1We are accustomed to reading fictions from cover to cover, following the story from beginning to middle to end. This “linear” fashion of reading is generally followed by readers who are patient enough not to skip the descriptive passages as they search for the action scenes or who don’t yield to the urge to find out how the story ends before deciding whether to read about how it begins. Of course, even linear readers will be confronted with analepses, prolepses, digressions, changes of level, etc. that proliferate in all but the most rudimentary narratives. Meir Sternberg, who is hardly alone in this respect, has devoted considerable attention to the intersequential relations between “the absolute dynamics of the causally propelled action” (fabula) and that of “the variable dynamics of the reading process” (sjuzhet) (Sternberg, 1978, p. 13). Needless to say, the text that follows rigorously isochronous relations between the two sequentialities, were that possible, has little chance of rising to the top of the New York Times Book Review’s list of recommended reading.

2The situation is different when it is the signifying medium itself that departs from the print-bound linear format to which we are accustomed. This is evident of course with mixed media forms such as the stage arts or comics, not to speak of the cinema or television, and now, more strikingly so, with the boom of digitally mediated narratives. But the media affordances offered by the mass-produced codex book in this regard must not be neglected or underestimated. Mallarmé’s Un coup de dés n’abolira jamais le hasard (1897) and Apollinaire’s Calligrammes (1913-1916) in poetry, of course, but also fictions such as Michel Butor’s Mobile. Étude pour une représentation des États-Unis (1962 ; dedicated to the memory of Jackson Pollock), William Gass’s Willy Master’s Lonesome Wife (1968) or Mark Z. Danielewski’s House of Leaves (2000) with their elaborate and sometimes perplexing graphic and typographic devices woven into the narrative are but a few examples. Brian McHale, in his Postmodernist Fiction (1987), devotes a section to works generated out of “a spatial displacement of words.”

3The competing narratives I wish to discuss are contained in Gabriel Josipivici’s 1974 short story “Mobius the Stripper,” subtitled “a Topological Exercise.” Two stories are presented in this text, not one after the other but one “over” the other or, from another perspective, one “under” the other, separated from one another by a horizontal line across the middle of each page from the first page to the last. The narrative is not immersive – quite the contrary – for one is constantly at pains to know where to stop reading one text and start reading the other. At the end of the top half of a page in the midst of a sentence in order to read a sentence on the bottom half, which itself begins on the previous page ? At the end of a paragraph in order to see what’s going on in a new paragraph in the other text, even if this provokes a disorienting rupture between two diegeses ? Or bouncing erratically from one text to the other in the hope, rarely satisfied, of coming across helpful hints and details ? Widely flouted are the rules of what text linguists call cohesion (the words of the surface text “mutually connected within a sequence”) and coherence of the textual world (“the configuration of concepts and relations which underlie the surface text, are mutually accessible and relevant”) (De Beaugrande and Dressler, 1981, p. 3-4 ; emphasis in original). One could of course seek to solve the problem by reading each text from beginning to end separately. However, this choice raises a few crucial questions. Which text is to be read first ? The one at the top or the one at the bottom ? According to the choice made, which text, logically speaking, takes precedence over the other ? What are the consequences of choosing one order over the other ? Such questions are ultimately moot, however, for it is unlikely that all but the most disciplined of readers will keep his or her eyes from wandering. Not to be ignored, then, is the fact that the two halves of the dismembered page are right there in plain sight – clearly by authorial design – begging to be read as such. The malleability of the printed page in Josipovici’s story is difficult to elude.

Simultaneity ?

4These and other questions clearly arise from the page layout of “Mobius the Stripper,” its “topology.” Putting the two halves of Josipivici’s story together in a single reading would, it seems, produce an effect of simultaneous narration. According to Uri Margolin, speaking of “the scenic representation of simultaneous[ly] narrated acts, or of acts of narration,” simultaneous narration consists in “[t]urning the text on the page from a one-dimensional into a two-dimensional object by splitting the page into two or more distinct rows or columns running in parallel, each representing one of the concurrent activities or acts of narration.” Margolin goes on to point out that concurrency, as in Beckett’s “I say it as I see it,” “is faced with the inexorable difficulty of configuring into a meaningful whole that which is still in the process of becoming at narration time” (Margolin, 2014, p. 778).

5Now to get around the conundrum posed by purportedly simultaneous narration in “Mobius the Stripper,” we might postulate the horizontal lines on the pages as paratextual markers. Gérard Genette defines paratext as the “threshold,” an “indecisive zone,” between the inside (the text) and the outside (the world’s discourse on the text). He goes on to quote Philippe Lejeune who describes paratext as the “fringe of the printed text which, in reality, commands every reading” (Lejeune, 1975, p. 45). Paratext is a zone of “transition,” but also one of “transaction” with the public (Genette, 1987, p. 7-8). As epitext, paratext lies outside the work properly speaking (interviews, correspondence, etc.) ; as peritext, it appears within the text (prefatory material, chapter headings, footnotes, graphic devices, etc.).

6Clearly the lines cutting across the pages “command” the reading of Josipivici’s story, but in doing so they compromise cohesion and coherence between the two apparently simultaneous narratives, impeding the constitution of a meaningful whole. On the other hand, separating the page into two halves – top and bottom – with a horizontal line would suggest that the text at the foot of the page is a commentary or a set of footnotes. This supposition is quickly discredited, however, for there are no numbers or other system of cross-referencing for connecting the two texts in this way, and thus no criteria for determining what might constitute a footnote to a given portion of the story about Mobius ; consequently, the two texts cut across and interfere with one another at random, with no indication by either as to how or where to segment the other into meaningful units. As a peritextual marker, the horizontal line across the pages drives a wedge between the two texts rather than serve as a transition between them.

