SciELO - Scientific Electronic Library Online

vol.15 issue1Between Accidents and Explosions: Indeterminacy and Aesthesia in the Becoming of HistoryLiterary Readers’ Education: Possible Dialogues between Conceptions of the Bakhtin Circle and Textbook Activities author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Bakhtiniana: Revista de Estudos do Discurso

On-line version ISSN 2176-4573

Bakhtiniana, Rev. Estud. Discurso vol.15 no.1 São Paulo Jan./Mar. 2020  Epub Nov 28, 2019 


Other Quixotes - Notes on the Discourse in Cervantes and Borges

Newton de Castro Pontes*

Edson Soares Martins**

*Universidade Regional do Cariri - URCA, Centro de Humanidades, Departamento de Línguas e Literaturas, Campus Pimenta, Crato, Ceará, Brazil;;

**Universidade Regional do Cariri - URCA, Centro de Humanidades, Departamento de Línguas e Literaturas, Campus Pimenta, Crato, Ceará, Brazil;;


This article discusses several of Bakhtin’s concepts regarding discourse in the novel genre (especially those found in his texts Epic and Novel and Discourse in the Novel) as a way to understand some discursive aspects of Miguel de Cervantes’ Dom Quixote. We then proceed to a comparison with the discourse in another literary work that partially reproduces Dom Quixote: the short story “Pierre Menard, Author of the Quixote,” by Jorge Luis Borges. Finally, we briefly discuss the problem of Dom Quixote’s almost absence in Bakhtin’s work about the novel genre, while presenting the hypothesis by Walter Reed that, as in Borges’ short story, Bakhtin’s theory itself also partially reproduces the Quixote’s particular way of thinking.

KEYWORDS: Heteroglossia; Dialogism; Miguel de Cervantes; Jorge Luis Borges


O presente artigo parte de várias considerações de Bakhtin a respeito do discurso romanesco (especialmente aquelas presentes nos textos Epos e romance e em O discurso no romance) a fim de compreender alguns aspectos discursivos do Dom Quixote, de Miguel de Cervantes. Procede-se, então, a uma comparação com o discurso de outra obra que reproduz parcialmente o Dom Quixote: o conto Pierre Menard, autor do Quixote, de Jorge Luis Borges. Por fim, discute-se brevemente o problema da quase ausência do Dom Quixote nas discussões de Bakhtin sobre o romance, apresentando-se a hipótese de Walter Reed de que, tal como no conto de Borges, a própria teoria bakhtiniana também reproduz parcialmente o peculiar pensamento do Quixote.

PALAVRAS-CHAVE: Heterodiscurso; Dialogismo; Miguel de Cervantes; Jorge Luis Borges

There were various reasons for celebrating Cervantes’ Don Quixote; two of them are practically above all the others: the language and the time. Two statements are necessary for this discussion: first, that Cervantes was a man of his time; second, that Cervantes is an author of his time - or, to refer to Bakhtin’s (1990b)1 terminology, the Cervantes that we are referring to is both an author-person and an author-creator. These statements may look identical, but they carry a fundamental difference: one of them speaks of ideas and ideals that were present in Miguel de Cervantes’ mind, the Spanish man who lived in Italy, enlisted himself in the Infantería de Marina in 1570, fought against the Ottomans in 1571 (taking three shots and losing the movement of his left arm), was imprisoned in Argel between 1575 and 1580, published the two volumes of Don Quixote in 1605 and 1615, died in 1616; the other refers to the figure of Cervantes that was created by more than 400 years of readings of Don Quixote, The Exemplary Novels and a few other texts, the Cervantes that exists as language, the Cervantes created by literature. Don Quixote is the result of the identity and the difference between those two Cervantes.

As a person of his time, Cervantes adhered almost naively to the patriotism and the Christian ideals exacerbated by Renaissance Spain in such a way that his beliefs never cease to sound sincere - be it in his military life or his literary one (none of his works attacks or tries to attack the cultural identity from that period). Thus, it is curious how the Cervantes author-creator is recognized as a much more critical, incisive figure: there are no antichristian or antipatriotic thoughts in Don Quixote (Cervantes person would not allow that), but the way in which Alonso Quijano’s journey is linguistically organized - its literary expression - is fully revolutionary. Without speaking a word against Spain or its costumes, the author succeeded in counteracting what the man believed; he succeeded in creating a literary version of Spain in which the relations between the people and their ideas are much more complex, more deeply problematic than what Cervantes himself may have thought. This contradiction between author-creator and author-person is an integral part in the configuration of the book and is what originated moments such as the Chapter XXXVIII of Part I, in which Don Quixote discusses the merits of the man of letters and the man of arms - and favors the second. A character in a literary work, and specifically a character that appreciates literature so much that he decides to conduct his life accordingly and gives a lecture that exalts the superiority of arms above letters - it all sounds like a suspicious eulogy and a celebration of Cervantes’ military career, which preceded his activity as a writer. It is almost as if the author-person tried to impose himself over the Don Quixote’s author-creator.

