It has become a commonplace of literary history to contrast Cervantes with the picaresque novels of his epoch, and to recognize in the allusions to and reflections of them in his works expressions of hostility. Since Américo Castro opposed the esthetic values of Cervantes to those of Alemán (in El pensamiento de Cervantes, 1925), much has been written to sharpen the contrast and to present it in terms which are not limited to literary devices, techniques, and characters, but bring out underlying differences of attitudes and human values. Many writers (myself included) have written of Rinconete y Cortadillo, the Coloquio de los perros, and the conversation between Don Quixote and Ginés de Pasamonte (DQ I, 22) as so many exercises in parody, or as criticism by doing what authors of picaresque might have done but failed to do. This essay has grown out of my talk that had as its title «Cervantes Deconstructs the Picaresque», and that focused particularly on the problems of the authority of the narrator.1 A recent book on the origins of the European novel, referring to the works of Cervantes «of a picaresque sort», says that they «function, like Quevedo's El Buscón, as sophisticated deconstructions —110→ of the ongoing novelistic series».2 These assumed oppositions deserve a closer examination.
The question of the authority of the text itself when it is produced as a first-person narrative is a function of the reliability or the trustworthiness of the narrator. This latter is an issue which has exercised critics since Henry James and Percy Lubbock made point of view central to the narrative discourse. The proliferation of autobiography, personal memoirs, diaries, and particularly political reminiscences has made us sensitive as never before to the bias of first-person narration, and so we find it more difficult than previous generations of readers, to accept any first-person fiction on its own terms.3 In our skeptical frame of mind we may fail to give due attention to the difference between the narrator who is honest within his limited angle of vision (e. g. Guzmán de Alfarache), and him who fudges the memory of his experiences in order to make a self-serving case (Lázaro). Are we being reliable readers? There is an evident risk that we will apply to earlier works of autobiographical fiction a rigor that they are not built to withstand. Their authors, and their readers worked within criteria of verisimilitude which satisfied them, so that the principal question was whether the rhetorical means had been used competently (one can reasonably ask whether Alemán, in his first and only novel, knew how to control it). If we fail to observe the difference noted above, and read all first-person narrators as unreliable or untrustworthy witnesses, rather than as a rhetorical means of presentation, we bring to fiction questions which should properly belong to history, not poetry, and end by confusing the two kinds of writing, and this in a period when the refinement of the epistemological distinction between history and poetry had acquired the greatest importance.4 Much modern criticism approaches —111→ its chosen texts in states of mind far removed from Keats' «negative capability», and the irritable (in the Keatsian sense) questioning of the narrator as to his bona fides, his prerogatives, his reliability and so on, is not always productive. Indeed, it may cripple our reading of his story, so that we look for the wrong kind of truth in fiction, like a Don Quixote in reverse.
In the argument that follows, I shall be only secondarily concerned with these questions of authorial and narratorial authority. My principal inquiry will be conducted into the prior question of Cervantes' relation to picaresque as a set of structural options, and in particular the common assumption that picaresque fictions could be seen, at Cervantes' historical moment, as a coherent genre. A genre implies a sizable group of works produced over a period of time, sharing formal and thematic characteristics and having, initially at least, a supporting ideology. I do not believe Cervantes could have seen those works which have come to be called picaresque in such terms, even if we leave aside the limited conception of genre that was part of the legacy of classical poetics. The encounter between Don Quixote and Ginés de Pasamonte is often read as Cervantes' rejection of picaresque, as a reductio ad absurdum of the autobiographical urge to tell all. In particular, Ginés' inability to finish his book because his life is unfinished seems to some readers to be a parody of autobiography's lack of formal control, the absence of such control being the result of another lack, that of an external perspective. But, as I shall ask later in this essay, is it really so clear that Ginés represents Cervantes' idea of the picaresque? Do the authors of picaresque fiction fail to frame their narratives? Why did Cervantes write stories which, if not picaresque, are a bricolage of picaresque formal and narrative devices? Even if we detect in these pieces an impulse to parody, we have to take into account the fact that parody involves a degree of complicity with its object, and a very evident intertextuality. What is parodied is incorporated, preserved, memorialized in the parody, even when the mimetic act of the parodist becomes a transfiguration. Just as Don Quixote would be impossible without Amadís and other books of chivalry, so Rinconete y Cortadillo, Coloquio de los perros, and —112→ La ilustre fregona would be inconceivable without Lazarillo and Guzmán de Alfarache.
It is now a quarter of a century since Carlos Blanco Aguinaga persuaded us to accept the judgment, expressed earlier by Américo Castro in El pensamiento de Cervantes, that there is a great divide between Cervantes and the writers of the picaresque.5 It would be pointless to chronicle here the vacillations among earlier literary historians in the presence of Rinconete y Cortadillo and the Coloquio, and to list the anthologies of La novela picaresca in which these works did or did not appear. What is important is that Blanco's article was decisive in convincing a generation of readers that Cervantes and picaresque were absolutely incompatible.6 The formalist typology of picaresque later worked out by Claudio Guillén, Fernando Lázaro Carreter, and Francisco Rico seemed only to confirm the necessary exclusion of Cervantes from the canon. So although my purpose is to try to understand a little better how Cervantes responded to the picaresque fiction available to him, we shall have to review Blanco's article and the generic model, since together they have had the effect of privileging Cervantes vis-à-vis the picaresque while limiting his freedom to appropriate its forms within any posture other than hostility.