Duck-Rabbit effect

7The scrambling of the two halves of “Mobius the Stripper,” projecting an uncertain sense of simultaneous narration while unsettling the diegetic ordering of the narrative and at the same time provoking, through paratextual means, a rupture in the unfolding of the story, inevitably produces a cognitive counterpart to these abnormal textual features. In seeking to process the text, as it is presented, how does the reader proceed ?

8To answer this question, it is useful to invoke the so-called Duck-Rabbit Ambiguous Figure. When looking at this figure, one will see a duck or a rabbit, but not both at the same time.1 A “gestalt switch” (also referred to as “perception reversal”) occurs which it is difficult for the onlooker to control, for when attempting to see a rabbit, one may unwillingly see a duck or, looking for a duck, will discern a rabbit. This is so even though the retinal image of the figure is the same in both cases ; what changes is the phenomenal experience.2

9In processing “Mobius the Stripper,” the reader experiences a similar gestalt switch : attention is drawn to one or the other part of the page to the provisional and wavering exclusion of the other part. As a result, the cognitive processing generally associated with the reception of narrative is impaired, unassisted by the textual features that typically guide narrative comprehension.

10Now to be sure, Josipovici’s story contains two diegetic universes, each of which, taken separately, stands on its own. As such, each is a “natural” narrative in a sense close to “spontaneous conversational storytelling.” (Fludernik, 1996, p. 13-17). Taken together, however, one is a “duck,” the other a “rabbit,” so to speak, that cannot be apprehended simultaneously nor, lest it be by random and quasi-involuntary gestalt switches, successively.

11The top half of the page is a third-person account of Mobius who works in a club as a stripper. His motivation is not sexual but metaphysical, an existential quest. Fat and Buddha-like, he talks while he performs. His aim is “[t]o take off what society has put on me,” to “[t]ake off the layers and get down to the basics” in order to “know the truth. Is all a matter of stripping.” (Josipovici, [1975] 1990, p. 134-147). Later he asks “For what is life ?” The answer : “Chance. And what is my life ? The result of a million and one chances. But behind chance is truth.” And Mobius goes on : “Beyond a man’s chance is his necessity. […] For each man is only one truth and so many in the world as each man is truths” (p. 149). Finally, he declares “So I come to myself at last. […] To the centre of myself” (p. 151). “Smiling to himself in the mirror,” just before taking his life, “Mobius waited for them [the voices] to finish” (p. 152).

12The bottom half of the page is narrated in the first person by a nameless young writer plagued by writer’s block. It begins with the sentence “I first heard of Mobius the stripper from a girl with big feet called Jenny” (p. 133). Jenny urges the writer, in vain, to go see a show by Mobius, telling him “You can’t write without experience, and how the hell are you going to gain experience if you stay shut up in here all day long ?” (p. 135). The writer resists this pressure. One day, however, when walking in a park he sees a peacock that reminds him of Jenny : “‘Tell me the truth,’ I said to the peacock with the big feet.” We see here that the peacock in the park and Jenny – both with big feet – come together for the writer in an epiphany. The writer exclaims : “Myth. Ritual. More than an idea. A metaphor for life. ‘It is !’ I shouted, suddenly understanding. ‘It is ! A metaphor for life !’” (p. 148). Inspired, he returns to his room to write. “And then, suddenly, out of the blue it started to come.” To Mobius’s “I come to myself at last” he counterpoises “Perhaps it was only one story, arbitrary, incomplete, but suddenly I knew that it would make its own necessity and in the process give me back my lost self.” Pinning a message to the door – “Dear Jenny. Dear Mobius. Dear Peacock. ‘Gone out. Do not disturb’” – he ends with “Then I sat down and began to write” (p. 151-152). What he writes can be none other than “Mobius the Stripper.”

Möbius strip

13It becomes clear, at the very end of Josipovici’s story, that it is with the final sentences that the two strands come closest to co-occurrence. What could barely have been surmised earlier is now evident : the concluding sentence of the writer’s narration joins up with the top half of the text, thirty pages earlier, which begins “No one knew the origins and background of Mobius the stripper” (p. 133). Similarly, the end of the narration in which Mobius, looking into the mirror and waiting for the voices he hears in his head to stop as he pulls the trigger, joins up with the beginning of the writer’s narration, the extinguishment of the character’s consciousness effectively coinciding with the incipient act of invention : “I first heard of Mobius the stripper from a girl with big feet called Jenny” (p. 133). In this way, the logic of the Möbius strip is built into the text : a strip of paper is twisted and joined at the ends in such a way that a single line can proceed along it in infinitum without changing sides. The question of the post hoc ergo propter hoc fallacy fades away.

14Use of the Möbius strip in prose fiction has taken on a more daunting form in John Barth’s Lost in the Funhouse (1968). Described by Barth as a “sequence or series” (rather than a mere “assortment”) of fourteen short stories, the book opens with a minimalist text entitled “Frame-Tale” : along the edge of a page, which the reader is instructed to cut out and fashion into a Möbius strip, are written the words “once upon a time there” on one side, and on the other side the words “was a story that began.” As Barth observes, “It happens to be, I believe, the shortest story in the English language (ten words) ; on the other hand, it’s endless” (Barth, [1968] 1987, p. VII). “Frame-Tale” is a mise en abyme : it serves as a miniaturized but prolific model for ordering and reordering the apparently conventional succession of autonomous stories into no fewer than fourteen recursive and reversible patterns reaching up to seven levels. Reading the book proves to be a labyrinthine experience.

