From: Cervantes: Bulletin of the Cervantes Society of America 7.2 (1987): 57-69.
Copyright © 1987, The Cervantes Society of America

NE OF THE THINGS that has always fascinated me about Cervantes' shorter narratives is the wide variety of story-telling strategies he employs in them. Even if we confine our investigation to the inset narratives found in Don Quixote, Part One, we can observe a surprisingly large number of different structures and recounting techniques. The unfortunate history of Marcela and Grisóstomo, for instance, is unfolded in four distinct sections, each with its own narrator and a different point of view. The goatherd Pedro begins the narrative in Ch. 12 with an account of the events that have led to Grisóstomo's infatuation and eventual suicide; the student Ambrosio proceeds to embellish the story somewhat in the next two chapters; then Vivaldo, the gentleman who stumbles upon the group of mourners as they proceed to the burial service, concludes the “objective” testimony by reciting the final canción desesperada of the deceased, verses which present a decidedly negative portrait of the object of the late Grisóstomo's affection. Vivaldo's recitation (Ch. 14) occurs mere moments before the appearance of the supposedly coldhearted beauty, Marcela, who

suddenly emerges —very much like a three-dimensional figure that pops forth from a flat backdrop— to offer us a totally different version of the events that have led to Grisóstomo's suicide. The sudden change of perspective introduced by the solitary shepherdess and the unsettling effect her contradictory testimony has upon the reader are what give special significance to the episode. Ultimately, then, it is Cervantes' unconventional narrative technique, rather than the events themselves, that confers literary value upon this adventure.
     A similar but slightly different strategy is employed a short time later in a more felicitous romantic intrigue. Cardenio's one-sided account of his luckless pursuit of Luscinda is presented in Chapters 24 and 26; this is supplemented shortly thereafter by Dorotea's narration of different but related events in Ch. 28. The reader is then held in suspense until Ch. 36, when Cardenio's rival Fernando provides the concluding section of narrative. In the meantime, however, still another romantic adventure, the purely fictional story of El curioso impertinente, is introduced, interrupted (the adventure of the wineskins) and concluded. Structurally, the most unusual aspect of the Cardenio-Dorotea-Fernando-Luscinda affair is the fact that all the complications and misunderstandings are resolved and the proper pairings made (Dorotea with Fernando, Cardenio with Luscinda) prior to Fernando's concluding narration. From a linear standpoint, the main plot action is brought to a halt before all the essential background material has been presented. Fernando's explanation, containing as it does information necessary for a complete understanding of the previous events, is cleverly appended as an epilogue in which the author casually explains —almost as an afterthought— the mystery of Luscinda's previous strange behavior. In what should be considered a bold departure from contemporary literary convention, Cervantes has designed his story to hold the reader in suspense even after the anticipated happy ending has been achieved.
     The aforementioned tale of the Curioso impertinente exhibits still another curious narrative pattern, this one of a serpentine nature as the story constantly doubles back on itself, each new block of narrative seeming to reverse the action of the previous section. The first part consists of Anselmo's plan to test his wife Camila's fidelity by having his friend Lotario make inappropriate sexual advances toward the young bride (Anselmo's Plan: A and L vs. C). But pretense quickly gives way to passion and soon the bride and best friend have become romantically involved in a conspiracy to deceive the foolish husband who has orchestrated this most unwholesome situation from the

beginning (L and C vs. A). An unexpected development causes Lotario to reevaluate the situation and return his allegiance to the unsuspecting Anselmo. The two friends are thus moved to conspire once more against Camila (Lotario's Plan: L and A vs. C), but before very long Lotario regrets his impetuous change of heart and decides to make a clean breast of it to his paramour. At this point it is Camila who concocts the final ingenious scheme —the most sophisticated and devious of all the plots hatched so far— that proves to be almost as foolproof as it is bold (Camila's Plan: C and L vs. A). Predictably enough, even this clever plan comes unraveled in an unforeseen turn of events; the result is death and unhappiness for all the principal players in this unholy masquerade. As a moral treatise, El curioso impertinente is not especially clever or insightful, but an example of ingenious plot complication and total narrative control, it is a masterpiece. The serpentine structure is evidenced on four different levels separated by three independent reversals of the action:

     A different (i.e., thematic, rather than structural) but equally interesting aspect of the inset story of El curioso impertinente is the criticism of that work offered immediately thereafter by the Priest. Pero Pérez questions the verisimilitude of this piece of fiction (he says the author “fingió mal” because no married man would ever be so foolish as to test his wife in the manner Anselmo did), but he does have kind words for the writer's style. As J. B. Avalle-Arce¹ and Bruce W. Wardropper² among others have noted, the Curate's criticism of the fictional Curioso impertinente ought to be contrasted with Fernando's later remarks (Ch. 42) about the purportedly true adventures narrated by the Cautivo. In this second case, however, the question of literary verisimilitude is never brought up; such plausibility is assumed to be inherent in historical narrations of this sort. And once again there is

     ¹ Deslindes cervantinos (Madrid: Edhigar, 1961), pp. 129 ff.
     ² “The Pertinence of El curioso impertinente,” PMLA 72 (1957), 591 ff.

high praise offered for the narrator's skillful manner of presentation. The critical point Cervantes underlines here is that the artist / novelist, in order to satisfy the rigorous albeit somewhat absurd neo-Aristotelian precepts of the age, is obliged to strive not only to produce an entertaining narrative, but also to maintain at the same time the appearance of factual or historical accuracy that will give his work the verisimilitude required for works of fiction. On the other hand, the historian, dealing as he supposedly does with incontrovertible facts, need not ever be concerned about manufacturing plausibility; he is required only to get his facts straight and narrate the events with an entertaining style.
     The difference between history and fiction is a theme Cervantes returns to in his novella, El coloquio de los perros, an unusual and unique item among his published works. The Coloquio has come to us in the form of a prose dialogue and combines the imaginative entertainment value of the fictional Celestina with the didactic critical acumen of El Pinciano's Philosophia antigua poética, two celebrated works that utilized the same dialogic format. But there is a structural parallel between Don Quixote and the Colloquy that merits even more attention at this time. I would like to explore now a particular narrative technique employed by Cervantes with considerable success in both his famous novel and in the less renowned novella.
     The structure of the Coloquio itself is relatively simple: Berganza narrates the story of his life while Cipión casually comments on the content and style —mostly the latter— of his companion's narration. However, the transcript of the dogs' conversation does not stand alone; it has been joined to the preceding tale, El casamiento engañoso, in what appears to have been a belated and most fortuitous inspiration. At some point prior to publication Cervantes decided, for thematic as well as technical reasons, to fuse these two tales. As we shall see, the thematic reasons have to do with the content of the Casamiento, while the structural alterations are made necessary by the distinctive techniques employed in the Coloquio. With regard to the latter tale, I perceive two different, though related, technical aims. The first is a desire to provide a proper narrative frame for the heretofore naked prose dialogue; the second —an even more significant decision, as it turns out— is to superimpose two extra layers of literary and social commentary upon the dialogue's original pair of critical strata, thereby reprising one of the author's favorite narrative devices from the Quixote.
     Ruth El Saffar has shown that the Coloquio consists of a series of

four narratives, one locked inside another, each containing a different narrator-critical listener combination. Furthermore, each of the four critical layers of the newly structured Coloquio is designed to provide a commentary on the style and content of the narrative section immediately below it.³ I would like to expand upon her observation to outline what I perceive to be Cervantes' ultimate scheme for the novella: a narrative spiral that first descends then ascends through four distinct planes of relation.⁴ At the top level there is the Licenciado Peralta, who in the frame tale reads and comments on the written narrative of Ensign Campuzano; immediately below that we have the eavesdropping Campuzano in the act of recording the conversation of the miraculously endowed canines and later offering his own judgments about the verisimilitude of the events he has witnessed and recorded. At the third level we leave the frame tale and enter the text of the colloquy itself; here speaker / narrator Berganza recounts his picaresque adventures to listener / critic Cipión, who continually interrupts to inject some personal remarks about the content of his friend's narration and the manner in which it is being delivered. The bottom level is reached when the Toledan witch Cañizares, in the course of one of her demonic reveries, relates to a somewhat incredulous Berganza the bizarre account of how a pair of children were transformed into canines by a sorceress' incantation. It should be noted that this strange tale of demonic forces and implied metempsychosis lies at the linear midpoint of Cervantes' novella and represents, literally and figuratively, the core of his labyrinthine narrative. Having descended through three layers before reaching this core, the narration begins to ascend again after the witch has finished her incredible tale. Back on the third level Berganza wonders aloud whether Cañizares' strange prophecy may somehow be responsible for the faculty of speech with which he and Cipión have suddenly been blessed. He then continues with his life story and concludes his account with how he came to enter the service of the good Mahudes at the hospital in Valladolid and miraculously received there the power of speech. Cervantes brings the tale to a swift conclusion by returning the