It will be more practical to take up the question of genre first, and note how the widely influential model is constituted. Claudio Guillén's essay «Toward a Definition of the Picaresque»7 listed formal and thematic properties of picaresque: the protagonist-narrator as half outsider in a world he can neither embrace nor reject; progress from innocence to corruption; personal discovery of values, «as if by a godless Adam» (79); episodic structure; stress on the material level of existence. Then, Fernando Lázaro Carreter argued that a picaresque genre existed once Guzmán de Alfarache had incorporated and transformed —113→ structural elements derived from Lazarillo de Tormes.8 From that moment, a set of options existed for subsequent writers and constituted a paradigm: «la novela picaresca surge como género literario, no con el Lazarillo, no con el Guzmán, sino cuando éste incorpora deliberadamente rasgos visibles del primero...» (05). As Francisco Rico has indicated, a genre is not constituted by the first work of its kind, but only when characteristic structures of the model are discovered to operate with generative energy in other works which follow.9 Lázaro Carreter holds that the principal elements in Lazarillo which are developed by Alemán in the Guzman are:
- The autobiography of «un desventurado sin escrúpulos», narrated as a sequence of episodes;
- Articulation of the autobiography by means of a succession of masters, who also provide the pretext for social criticism;
- The whole offered as an explanation of the final dishonorable state in which the protagonist finds himself (206-7).
This list may be added to, or modified, and Lázaro Carreter sees the modifications as evidence of the genre's flexibility. Individual writers can develop this or that formal element, or vary the relation of narrator to reader, while maintaining the generic integrity: «La picaresca cesa allá donde sus motivos y artificios constructivos han dejado de ser operantes para el escritor, es decir, cuando dichos elementos han perdido fuerza generadora» (201). He claims also that, whatever novelties and variations may be introduced by successive writers, the system continues to cohere by virtue of a centripetal force: «Se siente tentación de ver lo que sigue a Alemán como una actividad destructiva, como haces de fuerzas centrífugas, pero no; compensándolas hay otras que tienden al centro y que mantienen la relativa cohesión del sistema» (228). Lo que sigue a Alemán... That, of course, is where Cervantes appears, and the next question would be, how does Cervantes respond to that system created by Alemán, when the latter transformed the structural elements of Lazarillo?
At this point we must bear in mind that Lázaro's essay expressed the inevitable formalist reaction against the heavy stress in academic criticism on definitions of picaresque by reference to content. But in —114→ reacting against the pursuit of moral and social definitions, he gave little attention to the fact that the formal elements are semantically variable, not fixed. Within the autobiographical form of Lazarillo and of Guzmán lie very different ideological structures, as different in their way as those in Manet's painting Le Déjeuner sur l'herbe and the sixteenth-century print on which it was modeled, or the similarities of poetic form and voice in Petrarch and Wordsworth. The meeting of Guzmán and Lazarillo, it must be stressed, took place not on an open field of literary history, but in the mind of each reader. And there is no reason to suppose that it always occurred as a peaceful and polite formal merger.
In practice, typical characteristics of the genre have been difficult to isolate and to stabilize. Before Lázaro Carreter offered his «Revisión», it had long been debated whether Lazarillo was the generic initiator, or merely a precursor, suggestive but infertile. Was the generic model a very brief story in which a young boy is raised in poverty and given away as a servant? Or was it the very long story in which a boy raised in luxury leaves home by his own choice? Was it the story of the first boy, now grown up and priding himself on his independence and his status, although he is really dependent on the favors of a sleazy archpriest for his livelihood, and for his wife? Or was it the second, where the narrator undergoes a religious conversion, and discovers that human independence is a mirage? The first narrator justifies himself by reference to a hypocritical society whose models he has assimilated; his discourse is its own mirror. The second sees himself justified by Christ, in whom he finally recognizes the giver of form to his life; his discourse is therefore self-recognition in the other.
It is scarcely surprising that historians of literature found these two works as difficult to accommodate to each other as they did to select one to be the prototype of picaresque narrative. The solution propounded by Lázaro Carreter and, following him, by Francisco Rico, has been (as we noted above) to place Guzmán over Lazarillo and to say that where the formal elements coincide, there is the nucleus of the genus picaresque. Yet, as we have just seen, the tensions between Lazarillo and Guzmán de Alfarache are as evident as the correspondences. The correspondences conceal the differences. To be more precise, the formal correspondences are the means by which the oppositions are both concealed and expressed. By taking such formal correspondences as constituting the genre, by assuming, that is to —115→ say, that they are stable for a series of works, one sacrifices manifest differences to a conventional notion of genre. If we were to describe La pícara Justina and La vida del Buscón as we have just described the two earlier novels, we would see that they, too, form a relation which is less one of intertextual confirmation than of disconformity. The peculiarity of picaresque is that these works which we understand to have been most original are most antagonistic to their predecessors in the way that they make their shared formal elements signify radically different signifieds. Since, as we said earlier, there is no fixed, one-to-one correspondence between formal units and their thematic functions, a writer can as well alter minimally the established conventional units in order to disrupt their referential system, as he can substitute very different formal units in order to enhance or protect the traditional truth value of his total discourse. Quevedo's Buscón is a clear example of picaresque narrative disconcerting the reader by presenting familiar signifiers (the formal units of autobiography: boy leaves home; ignominious parents; closing the circle) in combination with a different social perspective and in an imperfect series.