15“Mobius the Stripper” can be seen as a variant of this pattern. On each side of the line that passes through the middle of the text from the first page to the last is a narration whose “end” merges ceaselessly into the “beginning” of the other. Moreover, the disruptive presence of this line produces, as we have seen, a number of unusual effects, making the text recalcitrant to the normal processing of narrative. So on the one hand is a haphazard succession of gestalt switches between the two halves of the page while on the other two narratives unfold and merge into one another as they are twisted into a Möbius strip. It thus emerges that the “topology” of Josipovici’s story spawns two logics : that of the Duck-Rabbit Ambiguous Figure and that of the more paradoxical Möbius strip. How these two figures interact, combine, and diverge underlies the very dynamics of the story, its discoursive action.

16Looking now more closely at the latter figure, one cannot but concede that paradoxes are widespread in literary works. Few would deny (in modern cultures) that in the real world dead people don’t recount their past (or present) experiences, and it would be odd, to say the least, that a detective turns out to be the very murderer he is looking for. In works of the imagination, by contrast, such anomalies and implausibilities abound. However, the paradox at issue in Josipivici’s story is not that of a tenet or set of circumstances running contrary to received opinion or belief. Rather, it is akin to a paradox of the logical kind : it derives from the sentence of Epimenides, a Cretan, who declares “All Cretans are liars.” The one-step self-reference of this sentence squarely contradicts the law of the excluded middle, as can also be seen in the two-step “The following sentence is false. The preceding sentence is true.” The Epimenides paradox, observes Douglas Hofstadter, “is a sentence about itself which would be true if it were false and false if it were true” (Hofstadter, 1979, p. 584). Describing such self-reference as a “Strange Loop,” (p. 691) Hofstadter comments on the connection between the Epimenides paradox and the mathematician Kurt Gödel’s Incompleteness Theorem, whereby sentence G in a closed system is unprovable but nevertheless true (p. 17-18). Strange Loopiness is illustrated visually by M. C. Escher’s 1948 Drawing Hands (which hand is drawing the other ?) ; it is also exemplified, three-dimensionally, by the Möbius strip.

Metalepsis

17The past quarter century has seen a growing body of research devoted to a device that has been associated with the Möbius strip : narrative metalepsis. First defined by Gérard Genette as “a shifting but sacred frontier between two worlds, the world in which one tells, the world of which one tells,” and described as “any intrusion by the extradiegetic narrator or narratee into the diegetic universe (or by diegetic characters into a metadiegetic universe, etc.)” (Genette, [1972] 1980, p. 234-236), metalepsis has since been refined in its theoretical dimension and adapted to the analysis of a broad corpus of works, both ancient and modern, in the various media. From this characterization it stands out that metalepsis, a violation or transgression of boundaries, is logic-resistant and paradoxical, but also that not all paradoxes found in narrative texts qualify as metalepsis.3

18Among the various models of metalepsis that have been proposed, one that seems well adapted to “Mobius the Stripper” has been elaborated by Sonja Klimek (2010, 2011). Adhering to the position as originally staked out by Genette in 1972 (with some modifications), Klimek stresses that metalepses are possible solely within literary works and are thus intrafictional.4 She then sets out the following typology :

1. “Descending metalepses” are downward movements from one narrative level to a deeper one :

1. from extradiegesis to intra- or hypodiegesis ;

2. from intradiegesis to hypodiegesis.

1. “Ascending metalepses” are upward movements from one narrative level to a higher one :

1. from intra- or hypodiegesis to extradiegesis ;

2. from hypodiegesis to intradiegesis.

1. Complex forms of metalepsis include

1. Möbius strip narratives (recurrence of descending and ascending metalepses) ; and

2. Illogical heterarchies (a figure at one level exercises power through a process of representation over a figure at its own diegetic level).5 [cf. Escher’s Drawing Hands]

19Josipovici’s story comes under Klimek’s Möbius strip narratives. But to fully grasp what this means, a number of additional factors must be taken into account. The title offers but a vague hint of what is to come. The most salient resolution of the quandary appears at the very end where, as the voices fade from Mobius’s head, the nameless writer sits down to write : the extinguishment of the protagonist coincides with the writer coming into his own. Or to put it in another way, whereas the existence of Mobius is dependent on the writer, that of the writer asserts itself only with Mobius fading away. But now which of the two strands of the story takes precedence over the other ? Given the material organization of the text, is it a narrative about Mobius that ends by tying up with the writer’s narration ? Or is it the end of the writer’s narration that throws back to the opening sentence of Josipovici’s story in which Mobius is introduced ? The answer to neither of these questions is conclusive, pointing to the conclusion that the work is structured along the lines of the Epimenides paradox – not, however, in such a way that the two strands logically invalidate one another but rather set up irreconcilable tensions between them. This state of affairs is undergirded by the insistent separation of each page into top and bottom with which the reader is confronted throughout. Such are the two logics – the Duck-Rabbit Ambiguous Figure and the Möbius strip – that, as suggested above, course through “Mobius the Stripper.”