In theoretical terms, this conflict was noted by Georg Lukács in The Theory of the Novel:

Cervantes lived in the period of the last, great and desperate mysticism, the period of a fanatical attempt to renew the dying religion from within; a period of a new view of the world rising up in mystical forms; the last period of truly lived but already disoriented, tentative, sophisticated, occult aspirations.

It was the period of the demons let loose, a period of great confusion of values in the midst of an as yet unchanged value system. And Cervantes, the faithful Christian and naively loyal patriot, creatively exposed the deepest essence of this demonic problematic: the purest heroism is bound to become grotesque, the strongest faith is bound to become madness, when the ways leading to the transcendental home have become impassable; reality does not have to correspond to subjective evidence, however genuine and heroic (LUKÁCS, 1971, p.104).2

In his liveliness to build a serious-comic, parodic and realistic literature, certainly inspired by Renaissance thought and, in certain measure, by picaresque novels such as The Life of Lazarillo de Tormes [La vida de Lazarillo de Tormes y de sus fortunas y adversidades, Anonymous, 1554], Cervantes ended up fictionalizing (intentionally or not) the form of the main problem of his time: a past that refuses to accept the present, an attitude and a content thought as eternal, but that do not find any place in a new world of values and, as a way to preserve themselves, reject reality and try to become self-sustainable.

There is also much to talk about the language of the Don Quixote. In this novel, not only there are various linguistic extracts (of different idioms, social classes, age groups, professions, of written and spoken expressions), but also these extracts are directly opposed and clarify each other mutually. Unlike the chivalric novels that it parodies, the world of Sancho and Don Quixote is not one that can be told through a single elegant language and a courtly style (as would be the case of Chrétien de Troyes’ verses): to make its comic style work, it needs the misunderstandings caused by the collision of distinct linguistic worlds. Famously, Erich Auerbach pointed this in a chapter of Mimesis called The Enchanted Dulcinea, discussing how the scene of the encounter between the Quixote and a woman who Sancho convinces him to be Dulcinea, at the beginning of the second volume, is built on the misunderstanding between the erudite language of the courtly novel (the idyllic) and the peasant oral expression (the everyday): “In Cervantes’ case, the two realms of life and style clash by reason of Don Quijote’s madness. There is no possibility of a transition; each is closed in itself; and the only link that holds them together is the merry neutrality of the playful scheme of puppet-master Sancho [...]” (AUERBACH, 2003, p.351).3

Above all else, the language is deeply rooted in the temporal dimension of the novel. As Mikhail Bakhtin discusses in the essay Epic and Novel, which does not refer to Don Quixote directly but mentions Cervantes among the most important stages of the novel’s development (BAKHTIN, 1981b, p.36),4 the heteroglossia or polyglossia of the novel genre must be understood from its specific axiological and temporal axis: its zone of maximal contact with the present in all its open-endedness. It is not by accident that Bakhtin pointed out the difference between epic and novel using the discourses about the living and the dead as a point of comparison: “[The dead] are removed from the sphere of contact, one can and indeed must speak of them in a different style. Language about the dead is stylistically quite distinct from language about the living” (BAKHTIN, 1981b, p.20).5 The discourse about the dead is, par excellence, the solemn elegy that elevates them; but for the living are reserved the witty, satirical, critical discourses that lowers them to the level of the living materiality of their peers. While the epic refers to an absolute past, that from the author’s point of view is received as a monumental legacy of a concluded age (the aedo and his listeners are separated from that age and can only refer to it as something distant that does not allow critical reevaluations), the novel turns itself to a present that is closer to the future than to the past and, above all, to the time of the author himself and of the responsive audience that contemplates his works. This is the great mark of the novel’s chronotope.

The destruction of the epic distance in the novel was realized, among other things, by its use of the comic and parodic, which possess a familiarizing function: the comical lowers the legendary to the level of the everyday, eliminating the hierarchical distance between the hero’s life and the ones who share his story, while the parody turns its attention to the conventionalisms and artificiality of the official discourses and of the canonical and perfectly concluded literary genres, writing in them a new linguistic perspective that sees and criticizes their stylistic traits through the current, living uses of the everyday language. That is why, when Bakhtin enumerates what he considers structural and fundamental particularities of the novel, he refers mainly to time and language:

I find three basic characteristics that fundamentally distinguish the novel in principle from other genres: (1) its stylistic three-dimensionality, which is linked with the multi-language consciousness realized in the novel; (2) the radical change it effects in the temporal coordinates of the literary image; (3) the new zone opened by the novel for structuring literary images, namely, the zone of maximal contact with the present (with contemporary reality) in all its open-endedness (BAKHTIN, 1981b, p.11).6

Such a temporal-linguistic change also implies that, to the novel, the contemporary being is open-ended; it is not completely incarnated in the socio-historical substance of its time. As a consequence, there is a strong presence of ideologist heroes in the novel’s tradition, heroes who individually think about the values that they intend to build (instead of being a repository for the national legend). It is a new initiative of the character which is both ideological and linguistic - in the case of Don Quixote, it can be thought about how Alonso Quijano, when arms himself as a knight, calls upon not just courtly values but also discourses that belong to the chivalric literature, even when those do not coincide with the ones he experiences in his reality; also, it can be thought about how Sancho Panza, in stark contrast, resorts to popular sayings as a way to interpret reality. The individual ideas and ideals of each character coincide with particular uses of the language, and those uses are placed in perspective through the novel: one use situates the other socio-historically.

