early modern poetics of neo-classicism with the doctrine of the three unities. However, Hölderlin elevates the philosophy of tragic time to another level when he says that the empty “transport” of tragedy conveys nothing but the conditions of possibility of appearance and disappearance:
For at the furthest frontier of suffering nothing else stands but the conditions of time and of space.
At that frontier Man forgets himself because he is wholly in the moment; and the God [sic!] forgets himself because he is nothing but time; and both are unfaithful [to one another]. (Hölderlin  2001: 69)
The “moment” this quotation identifies as caesura makes time perceivable, even though this moment is not “a part” of time. With regard to the spacio-temporal character of tragic caesura, o ne may say that whereas metalepsis lays bare the conditions of fictionality, because what distinguishes fiction from nonfiction is that in every fiction there is “at least the potential for narrative metalepsis” (Nelles 1997: 152), caesura exposes to the critical reader or viewer the conditions of appearing and disappearing of all phenomena that are to make a part of narrative, fictional or not, after narrative became at the turn of the nineteenth century the medium of all experience and reflection. One may, cum grano salis, apply here the distinction voiced by Niklas Luhmann: metalepsis reveals mediality of a form, whereas caesura makes visible the difference between the medium of narrative and particular forms (cf. Luhmann 2000: 30–32). Caesura’s difference from metalepsis explains why caesura’s representative in the world of man is the blind specialist in the underworld: Tiresias makes absence appear on the stage where he presents himself in his clairvoyant blindness.
The goal of tragedy, according to Hölderlin, is to stage the conditions of time and space and to make them palpable as a way of understanding human existence that is historical, i.e. dependent on Mnemosyne founded by the poets’ tale. Artistic forms that are characteristic for a given epoch or that, so to speak, grow out of their time exist “not just in order to learn to understand the spirit of the times but once that spirit is grasped and learned to hold it steadily and to feel it” (Hölderlin  2001: 118). Caesura in a narrative helps us to feel its flow, whereas the task of the critic would be to find or to provoke an interruption in a text. The critical dispersion of narrative with the help of caesura is a translation of the old forms of enclosing the world into a new perception that consists of shocks and break-offs, one which regards continuous narrative passionately but as something alienated from it, even irreparably lost and fascinating only because of this loss. Without adopting caesura in its midst, narratology runs the danger of becoming a dead science that deals with dead forms.
Making the eccentric theory of caesura a part of the academic organon of narratology means executing the will of Hölderlin, who began his notes on Oedipus with: “It will be a good thing, giving poets even in our country a secure social existence, if poetry, even in our country and notwithstanding the differences of the times and the political systems, is raised to the mechane of the Ancients” (Hölderlin  2001: 63).