Heuristic/Semiotic reading

20In their material manifestation, verbal narratives are normally so arranged that readers are prompted to follow the discourse continuously, registering prolepses, analepses, digressions, changes of speaker and register, etc. along the way in the process of constructing a narrative world. To account for this process, Marina Grishakova, with reference to Arthur C. Danto’s (1985) “narrative sentences,” has identified the “double-take” or “predictive-corrective” logic triggered by narrative discourse. Focusing on the delayed recognition of significant facts in the dynamic play between the known and the unknown, she outlines an exploratory or epistemic perspective on narrative suspense, curiosity, and surprise. “Narrative dynamics,” she writes,

translates into interplay between not knowing and knowing, mediated by the desire to know that prompts making inferences, exploring alternative paths and virtual scenarios, and identifying suggestive links. Some virtual scenarios contribute to narrative dynamics as alternatives to the actualized plotlines or versions competing for actualization ; others are closed or ruled out in the course of narration. (Grishakova, 2019, p. 379-380)

21Faced with a work such as “Mobius the Stripper,” this approach to narrative dynamics is hindered by the unhelpful graphic disjunction between narrative strands. In Josipovici’s story, inferences, links, and scenarios such as those invoked above are ephemeral, held indefinitely in suspense or must simply be abandoned, possibly replaced by other, provisional conjectures. The inferences entertained at the top or the bottom of the page may prove to be non sequiturs when transposed over the line, challenging any attempt to grasp an overall narrative coherence.

22Josipovici’s story generates narrative anomalies due to the unconventional page layout, and it is for this reason that its complex dynamics calls for a specific analytical strategy. This strategy is twofold, involving heuristic reading (or viewing), on the one hand, and semiotic reading (or viewing), on the other.6

23The reading (or viewing) of a narrative is heuristic in the sense that it is a prospective initial encounter of discovery, uninformed by a configurational “seeing-things-together” made possible by the intersection of a prospective heuristic reading and a subsequent retrospective semiotic reading – a process that is highlighted in “Mobius the Stripper.” Heuristic reading enlists what Umberto Eco, following Charles Sanders Peirce’s writings on abduction, has developed into a theory of textual communication centered around four types of abductive reasoning (Eco, 1983, p. 206 ; 1990, p. 156). “Overcoded” abduction occurs when, for example, the words “Once upon a time” trigger the assumption that the remainder of the story will follow the generic conventions of the fairytale : the general rule applies to a particular case in such a way that the conclusion flows “naturally” from the hypothesis. However, should an unanticipated turn of events pop up, incompatible with the fairytale genre, one will be forced to ferret out a plausible alternative, either by postulating a probable rule or taking an “inferential walk” outside the discourse, possibly in search of a relevant intertext. In this case, we have an “undercoded” abduction. It is out of the commerce between overcoded and undercoded abductions that fabulas are generated. Taken separately, the two stories contained in “Mobius the Stripper” proceed in this manner. As such, there is little in either of them that violates the received narrative conventions ; each has a beginning, a middle, and an end.

24Semiotic reading, which is primarily retrospective, presupposes heuristic reading and is characterized by higher-order abductive reasoning. When no tried-and-tested rule can adequately account for a troublesome textual detail, “creative” abductions come into play in the form of conjectures or fair guesses about possible facts or states of affairs (as in puzzle-solving) or as new laws are formulated to explain previously unknown data (as in scientific hypothesizing). Such facts and laws are of course subject to substantiation, an operation that takes place through procedures of inductive testing, or “meta-abduction.” Higher-order abduction tends to be elicited in zones of conjecture, uncertainty, and risk. In narratives, such factors are more likely to stand out in works of a certain complexity or unconventionality, thus engaging greater abductive effort, although these features may also be present in attenuated form in simpler works (in plot structure or style, for example) and call for a degree of abductive reasoning that is not immediately apparent.

The textual evidence

25Analysis of each of the two stories forming Josipovici’s text in accordance with these criteria of abductive reasoning would reveal a normal degree of narrative cohesion and coherence. On the other hand, applying such reasoning to the two texts conjointly as one proceeds along the disjunctive Duck-Rabbit trajectory will prove more arduous. Nevertheless, lower-order and higher-order abduction seem to be better suited to the exploration of narrativity in “Mobius the Stripper” than predictive-corrective logic, which is more closely bound to linear reading.

26Though extensive, the antinomy of the two strands is not absolute, as would be the case if story were to adhere rigorously to the Epimenides paradox. This is confirmed by the Möbius-strip “topology” of the text in which the end of one narrative flows endlessly into the beginning of the other. In addition, there are four identifiable and revealing instances of one narrative strand “overstepping” the line separating one from the other.

27Jenny, the big-footed friend of the procrastinating writer, has seen performances by Mobius and urges him, unsuccessfully, to go and see the stripper for himself as a source of inspiration. Although the three characters are never present in the same diegesis, Jenny’s role as intermediary sets in motion a train of undercoded abductions in response to various questions. Will Jenny convince her friend to attend a performance by Mobius ? Would such an encounter incite the writer to settle down to work ? Such questions that come to mind in the course of heuristic reading are addressed by the higher-order abductions brought about by the semiotic reading of the two competing narratives. By contrast, attempts at creative abductions in search of verifiable inferences as one flits from one half of the page to the other with little or no textual guidance frequently fail to get off the ground.

28At two other points the writer, speaking in the first person, enters the Mobius diegesis, both times metaleptically : “I first saw Mobius in a club in Buda. In Rio. In Albuquerque. […] I remember going to see him and. I first heard of Mobius the stripper from a kid down on the front in Marseilles. From a girl in Vienna” (Josipovici, [1975] 1990, p. 144). That the writer “first saw” and “first heard of” Mobius in this way (possibly in a daydream) directly contradicts the opening sentence of his own narration, where it is stated that it was from Jenny that he first gained knowledge of the stripper. Confrontation of the two narrative strands precipitates an unanchoring of deictic coordinates in such a way that, for this brief but meaningful instant, not only is the temporal ordering of the two diegeses rendered indeterminate, but also the diegetic status of the characters and actions portrayed.