Now let us consider the essay of Carlos Blanco Aguinaga that was mentioned earlier as having had a profound influence on our view of the relation between Cervantes and the picaresque. Blanco identifies the picaresque completely with Alemán, claiming that he presents the picaresque traits llevados a un extremo absoluto» (p. 314). Not only is Guzmán the most representative, it is the most extreme case. Blanco sustains this argument at considerable length, and it serves him the purpose of opposing Cervantes to the picaresque twice over: as the champion of the free creative spirit against the monolithic genre, and the open imagination against the closed dogmatism of Alemán. This opposing of Cervantes to a narrow dogmatic Alemán and a monolithic genre (these two being identical) makes evident that the force of Blanco's argument depends on some rhetorical strategies which require closer examination.
At the beginning of the article, he states that Don Quixote and the picaresque have conventionally been credited with establishing the modern novel, because they break away from the dominant idealizing forms of fiction. Blanco objects to this conjunction of Cervantes and the writers of picaresque, on the ground that they are not realist in the same way; far from it, their respective realisms are «dos maneras contrarias de concebir la novela». These two modes of realism are —116→ totally irreconcilable, the «realismo dogmático o de desengaño» of Alemán, and the «realismo objetivo» of Cervantes (p. 313: Blanco's italics). Blanco assures us that this opposition will be demonstrated in the pages that follow. However, there is no discussion of realism as such, or of the appropriateness of this term, nor is realism related to representation, to mimesis, or to verisimilitude, though these are more apposite in the context of the sixteenth and seventeenth centuries. For the reader, in any case, the «dos maneras de realismo» and the absoluteness of the antagonism between them are given a priori, and what was to be demonstrated becomes part of the demonstration. So Blanco reveals this pattern: a universe of representation in which realism is opposed to idealism; within realism, a further split between two kinds which are not complementary but «absolutamente antagónicos». This tight, tense conceptual world is, as can clearly be seen, an adversarial one, and it is sustained, first, by assimilating all picaresque fiction to Guzmán, and second, by repeated use of absolutes: the picaresque world is «sólo vanidad y gesto»; it is the «más bajo y opuesto al idea»; picaresque is «siempre autobiográfico», the picaro «siempre un vagabundo», «siempre solo...» (my emphases added). The frequency of siempre, nunca, todo and related absolutes is particularly evident, in a procedure of cumulative assertion which precedes and largely replaces the demonstration. This procedure closely resembles the method of Guzmán in his narrative, as exposed and denounced by Blanco, moving «de la definición a lo definido» (pp. 316-17). In other words, a world of sharp contrasts, of fixed positions, of adversarial dogmatism, underlies and directs the argument and drives it forward. I would not deny that Blanco has some brilliant things to say about both Guzmán and Cervantes, but that does not affect my argument. My point is that they subserve a rhetorical strategy of pitting the «open», «objective» Cervantes against all of the picaresque as represented in the Goliath created by Alemán. This scheme is kept in place, as we have seen, by an underlying system of absolute identities and equally absolute antagonisms.10
I don't believe literary history is like that. The reader of fiction in the period 1600-1610 could have seen Alemán's vast and inescapably —117→ serious romance11 followed by Lazarillo, all but forgotten, and now reprinted with great success. Brief, laconic, even in its irony, it exemplifies a morality of survival by paddling with the current. To the many readers who came to it for the first time after Alemán and Martí, it must have seemed more like a torpedo than a precursor. Next come La pícara Justina, and the manuscript Buscón. These three works can be seen as both subverting and exploiting Guzmán in their different ways, as creative parodies, as parasites burrowing into and feeding off its great bulk in order to create something startlingly new, variously skeptical of Alemán's esthetic and ideological ambition, which had also been startlingly new, even if the moralities are not. The interesting thing is that Cervantes goes so much farther in the creative recycling and transformation of this whole repertory of devices and motifs which had only so recently been assembled and was just as rapidly being dismantled and recombined. If we think autobiography and the single focus are indispensable, here is Rinconete y Cortadillo with two boys presented by a third person narrator who, moreover, occasionally appears uncertain of his story. If Rinconete y Cortadillo drops the autobiographical mode of presentation, the Coloquio, retains it, but abandons the human subject, thereby «making strange», in Viktor Schklovski's phrase,12 both the narrative convention and the world that it discloses.