     ³ Novel to Romance: A Study of Cervantes's “Novelas ejemplares” (Baltimore and London: The Johns Hopkins University Press, 1974), p. 81.
     ⁴ The phenomenon of descending and ascending trajectories in these tales is discussed by El Saffar in Cervantes: “El casamiento engañoso” and “El coloquio de los perros”: A Critical Study (London: Tamesis,1976), p. 38. Alban K. Forcione also points to this effect in Cervantes and the Mystery of Lawlessness (Princeton: Princeton University Press, 1984), pp. 102-03.

reader to the frame, where Campuzano remarks about what he has overheard (Level Two), and Peralta comments on the general verisimilitude of the Ensign's written account (Level One).
     The entire novella can therefore be diagrammed linearly (L = Listener; N = Narrator):

     This is not the only occasion on which Cervantes has opted to create such a multi-tiered narrative structure in his work; the decision here to link a pair of seemingly unrelated tales —and thereby move from two to four levels of critical discourse in the Coloquio— is very much akin to his belated creation of the Moorish historian Cide Hamete Benengeli at the beginning of Chapter 9 of the 1605 Quixote.⁵ With this previous ingenious insertion Cervantes was able to interpose between himself and the reader not just a straw man, a convenient target for critical barbs in the person of a so-called Original Historian (and a properly unreliable one, at that), but also two additional buffers in the form of a Morisco Translator and a Spanish Editor (sometimes called the Second Author). The revised plan for the novel would permit each of these contributors to, and therefore partial authors of, the novel to avail himself of several opportunities to comment on the verisimilitude, decorum, and various other literary attributes of the book as it was being written. In so doing, Cervantes invented an ingenious format that would allow him to write an experimental novel while commenting simultaneously about a few of the outmoded literary precepts he was pretending to invoke in the course of its composition. The renowned novelist clearly perceived that there would be more room for the kind of critical debate he wanted to inject if the fictitious narrator could be made highly visible to the reader, noticeably fallible in his judgments, and justifiably insecure about his critical opinions. Once he committed himself to writing a novel about how one ought to write a novel, Cervantes was simply obliged to create a bogus author like Cide Hamete. Appending the figures of the Translator and the Editor

     5 The technique of multiple layers or planes of commentary is also used toward the close of Book I of the Persiles, when several narrators are engaged in telling their life stories to a critical audience.

was still another stroke of genius, a carefully calculated move designed to provide yet one more layer of critical reflection, and therefore additional grist for the mill of his ongoing debate.
     Alban K. Forcione observes that behind the satire of the Coloquio lies an equally ingenious, if highly unusual, plan that he refers to as a “deliberate cultivation of formlessness.” The reader is obliged to wade through a series of intrusive elements (constant interruptions, annoying repetitions, needless retrogressions and qualifications, etc.) designed to interfere with his desire to arrive at some satisfactory explanation of Berganza and Cipión's sudden and miraculous acquisition of the power of human speech. The critic states that Cervantes' intention was to compel his reader to experience a sense of narrative disintegration as he is confronted with “an artistic creation that is outrageously and self-consciously unartistic.”6
     To judge from the dates of publication of the respective works, it is safe to assume that the multi-layered narrative device as we find it in the Quixote antedates its use in the Coloquio. It is also worth noting that Cervantes utilized the technique of layered critical commentary on various occasions and in both parts of his masterpiece, as George Haley was the first to document.⁷ Haley's thesis, brilliantly developed and presented, focuses on the parallel relationships that exist between a series of characters involved in the grand-scale composition of Don Quixote's story and the participants in Maese Pedro's relatively fragmentary retablo. Professor Haley has shown (pp. 146-47; 161-62) that the early division of the novel into four levels of narrator-listener / reader engagement (DQ I, 9) is echoed again precisely and in microcosm during the adventure of the puppet show (DQ II, 25-26). The narrators who would ordinarily speak or write in the first person (Cervantes and Ginés de Pasamonte, respectively) hide behind new identities (the created roles of Cide Hamete and Maese Pedro) that enable them to render their creations with artistic self-sufficiency in the less subjective third person. Maese Pedro's young assistant, therefore, can be seen to mirror the role of the Translator and Editor / Second Author (who function as one at the third level) in transmitting the narration to the fourth and final component, the intended audience (the reader of the novel; those gathered to witness the retablo). Haley's observations can be summarized in the following scheme:

     ⁶ Mystery of Lawlessness, p. 37, emphasis Forcione's.
     ⁷ “The Narrator in Don Quijote: Maese Pedro's Puppet Show,” MLN 80 (1965), 145-65.

     Unfortunately, Haley's ingenious analogue, for all its theoretical correctness, cannot be applied comfortably to the Coloquio. The latter operates under a slightly different narrative plan —there is no “hidden” narrator, for example— although it does revive the Cervantine practice of utilizing multiple layers of critical commentary. With regard to structural pattern, the Coloquio is more like the Cave of Montesinos episode than any other, as I shall presently demonstrate.
     In Ch. 24 of the Second Part of Cervantes' masterpiece, (1) the Translator and the Editor (twin voices, as I have noted) decide to let stand (2) the marginal comments of Cide Hamete regarding the credibility of (3) Don Quixote's affirmation as absolutely true (4) his account of the events he witnessed in the Cave of Montesinos. Working backwards, then, we find at the core Don Quixote's rather fantastic account of his visit to the enchanted world of Montesinos; atop that there is Sancho's understandably skeptical reaction. Moving onto the supra-textual plane, we encounter at Level Two the marginal comments of the dubious Moorish Historian regarding the questionable veracity of the adventure he himself has just recorded. And finally at the topmost layer there are the ambivalent remarks of the hypercritical Translator and Editor who have opted with some misgivings to publish the entire record and let the reader decide for himself whether or not to believe Don Quixote's story. Correspondingly, at the core of the Coloquio is the utterly incredible story narrated by the witch Cañizares, atop which we find Berganza's expression of complete disbelief at what he has just heard from the old hag's lips. As we ascend to the supra-textual zone we encounter the Ensign's marginal expressions of doubt as to the truthfulness of what he himself has transcribed and, ultimately, the critic Peralta's begrudging acceptance of his companion's version of the dogs' colloquy. The following scheme illustrates how each of the four levels of the Cave of Montesinos episode corresponds precisely to a similar narrative plane of the Coloquio:

     It is no mere coincidence that Cervantes should choose a pair of characters like Campuzano and Peralta to be his commentators on the Coloquio, or that the unfortunate Ensign should ultimately find himself in an untenable Quixote-like position. In the preceding Casamiento, Campuzano tells a similarly incredible tale about a contrived counterfeit marriage that has ultimately backfired on him. At the close of Campuzano's narration Peralta demonstrates a reasonably skeptical attitude toward the Ensign's story. Asked now to read an even more fantastic tale about two talking dogs, Peralta declares that he is not prepared to believe anything from Campuzano's lips —or pen, for that matter. The Ensign is obliged to admit that this second story is also quite strange, that he himself had difficulty believing it, even as it was unfolding, and indeed had initially believed it to be a dream. Upon reflection, however, he has concluded that on his own he could not have imagined or invented such a voluminous amount of incredible dialogue, so he affirms that the dogs really did speak that night.
     The compromise reached at the close of El coloquio de los perros, besides effectively replicating the denouement of the Cave of Montesinos episode, in many ways also echoes the theme of history vs. fiction we saw in the judgments passed by the Priest and Fernando on El curioso impertinente and the Cautivo's story, respectively. Peralta dismisses Campuzano's transcript of the dogs' colloquy as “fingido,” but praises it for being well composed and urges the Ensign to continue with the promised second part. Campuzano insists that the entire account is true (i.e., history), but the more sophisticated Licenciado refuses to accept it as anything but fiction. In the end they decide not to dispute any further as to whether the dogs actually spoke (history) or not (fiction). All roads, it would seem, lead from the Coloquio back to the Quixote.
     Turning now to the thematic reason for Cervantes' linking the Casamiento to the Coloquio, let us consider the hypothesis put forth by