29Further on, the abrupt switch from third to first person confirms the writer’s metaleptic incursion into the Mobius narrative.

He sat on the bed and stared at the wall, eating bananas and dozing. I first saw. I first heard. I remember His Excellency telling me about Mobius the stripper. In Prague it was, that wonderful city. I was acting as a private secretary to the Duke and had time on my hands. I was down and out in Paris and London. A girl called Bertha Pappenheim first mentioned Mobius to me. (p. 148)

30It emerges from these passages that the writer’s first encounter with Mobius occurs in several places through the agency of various individuals and at wholly unpredictable and indeterminate times. These details are inconsistent with and difficult to reconcile with the writer’s assertion that Jenny was the first to speak to him about Mobius.

31In the absence of any plausible pre-existing explanation to resolve these discrepancies within their immediate context, creative abductions will be sought out so as to view the situation from a broader perspective. As it turns out, the passages in question reflect the disruptive influence of the line separating the two halves of the page, doing so at particular points and on a reduced scale. Just as cohesion of the two narrative strands falters under the effect of the graphic design that marks their division, so the irreconcilable elements deployed in the two passages destabilize narrative coherence. It is then on the basis of these passages that the question of the metaleptic configuration of the narration as a whole is hinted at, even in the course of heuristic reading.

32It is on only one occasion that Mobius makes an appearance in the writer’s narration, on the bottom half of the page, and this while both are walking in the park : “A little group of people was standing under the trees some way along the path. One or two park wardens. A fat man with one of those Russian fur hats” (p. 148-149). How, from this utterance (which is not metaleptic), can it be determined that the man wearing a Russian hat is Mobius ? Referring back a few pages to the top half of the page, we find that “he [Mobius] just stayed in his room the whole time except for the occasional stroll down to the park and back, heavily protected by his big coat and Russian fur hat” (p. 144). Even though the two men are unknown to each other, this is the clearest, and in fact sole moment of diegetic continuity between the two halves of the page in the entire story, for it is during one of his strolls that Mobius is perceived within the diegetic space of the writer’s narration. And indeed, it is the one and only time that the writer lays his eyes on the stripper, albeit with no knowledge of his identity. By its very singularity, the incident serves to underscore the disruptive features that dominate the work as a whole. In contrast to many portions of the text, a lower-order abduction here enables a retrospective inference that connects the two utterances, an operation that otherwise remains largely impeded throughout the interactions between the two narrative strands.

33As narrative, Josipovici’s short story engages a process of heuristic reading and semiotic reading that induces constantly varying degrees of abductive reasoning, due in large part to the material layout of the page. The crucial turning point comes with the sudden epiphanic realization by the writer, while walking in the park, that myth and ritual are “a metaphor for life” (p. 148). From here on, the two strands become more cohesive, requiring less abductive effort than before, as they now run parallel to one another. This is associated with a heightened sense of simultaneous acts of narration despite separation of the two halves of the page, but also with the near-concurrence of diegetic time in the two strands, facilitating greater coherence between what is expressed in the two halves of the page : “‘So I come to myself at last,’ Mobius said. ‘To the centre of myself.’” (p. 151) ; and the final sentence : “I knew that it [writing] would make its own necessity and in the process give me back my lost self” (p. 151-152). Whereas higher-order abduction is elicited through much of the text (intensified by the Duck-Rabbit effect), lower-order abduction is affirmed toward the end. This is not to say, however, that the stories of the two characters merge into one. Each finds his “self,” but the existence of one is dependent on that of the other : a fictional self-projection by the writer, the stripper puts an end to his days as the voices fade from his head, just at the moment when the writer, stripped of his distractions, regains his self by writing what follows of its own necessity. Both, in effect, have fulfilled “a metaphysical need. To strip. To take off what society has put on me” (p. 133-134). Such is the goal announced by Mobius from the outset, even though their destinies diverge : Mobius, a stripper, is the performer of rituals ; the writer, anonymous, is a creator of myths.

Topological exercise 2.0

34We might now look at Josipovici’s story in a different light by tinkering with its topology. Rather than a text published recto-verso on fifteen sheets of paper bound together in a book, we could quite feasibly print the entire thirty pages of text (some 13,000 words) on a long strip of paper, with the Mobius story running as one continuous text along the top and the writer’s story as one continuous text along the bottom ; this lengthy “page” (with no page numbers, but still with a line in the middle separating the two stories) could then be fashioned into a Möbius strip. Such a transformation would bring about a number of substantial changes to how the narration unfolds.

35In particular, the “end” of each narration on the last page of the published version of the text would join up with the “beginning” of the other narration on the first page : the two strands would then flow into one another in infinitum, without beginning and without end. Where to begin ? Where to end ? The answer : wherever the reader starts and stops. As for the middle, were there any, it might just as well be located in the empty space surrounded by the Möbius strip as at some unspecifiable or arbitrarily chosen point along the two narratives that perpetually merge into one another. Perhaps the greatest quandary in the story is that when, in the final lines of the published version, the writer sets out to write “Mobius the Stripper,” it remains unknown whether the recounted events take place during the writing (possibly as an interior monologue), in a supposed but unidentifiable past, or maybe in an imaginary future. This temporal indeterminacy ties up with the effacement of the story’s beginning, middle, and end, and it is made yet more salient by the redistribution of the thirty pages in the form of an unwieldy Möbius strip. Redesigning “Mobius the Stripper” in this way underscores what is inscribed in the published version of the text, marked typographically with an identifiable beginning and end, by pointing to the logical conclusion : the Möbius strip, as paradox, serves to overcode the entire narrative and thus acts as a mise en abyme of the entire work.