The encounter between Don Quixote and Ginés de Pasamonte (or Ginesillo de Parapilla, as the guard calls him) is so often cited as an example of Cervantes' rejection of Alemán's esthetics that it may —118→ seem idle to mention it once again. On further examination, however, the issue seems to me to be less clear cut. Cervantes' convict on his way to the galleys cannot but remind us of Alemán's, whose narrative is written in the galley where he has just earned a royal pardon. Both are notorious criminals. Both are writing their life stories. It will be recalled that Ginés explains with great pride how he has written his life «con estos pulgares», that it is so good that «mal año para Lazarillo de Tormes y para todos cuantos de aquel género se han escrito o escribieren», that «trata verdades tan lindas y tan donosas que no puede haber mentiras que se le igualen», and that it is not finished because his life is not yet finished. Here is an arrogant rogue who wants to shine as a writer, and so he claims that truth is better than fiction. Indeed, he wants to have it both ways, as we see in his assertion that his «verdades» are «lindas»: his no-nonsense claim on behalf of unvarnished facts over the «lies» of fiction, turns into the claim of the artistic attractiveness of unvarnished nature. There are differences between Ginés and Guzmán, of course, the most noticeable one being the fact that Ginés is on his way to the galleys to serve his second sentence, but this encounter is usually read as expressing Cervantes' dissatisfaction with the picaresque. So, Ann Wiltrout asserts that «with Ginés de Pasamonte, the perpetual outsider, Cervantes takes his most conclusive stand against the picaresque novel».13 Claudio Guillén is more cautious, observing what other commentators have not stopped to consider, namely that Ginés de Pasamonte is a reader, and in this episode Cervantes gives an encounter between two readers.14
One would like to know how Ginés de Pasamonte read Lazarillo de Tormes and the rest, including, presumably, Guzmán de Alfarache, in both its authentic and its spurious parts: whether as fiction («mentiras») or as true autobiographies. It is not clear whether the «mentiras» is a broadly dismissive word applied to other kinds of fiction, tall tales and the like, which are universally recognizable as such. It is tempting to suppose that he, like the mad Knight, has taken fiction for truth, but if Cervantes leaves this point unclear and arguable, it is because the resolution is not essential to our understanding. What does seem clear is that, in the scheme of pairings, matchings, and —119→ balances that Cervantes elaborates within Part One, Ginés de Pasamonte's relation to his favorite reading (picaresque) is homologous with the relation of Cardenio to his (amorous romance). The convict has read Lazarillo and Guzmán as desirable modes of action, as models for the conduct of a life. They must, then, have mediated to him a world of possible adventure, rather as amorous fiction had mediated to the young Cardenio, through its deceitful tropes, a world of romantic longing and anguished separation by which to conduct his desperate courtship of Luscinda («Vivía en esta mesma tierra un cielo...», Luscinda's parents «casi imitando... a los padres de aquella Tisbe tan decantada de los poetas» I, 282; «Lo que levantó tu hermosura han derribado tus obras: por ella entendí que eras ángel, y por ellas conozco que eres mujer» I, 274).15 Ginés de Pasamonte's «verdades» become «lindas» insofar as he succeeds in surpassing those models, first in his life, then in his writing. The attractiveness (for him) of his narrative derives from the assumption that there is an exact correspondence between the life and the narration, life becoming language as an act of will. Here we see the parallel with Don Quixote who narrated his first setting out in Part I, ch. 2, translating the act into the word, as his words determine his acts: «'Apenas había el rubicundo Apolo tendido por la faz de la ancha y espaciosa tierra... cuando el famoso caballero don Quijote de la Mancha, dejando las ociosas plumas, subió sobre su famoso caballo Rocinante, y comenzó a caminar por el antiguo y conocido campo de Montiel.' Y era la verdad que por él caminaba» (I, 94). Ginés, like Don Quixote, is both writer and reader of his life, he creates himself, looks upon his work and sees that it is good. So, in order to become his own ideal reader, he has eliminated the critic that every writer must nourish within himself. And like the Knight, he aspires to make his life total discourse, to abolish the difference between story and diegesis, between the teller, the telling, and the told.