Ruth El Saffar in a 1973 article: the Casamiento is actually a reworking of the Coloquio, one which casts light on the dogs' dialogue and transforms its meaning in the same way that Part Two of Don Quixote represents a revaluation of Part One.⁸ In this critic's view, the fusion of the Casamiento with the Coloquio allows the lesser of the two stories, the dogs' colloquy, to become part of a greater and more meaningful whole. In the process an otherwise implausible conversation is incorporated into a broader and much more believable tale dealing with an author's moral and physical recovery —or resurrection, if we consider the symbolism of the hospital's name— from the illness, alienation, and disgrace that have recently plagued his life.⁹
     Just as Haley views the Maese Pedro episode as an analogue of the narrative scheme for Don Quixote as a whole, E1 Saffar sees the purgative effect of the unfortunate Ensign's experience in the combined Casamiento and Coloquio as an echo of Don Quixote's equally unsettling adventure in the subterranean enchanted world of Montesinos' Cave. She points out in her article that both the Ensign and the Mad Hidalgo are suffering, alienated, and intensely disoriented individuals at the time they begin their adventures.¹⁰ They are consequently obliged to descend in their dreams to a nether world where they will undergo an intense spiritual transformation that represents the start of their journey back to good physical and mental health.¹²

     ⁸ “Montesinos' Cave and the Casamiento engañoso in the Development of Cervantes's Prose Fiction,” KRQ 20 (1973), 451-67.
     ⁹ Forcione at one point appears to concur with El Saffar, observing that the Casamiento is an independent confession framing and containing a subordinate one (i.e., the core episode of Cañizares' revelations); he views the frame tale's elaborate set of imaginative correspondences as a link with the episode of the Toledan witch (135). At a later point he parts company with El Saffar, labeling the Casamiento a powerful counterforce to the narrative movement of the Coloquio, which he declares to be the more dominant of the two stories (146).
     ¹⁰ “Montesinos' Cave . . . ,” p . 464.
     ¹¹ Forcione, disagreeing, prefers to interpret the Cave of Montesinos adventure as a parody of the romance conventions of anagnorisis, an ironic treatment of the descent and recognition themes of the romance tradition. These same conventions later reappear in the Coloquio during Berganza's strange encounter with the witch Cañizares (48-49). In Forcione's view, the theme of Don Quixote's descent into the cave is man's adversarial relationship with time and its consequences (e.g., mortality and decay). The Mad Hidalgo is confronted in his dream with the fact of man's inescapably transitory nature and the ultimate futility of his own dreams of immortality. The Coloquio, on the other hand, is said to deal with the moral theme of demonic powers in conflict with divine purposes (51-55).

     El Saffar demonstrates furthermore that each man's experience in these episodes is bipartite, consisting of one realistic and one fantastic encounter, but recounted in a different order.¹² The first part of Don Quixote's cave adventure is utterly implausible (the encounter with Montesinos, Durandarte, Belerma, and the others in a centuries-long state of suspended animation), while the second portion has a decidedly realistic tone (the enchanted Dulcinea begs Quixote to lend her six reales, a request he is unable to fulfill). Campuzano's ordeal runs in the opposite direction: it opens with a clearly picaresque account of his disastrous courtship and marriage (and the syphilitic condition that results from his sexual union with the cunning Estefanía), then comes to a close via his dream-like encounter with two talking dogs.
     As El Saffar sees it, each narrator's strange experience is a fortuitous liberation from previous patterns of action that have caused his intense feelings of failure and alienation.¹³ Moreover, through the process of verbalizing their hallucinatory adventures, both Don Quixote and the Ensign actually begin the long journey to recovery from their recent moral and emotional afflictions. The critic convincingly argues that Cervantes must have decided at some late point (ca. 1612-13) to join the Coloquio to the Casamiento in order to duplicate, thematically at least, the Cave of Montesinos adventure.¹⁴ In this new arrangement, Campuzano's act of transcribing the dream-like colloquy is supposed to inspire his subsequent purgative confession to Peralta about his ill-advised Casamiento engañoso.
     But we are then left with a nagging technical question: why has Cervantes chosen to present the events of the combined stories out of chronological sequence, through the repeated use of “flashback” narratives? Regrettably, the El Saffar article is not concerned with such technical issues. For my part, I believe the solution can be found by examining a special quirk in Cervantes' style and his marked preference for one particular story-telling strategy. We have already observed that in the combined Casamiento and Coloquio Cervantes reprises the practice of constructing multiple levels of narration and criticism, a pattern he established originally in the Quixote. I would now like to suggest that the same device of layered critical commentary provides us with an important clue as to why Cervantes, in combining these two tales, chose to begin in medias res, rather than present the events in their normal chronological order.