36Desegmenting the narrative text by redistributing it in this fashion can be carried a few steps further. First, the line separating the two narrative strands could be removed, effectively collapsing the two competing narratives, Mobius’s and the writer’s, into one continuous Möbius-strip text. Were we now to eliminate punctuation and the spacing between the words, textual cohesion and coherence would all but disappear, giving way to a high degree of randomness. Consider the following example in which the last line of the top half of the page runs directly into the first line of the bottom half and then the last line of the bottom half of that page joins up with the first line of the following page :

himobytheyshoutedhowsyourdicksincemymorningswerespentdeliveringlaundryfornunap
[…]
littleselfioncepossessedtobedangerouslythreatenedbyiskeepinguphewouldreplyhowsyours
(p. 136-137)

37With effort, the words in these two fragments can be picked out. However, even those gifted with the hardiest of cognitive powers will soon struggle to find regularities and patterns of meaning over larger stretches of graphemes. Imagine now the entire text of the story presented in this manner, fashioned, moreover, into a Möbius strip on each side of which one of the narratives is printed. Clearly the disruptive gestalt switches between the two halves of the page triggered by the Duck-Rabbit Ambiguous Figure would cease to exist. At the same time, attempting to apply abductive reasoning to this endless string of graphemes along the Möbius strip would quickly come to naught : any sense of the story of Mobius and that of the writer would be obliterated, and indeed the overall intelligibility of the mass of marks covering the Möbius strip, despite the succession of more or less discernible (but syntactically precarious) words, would be annihilated. Artificially pushing the material organization of the text to this extreme all but neutralizes the two logics at work in the story, but in doing so it underscores their centrality.

Nonlinearity and entropy

38To extricate ourselves from the impasse into which this experimentation has led us, we now turn to the phenomena of nonlinearity in dynamical systems and entropy (a measure of information).

39If “Mobius the Stripper” is a “nonlinear” narrative, this is hardly in the sense of a story written in a meandering, roundabout or digressive way without a linear plot, as oral stories are sometimes told, for instance. More fitting in this case is the principle of nonlinear dynamics as understood in the complexity sciences, where entities cannot be broken down into smaller units (as in linear systems), but rather are examined taking account of the interactions of their constituents across several scales. “A striking difference between linear and nonlinear laws,” wrote Henri Poincaré,

is whether the property of superposition holds or breaks down. In a linear system the ultimate effect of the combined action of two different causes is merely the superposition of the effects of each cause taken individually. But in a nonlinear system adding a small cause to one that is already present can induce dramatic effects that have no common measure with the amplitude of the cause. (Quoted in Nicolis and Prigogine, 1989, p. 59)

40Read separately, the two strands of Josipovici’s story represent the superposition of two “linear” narratives. Given the graphic presentation of the text, however, the strands are deployed simultaneously. In this way, sensitivity to initial conditions, a feature characteristic of nonlinear systems, sets in from the very outset, triggering random switches between the two competing narratives (the Duck-Rabbit effect) together with the intensification of abductive processes.

41Fluctuating and unpredictable, nonlinear systems form a part of dynamical systems (or dynamics), a field pioneered by Poincaré which, according to Melanie Mitchell, “concerns the description and prediction of systems that exhibit complex changing behavior at the macroscopic level, emerging from the collective actions of many interacting components” (Mitchell, 2009, p. 15). It is noteworthy to what degree the numerous incongruences and overlaps permeating “Mobius the Stripper” reflect dynamics in this sense as a result of the tenuous relations between the story’s interacting components and the evolving larger macroscopic level. In effect, the competing narratives are the product of the juxtaposition of registers and levels, of colliding streams of information, of a clash of differently encoded series.

42When discussing nonlinearity, it is necessary to take into account the phenomenon of entropy, a topic of substantial complexity in itself. For the purposes of literary and narrative analysis, it is appropriate to narrow the subject down to what Demian Battaglia has termed entropy “flavors.”7 The first, labelled “surprise and chaos,” relates to how disorder and uncertainty, growing over time, are perceived and disambiguated, sometimes producing an element of surprise. The second entropy flavor is “synergy, or more than the sum of the parts,” where the collision of flows of information results in “information modification” and the extraction of synergistic surplus, causing a semantic “explosion” or “synergistic surplus” through the juxtaposition of registers and codes. And the third, “modeling complexity,” is an algorithmic entropy in which “de-compression” transforms noise into information, into “organized” complexity as opposed to “random” complexity.

43Though all of these entropy flavors are at play in “Mobius the Stripper” (and, it could be argued, in narratives at large), the most prominent is the synergistic variety, defined by Battaglia as “the non-trivial part of total entropy arising from colliding entropy streams.” When there is a high level of informational collision and clashes in a narrative fiction, a series of semantic “explosions” or “synergistic surplus” is triggered, resulting in information modification. A product of nonlinearity, synergistic entropy is not fully compressible : intertexts may inject unanticipated meanings into the narrative by merging disparate information ; intermedial mismatch between a text and figurative elements may generate meanings that cannot be extracted from the information contained in the two source media ; unconventional page layouts and/or (typo)graphic devices will call for resourceful discovery procedures.