Here the importance of Ginés' name becomes apparent. Is he really Ginés de Pasamonte, as he insists, or Ginesillo de Parapilla, as others declare? The name Ginés de Pasamonte, it is easily observed, bears a resemblance to that of Guzmán de Alfarache: note particularly the linking de; the same number of syllables (2 + de + 4); the same pattern of stress; the similar placement of the vowels a and final e. We may read this echo of Guzmán as parodic, but that does not take us —120→ far, since there is no frame of parodic structures built around him as there is around the figure of the Knight. The inference must surely be that Ginés has chosen to name himself thus. He has adopted a name evocative of the literary picaresque in order to incorporate himself into that world, just as the hidalgo Quijada (or Quesada, or Quejana) named himself Don Quixote in order to pass into the world of chivalry. Naming as an act of identification is more than appending an identity tag; it becomes an act of symbolic assimilation of the ideal represented by the name, so that to change a name is then to change more than a label, more even than status. Change of name is essential to rites of passage, and the change signals a transformation, a desire for new ontological definition. Thus, when the hidalgo's rocín becomes Rocinante, Cervantes tells us that the new name is «significativo de lo que había sido cuando fue rocín, antes de lo que ahora era» (I, 90), and when Aldonza Lorenzo is called Dulcinea del Toboso, the new name is, once again, «músico y peregrino y significativo» (I, 91). That is to say, her name is a sign which announces the semiotic system of chivalric romance, and draws her into it; and since that system is a Platonizing one, the name discloses an essence which she shares with the system, which was latent in her, and which has waited for just those magic syllables to evoke it. The bearer of the name will forever be known for the quality designated by the name, and Don Quixote can sally forth, confidently asserting that things are not what they appear, but are what their names (i. e. his names for them) evoke, which is of a piece with that 'other' more 'real' world. Ginés de Pasamonte's name is significativo in this quixotic sense. Indeed, it is fully quixotic, being not only significativo, but «alto, sonoro y significativo», like the name Rocinante, or «músico, peregrino y significativo», like the name Rocinante, or «músico, peregrino y significativo», like Dulcinea del Toboso.16
Claudio Guillén has argued that this episode represents Cervantes' rejection of first person narration:—121→
La vida de Ginés de Pasamonte is presented by its author, with the commissary's consent, as a truthful autobiography. Nevertheless, Cervantes stresses most explicitly the problem of narrative structure. A dramatic or epic character possesses, to be sure, some sort of identity; but how does one shape a «life»? The supposed proximity to «life» of the autobiographer is exacted at a very high cost: that of formlessness -and perhaps, as a consequence, of meaninglessness. Any life that is narrated by its own subject must remain incomplete and fail to achieve artistic unity or, very simply, the status of art. Narrative form demands a «second» or «third» person expressing a consciousness that is extrinsic to the sequence of events. Only such a consciousness can make possible the writing, in Aristotelian terms, of either «poetry» or «history».17
This reading, with its conclusion that an autobiographical mode of presentation is inadequate as narrative art is taken up by more recent commentators.18 But two questions can be raised here: does the «real» autobiography of Ginés stand for artless autobiography, or for autobiographical fiction in Cervantes' argument? and is a second or third «consciousness that is extrinsic to the sequence of events» really —122→ demanded by narrative form? Ginés' autobiography may be said to parody picaresque fiction only to the extent that it is unfinished, and authors of picaresque narratives always tell us that there is more to come. If it is a parody of fiction, then, it is so by virtue of that fiction's search for a kind of mimetic truth to life which is excluded by the well rounded story, the geometrical plots and climactic endings of romance. But whereas Ginés wants to hoard his pages until his life reaches its natural end, readers of picaresque narratives know very well that there will not be a continuation (though this knowledge has not stopped impostors and hacks from the attempt). The autobiographical fiction, in other words, simulates an abrupt ending, an arbitrary closure which, as an abundant critical literature shows, is far from being arbitrary. The ending of Lazarillo comes where it does because that is where its internal poetics demands that it should be. I do not believe that Cervantes was so tendentious as to let us believe he could not see the difference. The question of «how to shape a life» need not be of a different artistic order from that of third-person narration.19 The classic models, St. Augustine's Confessions and Apuleius' Golden Ass do achieve it. It is shaped in the first case by providing an internal pivot, the moment of conversion, which organizes past and present so as to generate meaning in the events, and to turn a life of waste into one of plenitude. Apuleius' life as an ass, of course, sets its own temporal bounds, as a significant portion of a life. Once the temporal segment is marked off, the writer is compelled to disclose an artistic necessity structuring it within formal bounds; not, of course, expressing the randomness of life, but allowing the reader to perceive that what are represented as life's vicissitudes are the vehicle of important determinants such as heredity, fate, divine providence, or other forces. Alemán's Guzmán looks back from a moment of conversion and re-reads his life as he narrates it, covering it with commentary. The «other» viewpoint called for by Guillén is contained within the book as one of the layers of consciousness. In Lazarillo also it is there, and not only «extrinsic to the sequence of events», but to the narrator, in that Vuestra Merced to whose point of view Lázaro is continually adjusting his narrative and, in particular, his prologue. These procedures are of the most artful artlessness, —123→ and quite the opposite of Ginés' naiveté, as the bibliographies of criticism of first-person novels from Lazarillo to Proust clearly indicate. A Life written while the life is in progress, concurrently with it (Ginés' way) will fail to rise to, or even to notice, the challenge of art. This cannot be said of Alemán who organizes his narrative from the Janus-like perspective of a climactic or visionary moment, from which the narrator can see himself as 'other' and from which structure is produced. Alemán does face the problems of representing the retrieval of past experience and of justifying the writing. For these reasons I find it irrelevant to regard Ginés' Vida as an attack on Guzmán's.20
Cervantes did not write first-person narrative except within a third-person frame. The picaresque did not gain his fervent participation as pastoral did, and as the long, intricate «Byzantine» romance also did. This much can easily be conceded. Yet, without Lazarillo and the others he would not have written Rinconete y Cortadillo at all, nor the Coloquio, nor La ilustre fregona as we now have it. In Joaquín Casalduero's phrase, «roza Cervantes el género picaresco sin querer entrar en él».21 In a slightly more analytic mode, Gustavo Alfaro distinguishes «la picaresca», («una actitud ante la vida que asociamos con el espíritu antiheroico y rebajador de los valores morales») from «lo picaresco» seen as «materia novelable».22 That materia novelable is integrated into his third-person narratives with great variation and subtlety, and makes possible new structures as well as new thematic combinations and new modes of judgment. Let us briefly review the picaresque motif «boy leaves home». In La ilustre fregona, Carriazo escapes from his noble and wealthy family to pursue the picaresque way of life at the tuna fisheries. Cervantes stresses two facts: first, that the boy left home of his own free will, by «inclinación picaresca», without compulsion; second, that he was not corrupted by that life but remained, in the well known words, «un pícaro virtuoso, limpio, —124→ bien criado, y más que medianamente discreto».23 How this oxymoron was received by Cervantes' contemporaries we cannot tell, but it is the signal of others to come, and through them is deeply implicated in the structure of the story. The narrator who informed us how Carriazo left home in obedience to an inner impulse which he could not resist, later introduces an aging nobleman who, many years earlier, had also yielded to an overwhelming impulse: he had raped Costanza's mother. This nobleman, it turns out, is the father of Carriazo. Costanza, the fregona, is his abandoned daughter, and she has shunned the occasions for revelry and loose living that the inn affords. Certainly, her innkeeper foster parents are uncommonly virtuous, as was her mother, so the question of sangre / crianza, heredity versus nurture, is left open, evenly poised but without comment. A similar case occurs in La gitanilla, where the stolen blue-blooded infant girl grows up among gipsies and acquires their talents for singing and dancing, is preeminent in piquant charm and ready wit, but is so virtuous that she will not permit a naughty word or gesture in her presence. There again, the force of 'blood' is sustained by the very special care taken in her upbringing by the old woman who stole her.