     ¹² “Montesinos' Cave . . . ,” pp. 451 and 465.
     ¹³ “Montesinos' Cave . . . ,” p. 456.
     ¹⁴ “Montesinos' Cave . . . ,” p. 460. [In the original version of this article, this footnote appeared on p. 68. -FJ]

     The simplest possible linking arrangement would have dealt exclusively with Campuzano. We would have observed him first in the painful process of courting, wedding and losing the shifty Estefanía, and then at the hospital of the Resurrection during his hallucinatory descent into the realm of conversing canines. But such a bare-bones structure, featuring as it does a Coloquio with only three narrative levels (Campuzano overhearing Berganza and Cipión; Cipión listening to Berganza; Berganza witnessing Cañizares' reverie), would have made it difficult for Cervantes to point to the moral lesson that both the Ensign and the reader were supposed to extract from these two closely related experiences. The missing element, the catalyst designed to infuse moral exemplariness into the new scheme, is the sympathetic ear (and later, the acquiescing eyes) of Peralta. In the course of recounting to his companion the bizarre particulars of these two strange experiences, the narrating Ensign is obliged to confront the unpleasant truth about his previous personal comportment, draw the proper moral conclusions, and pledge himself to a reform of his personal conduct. Thematic considerations aside, the creation of the listening Licenciado can also be applauded as a sparkling demonstration of the author's technical virtuosity: Peralta's presence as a commentator on a much-desired fourth level of narration illustrates Cervantes' intense commitment to greater narrative sophistication and structural complexity in his novellas.

     To summarize, for some reason Cervantes had a fondness for —perhaps even a fascination with— the number four.¹⁵ In blending the Casamiento with the Coloquio Cervantes returns to a strategy he introduced earlier in the Quixote: the pattern of having his intercalated narratives take place on four distinct levels. As we have already seen, there is a quadrumvirate of narrators in the Marcela episode (Pedro, Ambrosio, Vivaldo, and Marcela herself); likewise, we observe four separate conspiratorial plans (caused by three sudden reversals in the plot) within the story of the Curioso impertinente; and we count a total of four narrative divisions in the love intrigue of Cardenio and Luscinda (Cardenio's story, Dorotea's tale, the climactic encounter at the inn, and Fernando's concluding epilogue). Also worthy of mention is the fourfold construction that Haley signalled in the Maese Pedro

     ¹⁵ A plausible psychological explanation of this preference can be found in El Saffar's Beyond Fiction: The Recovery of the Feminine in the Novels of Cervantes (Berkeley: University of California Press, 1984), pp. 10-11.

adventure and its undeniable link to the four narrative tiers Cervantes established in DQ I, 9, when he created the figures of Cide Hamete, the Moorish Translator and the Second Author / Editor. And to complete the listing we have the four critical planes I have discerned within the Cave of Montesinos episode. Cervantes recognized that by appending the dream-like Coloquio to the more realistic Casamiento he could create, thematically at least, a situation analogous to the one faced by Don Quixote in the Cave. He also realized that in order to maintain the four-tiered structure of that model and others from the Quixote, certain technical adjustments would have to be made. Most notable of these was the need to invert the chronological order of some of the events, e.g., to begin with the Ensign's emergence from the hospital and fuse the overheard Coloquio (which actually occurs before Campuzano's recovery) to the very end of the Casamiento through the mediating figure of the witnessing Licenciado. Such a rearrangement would then enable Cervantes to utilize the “flashback” device of which he was so fond and add the desired fourth critical level. Most of us today would agree that, all things considered, it was a most curiously pertinent and fortuitous decision on the part of the distinguished writer.