44We turn now from the experimental manipulation of Josipovici’s story presented in the previous section to the topology of the text in its published form, subtitled “a Topological Experiment.” As observed earlier, this topology, resulting from the particular material disposition of the text, takes shape in the form of the Duck-Rabbit Ambiguous Figure and the Möbius strip. Operating at the macroscopic scale, it effectively opposes the entropic forces set in motion as from the early pages, thereby weakening narrative cohesion and coherence, to patterns that gradually emerge on a larger scale, discernible only under scrutiny. The Möbius strip, a logical paradox, on the one hand serves to “structure” the incongruous unfolding of the two competing narratives, and on the other hand acts as a mise en abyme that, bit by bit, lends an enigmatic coherence to the overall narrative. As for the Duck-Rabbit effect, it has been seen how the two narratives in Josipovici’s story, separated by a line, impede attempts, through varying degrees of abductive reasoning, to reconcile the disparities and clashes between the differently encoded sequences, the generator of synergistic entropy. From the perspective of nonlinear or complex dynamical systems, it can be seen here that predicting how the macroscopic scale will emerge from the collective action of smaller entities or units can proceed only probabilistically at bifurcation points occurring at the “edge of chaos.”8 Gradually, however, abductive effort decreases as the gaps between the two narrative strands diminish in the final pages and as overcoding by the other topological element of Josipovici’s story, the Möbius strip, becomes more apparent.

Mediality

45The format in which we are accustomed to reading narrative fiction (and non-fiction) is the book, consisting of a certain number of geometrically uniform printed pages to be read line by line from the top left to the bottom right. This has contributed to generations of readers associating narrative with linearity – a tenet one would be hard put to carry over to oral and other forms of storytelling. There have, of course, been exceptions to this received idea going back as far as Tristram Shandy, if not to pre-Gutenberg times, but by and large the idea of the linearity of narrative has prevailed and continues to do so. Modernism, and in particular postmodern literature, brought in a proliferation of works that have put narrative linearity to the test, a development propelled to no small degree by the proliferation of electronic and digital media, and that have produced a broad gamut of what Grzegorz Maziarczyk (2011) calls “typographic experimentation” and Brian McHale, adopting a phrase from Raymond Federman, “a spatial displacement of words” : the space of the printed page having been long left functionally invisible, the idea was to restore the technological reality of the page and its interaction with the word (McHale, 1987, p. 181).

46This point is not negligible with regard to “Mobius the Stripper” and other narratives characterized by graphic experimentation, for it raises the thorny issue of the relations between differently encoded series in narrative fiction falling under the framework of nonlinear complexity and the affordances and constraints peculiar to the medium in which it is couched. Marie-Laure Ryan has observed that once authors begin to play with the graphic layout and spatiality of print in written narratives, such that print is no longer regarded as the mere translation of spoken language, then closer attention must be paid to the relations between media and narrative (Ryan, 2006, p. 16-30). Accordingly, she proposes to examine media and mediality in their interrelated semiotic, technological, and cultural/institutional dimensions.

47Contemporary studies devoted to mediality in literature and the arts break down into two basic approaches. One is transmediality, defined as “medially unspecified phenomena that are not connected to a specific medium or its mediality and can hence be realized using the means of a large number of different media” (Thon, 2014, p. 335). The other approach is intermediality, where medial transposition refers to the passage of a work from one medium to another (film adaptation of a fictional or a factual text), media combination to a constellation of two or more media, ranging from contiguity (comics) to full integration (opera), and intermedial reference to the evocation of a painting in a novel, for example, or more comprehensively, the musicalization of fiction (Rajewski, 2002, 2005).

48How is “Mobius the Stripper” to be positioned in relation to these forms of mediality ? Clearly the story is in no way transposable to the oral medium – or for that matter to the stage or the visual arts. The sequence of experimental manipulations carried out above could conceivably be visualized three-dimensionally through digital means for illustrative purposes, but the result would be remote from both the narrative content of Josipovici’s story and the affordances peculiar to the pages as they appear in the published text. Nor can we say that the story is an example of media combination (as in the graphic novel), since only one medium is employed, that of the printed page ; or that the story refers to or imitates artefacts in other media. This being the case, it is intramediality that comes closest to describing the materiality of the story. In what sense is this form of mediality to be understood ?

49As defined by Irina O. Rajewsky, intramedial reference occurs between text and “one or more (real or fictitiously existent) single text(s)” (e.g., mise en abyme, intertext) or between a text and “one or more semiotic systems” (e.g., genre, text type) (Rajewski, 2002, p. 71). Presented this way, intramediality resembles Gérard Genette’s transtextuality. Rajewsky’s first type combines Genette’s subcategories intertextuality (inclusion of one text in another) and hypertextuality (relation between a text and its antecedent text) ; her second type is partly comparable to Genette’s architextuality (taxonomy). As for Rajewsky’s intermediality, it corresponds roughly to Genette’s transmodalization.

50This excursus is revealing in a number of ways, two of which concern us here. Rajewsky’s focus is broader than Genette’s to the extent that she is concerned with how different media interact in all forms of artistic expression, not merely in scriptorial media. On the other hand, Genette’s explorations of various forms of transtextuality are concentrated mainly on the hypertextual practices of transformation and imitation (through satire, parody, pastiche, etc. ; 1982) and paratextuality (1987). His interest is not in the nature of the relations between media ; in Rajewsky’s system, transtextuality would be a subcategory of intermediality restricted to a single medium. It is also important to note that Rajewsky understands medium in the very general sense of channels of communication, extending from oral to scriptorial cultures and on to the techniques of the modern mass media (Rajewski, 2002, p. 7). On the other hand, it is striking how little attention is paid to the constitutive features of the various media, to what distinguishes oral from scriptorial media, scriptorial from filmic media, etc. This shortcoming is not inconsequential, as any attempt to transpose the text of “Mobius the Stripper” from the scriptorial to the oral medium will readily show. The relations in the material text of Josipovici’s story can thus be described as intramedial, involving one medium. But how is this intramediality to be characterized, and in what ways does it contribute to the nonlinear quality of the narrative ?