Cervantes' presentation of the origins of his characters and the relation of those origins to the present story has been taken as contrasting his openness, his preference for allowing characters to chart their own course and follow their own will and inclination, with «picaresque determinism».24 Yet, as we have just seen, Carriazo is the son of a father who exercised his inclinación in an act of shocking violence. When Lazarillo is put out as a servant, he has no choice. Cervantes tantalizes the reader with the possibility that freedom of choice may be limited by an inherited tendency toward deviant behavior, «breaking out» on the one hand, pulling against the acquired values of his class on the other. In this story, the pícaro virtuoso, limpio, bien criado can easily be read as the resultant of the two vectors, sangre (the pícaro) and the decorum of noble upbringing (limpio, bien criado) operating, evidently, within an individual personality.
The two boys in Rinconete y Cortadillo refer briefly to their escapes from home, which characterize them as cheerful teen-age delinquents. —125→ And here it is instructive to note that Cervantes' third-person narrator simply transmits what they say. The devious way of explaining and excusing by means of a few selective details, and couching the recital in high flown language and evasive euphemisms reveals Cervantes' debt to Lazarillo. But, contrary to what we might expect from reading commentators who oppose Cervantes' perspectivism to picaresque monocular vision, the narrator says nothing to correct the boys' self-presentation. Their change of linguistic register when they realize that they have no need to impress each other, and could work together to their mutual advantage is a shift of perspective which, it is true, can be conveyed most effectively by means of the third-person narrator, but in his function as tape recorder rather than as external reference point. Indeed, the principal purpose of the narrator in this story is not to make us see Rinconete and Cortadillo differently from the way they see themselves, but to enable a slice of life rather than a whole life-sequence to be narrated. Cervantes did not want to narrate a whole picaresque career, which requires that the life arrive at some critical moment or climactic event which will motivate the actor to become a writer. Instead, he presents the beginning of such a possible career and the first encounter of the young opportunists with institutionalized evil in the house of Monipodio, and leaves the reader to speculate on the relation between them. The narrator stands equidistant between the speakers, thus guaranteeing the true balance of the dialogue -assuming, of course, that he himself does not intrude to the extent that we may suspect bias, and that he does not give evidence of bad memory or defective hearing: all this has been well stated by others. But it is unlikely, I think, that one would find an evaluative distance between the narrator and the boys. Their cheerful insouciance, their fascination with the easy life, their ready wit, their perception of all life even in its most sordid aspects as a spectacle, is conveyed through the narrator with an exceptional transparency. In fact, the pleasure they take in acts of thieving and conning seems to have its counterpart in the narrator's amusement as he reports those acts. The way that Joaquín Casalduero (a critic of exemplary sensibility) has written about this novela confirms my impression: «En el mundo reinan la desconfianza y el engaño, pero es un engaño infantil con la trampa a la vista»; «Cervantes nos hace ver toda la limpieza y gracia con que Cortado quita un pañuelo al sacristán...»25 This section of Casalduero's chapter is —126→ called significantly «Alegría de Rinconete». Similar observations are offered on the events that take place in Monipodio's den: «A pesar del porte y la faz de Monipodio, a pesar de los bravos, tenemos la sensación de hallarnos en un mundo infantil, en un juego, en que la puerilidad de los jugadores les impide ver el engaño. Tan inocente es ese mundo que Rincón y Cortado lo dominan por completo».26 This is probably how Rincón and Cortado experience their world, their own misdeeds and those of others, and Casalduero has captured the impression with great clarity, because the narrator is transparent, or more accurately, his humor, his light facetious style and his nonchalant stance coincide with the boys'. He does not rectify anything in the telling. But I think this transparency of the narrator needs discussion. It is amusing to read of the boys' meeting, of their attempts to impress each other with formal address (señor gentilhombre; vuestra merced; señor caballero) and well turned phrases; it is entertaining to see them fleece the muledriver who thought he could fleece them; so far, this is standard comedy and does not touch the reader as a moral being. But their next act is to ingratiate themselves with some travellers who are going to Seville, and who let the boys ride most of the way, and just as they arrive, they rifle their hosts' baggage and run. There is a great qualitative difference to be observed between fleecing an over-confident adversary at cards, and robbing the people who have just given you hospitality or done you a favor. But the narrator's tone does not vary; there is no comment on this breaking of a fundamental taboo, and consequent advance in social unacceptability. The stance of the amused observer is not altered; but the careful reader will notice a slight backward glance of objective appraisal as the narrative presses on: «Habíanse despedido antes que el salto hiciesen de los que hasta allí los habían sustentado, y otro día...»27 They took leave, the narrator insists quietly, of those who were their benefactors, in case we had forgotten who their victims were. What Casalduero has overlooked as he notes the alegría, the boys' sleight of hand and the laughter, is that the narrator always makes us aware of the victim: los que hasta allí los habían sustentado tells us that the offence is greater than the mere theft of some shirts; later we see not only skilled picking of pockets, and the hilarious babble that keeps the —127→ victim's bemused attention, but the anguish of the sacristan whose church collection money was in the purse. Indeed there is laughter in the story, and laughter for the reader, and yet the narrator's evenness of tone in dealing with matters of increasing gravity invite us to rectify this manner of the telling.
We cannot separate the teller from the told, and once we look below the level of the discourse, what is 'told' is not just a series of tricks and deceits followed by a costumbrista episode in Seville and a tourist visit to Monipodio's place. It is the journey of the boys to Seville, their arrival and what they do and see. Seville was their goal from the beginning, and Seville, «gran Babilonia de España» was, of course, the picaro's paradise, the center of great wealth, and «crime city» for Spanish readers. The increasing gravity of events (which, we have observed, does not disturb the levity of the telling) -from word play to card play to abuse of trust- follows the road through the customs gate of the city, penetrating further into the city square, and finally into the enclosure of Monipodio's house which is the thematic center of the story. Given Seville's significance on the moral map of Spain, this house must be understood as its center. So the journey to Seville is the familiar journey to the underworld, but in a literal picaresque mode: from the freedom under the sky of Castile, in the open air all the way until they arrive at the center of Monipodio's operations. Their language finds its paradigms in Seville, for if they used language first for mutual deceit and later to bamboozle victims, Monipodio has a complete system of language which erases the values of the «upper»; world. Seeking picaresque freedom leads to the negation of freedom; facilis descensus Averni. At each level of analysis (linguistic, thematic, structural) we can see that Cervantes is representing a destiny implicit in an initial disposition. To choose the picaresque road is to choose Monipodio in the end. The narrator's smile and our laughter assure us of the smoothness of the journey and of the almost domestic banality of evil.
It has not been possible here to discuss any other aspect of Rinconete y Cortadillo than the use of third-person narration and its immediate effects. To sum up, first it enables Cervantes to avoid the problem of justifying his text (by what authority does this narrator make us read his life?) since third-person storytelling traditionally needs no specific defense. Among justifications for first-person narrative, Ginés de Pasamonte's Vida would be one extreme («I write it because it's mine») with Guzmán at the opposite pole («I write it —128→ because God moves me»). Then again, to be significant, autobiographical narrative requires a climactic event, a turning point, a retrospective analysis, a layered consciousness in the narration. Claudio Guillén argued that autobiographical narration risks formlessness, but in Rinconete we find Cervantes removing the usual props which hold up the fiction in a third-person mode. The presence of a consciousness «extrinsic to the sequence of events» has enabled him to present a small slice of a life (two concurrent lives, in fact) which may never have that climactic event or turning point that compels retrospection. He has done something much more extraordinary, which is to disclose significance in a beginning, not an end; and while narrating not the past but the present of his boys, he has created for them a virtual future.
If Rinconete y Cortadillo dispenses with autobiography and the climactic moment, with the revision (re-vision) of the past, the Coloquio de los perros explores just such a moment. Between the call to write and the last word being penned, Lázaro and Guzmán have felt their past take shape, pressed by the need to justify themselves. They have played the game of «telling all», which is also a game of reticences beyond the revelations. Lázaro tries to adjust his narrative to the expectations of Vuestra Merced. Guzmán tells his in a confessional mode, laying himself bare but then covering his nakedness with a sheet of moral and spiritual commentary: his judge will be God and the Wise Reader. Now, there is nothing of this in the Coloquio (the «confession» having already been displaced into the Casamiento engañoso), no call to speak either from an external authority, or from an internal witness. Neither of the dogs claims to have lived a life of exemplary or significant acts. In fact, neither of them writes anything or is aware that what they say will be written down by a third person. This puts them in the paradigmatic situation of characters in any third-person fiction, (or at this primary level, of any dramatic dialogue). But the situation imagined by Cervantes is even more radical than this because, as dogs, they have been dumb creatures until the moment when their dialogue begins. This means that they are responding not to a demand but to the fact of speech itself. Instead of self-defense, justification or apologia, we are given the question which is at once so simple and so absolutely overwhelming: I have this power to speak; what shall I say? I did not ask for it and cannot avoid it; it comes as an act of grace. Dialogue declares that if speech is the characteristic act of the self it becomes so by recognizing —129→ and responding to the other. In narrative the self organizes experience, and Cervantes' tale of the dogs is therefore an extraordinary insight into the fact that we do not need an event to make us structure our past.28 We do it all the time in our internal narratives, and in our dialogue with others, and in that debate with the critic within us as we speak and write. The relation Berganza-Cipión can be split many ways and viewed as self-other, speaker-hearer, narrator-implied reader, writing self-critical self.