Writing systems

51It is a widespread article of faith, rarely challenged, that glottic forms of writing are “linear,” just as speech, and that the signifier proceeds linearly. This assumption stems in large part from Ferdinand de Saussure’s theory of communication where the written word is understood as a visualization of the spoken word, a pale derivative of speech.

52This doctrine is contested by Roy Harris in Signs of Writing (1995), where a theory of communication, a theory of the sign, and a theory of writing systems are developed. A theory of writing, Harris stresses, is not a linguistics of writing, but a semiology of writing. Accordingly, he outlines a syntagmatics of writing based on four interdependent factors that simply do not exist in speech :

(1) the biomechanical dimension of communication (unlike auditory signs, which are transmitted by sound vibrations and are ephemeral, written signs require the use of writing implements and a prepared surface, leaving a durable trace) ;

(2) writing surface (sheets of paper, which have no auditory equivalent, are better adapted to the printing press than clay or marble) ;

(3) graphic space (“the disposition of written forms relative to one another and to other forms within the same graphic space” ; p. 121) ;

(4) directionality (succession of signs in a graphic space).9

53Speech occurs as a modifiable stream of sound, rich in expressivity, but it lacks the two- and three-dimensional resources of space. It is for this reason that the syntagmatics of writing enjoys “far greater variety and complexity than the syntagmatics of speech” (Harris, 1995, p. 46). This point is illustrated by three different page layouts of George Herbert’s poem “Easter Wings” : the modern version in which the first and the last lines of the two stanzas are long, narrowing down in the middle ; the original version where the same text is rotated 90 degrees and printed on facing pages ; the text printed as a standard paragraph without the verse structure. Far from being variants of one and the same writing system, Harris insists, these versions represent three writing systems that exploit graphic space differently (iconographically in the first two cases) and must be scanned by the reader in different ways ; they confirm the fact that written signs are not a visual transposition of spoken signs (Harris, 1995, p. 60-62).

54Another essential attribute of the syntagmatics of writing is the surface dependency of writing systems. One example is the following polygraph :

55LoUiSiAna

56Due to their varying fonts, the characters play a dual syntagmatic role, obliging the reader to deal simultaneously with two different scannings. In no way can the polygraph, whose relations are clearly intramedial, be assimilated into any of the forms of intermediality identified by Rajewsky. At the same time, the polygraph illustrates in a nutshell the pertinence of the syntagmatics of writing to one form of intramediality (Harris, 1995, p. 119-120).

57Closer to the topology peculiar to “Mobius the Stripper” is the Necker cube.10 Somewhat similarly to the illusion produced by the Duck-Rabbit Ambiguous Figure, four distinct cubes can be made out, depending on which angle one is looking from, but all four of the cubes cannot be seen simultaneously. Were the Necker cube constructed as a three-dimensional object, the illusion produced by its presence on a flat surface would disappear.

58With regard now to the Duck-Rabbit Figure, nowhere in Josipovici’s story does this figure appear as such. As discussed earlier, however, separation of the pages into two halves by a horizontal line results in an erratic series of gestalt switches, characterized as the Duck-Rabbit effect.

59There is an unspoken assumption that writing is visual. Visual, though, it is not : writing begins with graphic space and the disposition of written forms relative to one another within that space – a condition for the syntagmatics of writing that has no parallel in speech. It is also the case that the supposed “linearity” of speech is seriously marred by confusion between the visual and the temporal, the result of superimposing the linear on the auditory (cf. the famous “linearity of the signifier”). For biomechanical reasons alone, then, writing cannot be modeled after speech. Indeed, linearity is no better suited to describing the relative position of characters in the graphic space of writing than it is to characterizing the temporal succession of the sounds of speech.

60To describe the “geometry” of graphic space, Harris adopts the principle of alignment : a straight line as the shortest distance between two points. In writing, alignment extends either horizontally, forming rows (in European languages), or vertically, forming columns (in Chinese). The direction of rows is either left to right (European languages) or right to left, and that of columns top to bottom (Chinese) or bottom to top. Now, at the end of a row or of a column, the path is inevitably interrupted and taken up again on the adjacent parallel row or column, proceeding in the same direction in successive rows (left to right) or columns (top to bottom). As a consequence of the constraints of graphic space, glottic writing cannot be said to be linear. To be truly “linear,” textual organization would have to proceed along a spiral path, though writing (and reading) in this fashion, as Harris understandably points out, poses a set of insurmountable difficulties of its own.

61But what about Barth’s “Frame-Tale,” a story in ten words printed on a Möbius strip that proceeds linearly in infinitum ? Or “Mobius the Stripper” which, as the topology of the story shows, also follows this pattern ? The line cutting across each page from the first to the last, a generator of entropic noise, comes as a stark reminder that neither the printed page nor the development of the competing stories proceeds in a linear fashion. Textual cohesion and coherence are further called into question by a faltering sense of simultaneity between the two diegeses, triggering an irregular series of higher- and lower-order abductions on the part of the reader, but also by metaleptically induced paradoxes and indeterminacy as to where the overall story begins and ends. As in dynamical systems, the interactions of the parts, subject to entropic forces, tally imperfectly with the whole, which is itself in a state of flux. Curiously, the presence of the disruptive line confirms the nonlinear nature of syntagmatics of writing while it also serves to complement the nonlinear series of elements at the level of narrative content. It is then all the more noteworthy that this line should re-emerge as the line running endlessly along the Möbius strip, as illustrated by the experimental manipulation of the text, demonstrating that Josipovici’s story is overcoded by this figure.11