These relations are transformed when the conversation is written down as the Coloquio by the alférez Campuzano and incorporated into his conversation with Peralta who reads it skeptically, and transformed again by us as we read it, knowing it to be fiction. At this point the Coloquio will be read as the response to a climactic moment, not in the life of Berganza, but rather in that of Campuzano. The dialogue becomes Campuzano's prophetic dream of life as a dog's world, prophetic in the sense of penetrating into the mystery of the world as it is. This dog's dream can frame Campuzano's own experience as exemplified in his Casamiento engañoso, and confirm that this life is, indeed, a dog's world. On the other hand, it could bear a message for Campuzano, as the visit to Monipodio's den bore a message for Rincón and Cortado, namely, the implications of his surrender to deceit and self-deceit for the world at large. We could then read the dream colloquy as the climactic moment in the life of Campuzano, which has enabled him to narrate his Casamiento engañoso as a confession to his friend.29 Does it justify him, because that's how life is, and we're all in it? Or does it alert him to the world he is helping to make, challenging him then to change himself for his and the world's sake? These are two sides of Guzmán's narrative, but Cervantes, once again, as in Rinconete, illuminates the moment of choice, experienced as —130→ a moment of doubt and a moment when the truth is not yet clear. For Campuzano, at any rate, the new beginning is modest enough: a reunion with an old friend, a meal, a talk, a shared experience, a story, going to church, a walk together in the open air among other people. No melodrama, but a rich texture of communication. And for Cervantes, communication means communion.
I conclude this very sketchy review of the question by insisting that we underestimate both the thinking and the varied art of Cervantes when we pit him in bitter ideological animosity against Alemán. Maurice Molho has contrasted Cervantes' «mundo abierto» with Alemán's «mundo irrespirable», and presented Rinconete as «un mentís personal a la problemática leamaniana».30 It is clear, however, that Cervantes was fascinated by all literary forms, and that whatever his distaste for preaching, «mezclando lo humano con lo divino» (DQ, «Prólogo») this new vogue was deeply pondered. All fictional stories offer a hypothetical situation and a question «What if X?» and we, as readers, must be careful to recognize that artistic rivalry is not always a matter of producing a truer or falser picture of reality, but a different reality. Lazarillo's anonymous author proposed a «What if...?» that supposes the internalizing of experience. Alemán's use of first-person narration matches experience and the bad conscience of the narrator. Neither author denies freedom; in fact the insistence with which Guzmán writes, in the early chapters, of his yielding not to 'blood' or family but to peer pressure and fear of losing face makes that book 'modern' in ways that Cervantes is not. For whatever reason of artistic judgment or personality, Cervantes preferred to work with problems of judgment and perception rather than those of awareness, conscience and retrospection. The place where Cervantes brilliantly outmaneuvers and thereby deconstructs the picaresque autobiography is in the location of authority. An author knows that his authority for writing fiction, however compelling he makes it appear, is but another fiction, a sham, though he conceals this from his readers. Unless, of course, his «What if ...?» contains some supposition about living a fiction, some figure who claims to be author of his reality, in which case the real author's strategies and evasions are inevitably reflected upon. But even the baring of strategies and evasions does not get down to the radical question of authority and the self-certainty of the writer, for which there is no —131→ certain ground. Can the writer of autobiographical novels hide behind the authority of the fictional narrator, whose claim is that he stands there to be judged? This is the question that Cervantes will not let rest, as he makes narrators doubt other narrators, makes his readers believe, doubt, question his narrators, and thereby acknowledge the absolute mastery, the absolute arbitrary authority of Cervantes.
My topic is made particularly slippery by the fact that both genre theory and literary history are in disarray. Implicit in this paper has been the assumption that we can no longer talk usefully about «the picaresque novel» as a well defined genre. In a very short period of time a small group of authors do remarkably different things with a small range of new materials and themes. Interaction is intense, and there is no common direction, so that to characterize a picaresque typology to which Cervantes was opposed would be to falsify his creative responses at different moments. Genres exist only in consensus, and since a consensus on what are the constitutive traits of a picaresque genre is impossible, it will be more practical to look in Cervantes' work for the new perspectives which he achieved by incorporating motifs, narrative points of view and social reference from these other works. All his best fiction is intergeneric, and we as readers have to begin by deconstructing that nineteenth-century invention, the picaresque, and the criticism that has kept it in place.31