Another of Bakhtin’s essays, Discourse in the Novel, is one of the rare occasions in which the Russian thinker discusses Don Quixote directly. Already in his first mentions to the book, he emphasizes the heteroglossia found on it, besides the linguistic perspective created by the polemical coexistence between the forms of the elevated discourse (in a version idealized by chivalric literature, in particular the courtly novel) and the vulgar or even coarse forms of language used in 17th century Spain:

Cervantes excelled in describing encounters between a discourse made respectable by the romance and vulgar discourse - in situations fundamental in both novels and life. In Don Quixote, the internally polemical orientation of “respectable” discourse, vis-a-vis heteroglossia, unfolds in novelistic dialogues with Sancho, with other representatives of the heteroglot and coarse realities of life and in the movement of the novel’s plot as well. The internal dialogic potential embedded in respectable discourse is thus actualized and brought to the surface - in dialogues and in plot movement - but, like every authentic manifestation of the dialogic principle in language - it does not exhaust itself completely in them, and is not resolved dramatically (BAKHTIN, 1981a, p.384).7

Among other aspects of Cervantes’ novel, Bakhtin notes the polemical dialogism between stupidity and intelligence, emphasizing that it is based on the incomprehension of the other’s discourse (similarly to what was proposed by Auerbach): the episode of Sancho’s governorship is directly mentioned as one of the most apt moments for the development of a diversity of dialogic situations (BAKHTIN, 1981a, p.403).8 Besides, he points out the encyclopedic potential of Don Quixote, the way it represents a multiplicity of languages from that era with the purpose of incorporating non-literary languages into the novel without any intent to literalize them; according to Bakhtin, the intention is, indeed, not to ennoble the extraliterary discourses (as would happen in elevated genres), but to represent a multifaceted and socio-ideologically varied reality: in other terms, the novel is built as an encyclopedic microcosm of the heteroglossia historically defined and materially configured through language.

Cervantes’ novel appears as one of the main works in which the novel genre “unfolds its fullest potential” and “The author encases his own thought in the image of another’s language without doing violence to the freedom of that language or to its own distinctive uniqueness” (BAKHTIN, 1981a, p.409).9 This results, according to Bakhtin, in a parodical potential of a special caliber in which the parodied language shows resistance to the other’s parodical intentions. When the language of the chivalric novels, especially Amadis de Gaula, is expressed by Don Quixote, its full meaning and historical situation is only revealed if this language is seen in the totality of the dialogues from Cervantes’ time - that means the courtly language, which was a stable expression of the dominant social class and appeared isolated from other discourses in the courtly novels, suddenly only has any meaning in the polemical and transformative heteroglossic wholeness of Don Quixote, in its contraposition to other discourses from the book.

The main question for Bakhtin is in how the discourses of others are taken by novels of the same stylistic line as Don Quixote: those do not appear intact in these works; in fact, they are submitted to an intonation that pertains to the author, through which they are reevaluated and parodied. It’s not a simple set of quotations from social discourses - or, referring to the linguistic cliché popularized by Julia Kristeva (1986, p.37),10 a “mosaic of quotations” -, but a response to them, which sets the properly dialogic and responsive aspect of the discursive genres in general and the novel in particular. The novel’s heteroglossia is not about a mere repetition, mention or quotation of those discourses, neither their reduction to a single literary discourse, but about the fact that there is an emotional-volitional inclination by the author in relation to them, in a way that the author’s will does not confuse itself with their original will - as the author responds to them, we recognize echoes of a dialogue that refers to preexisting forms of life and thought, and thus have their own intentions, conceptions and linguistic expressions; organized in a dialogue, they try to situate themselves and mutually clarify each other. The author’s response does not express itself in a single monologic discourse, but in the dialogic heteroglossia - his voice is not a language, but a literary system of languages. In Bakhtin’s terms:

The language of novels of the Second Line [to which belongs Don Quixote] is not one language formed genetically out of the mixing of languages, but, as we have emphasized many times, is rather a unique artistic system of languages, all of which do not lie on the same plane. Even when we exclude character speech and inserted genres, authorial language itself still remains a stylistic system of languages: large portions of this speech will take their style (directly, parodically or ironically) from the languages of others, and this stylistic system is sprinkled with others’ words, words not enclosed in quotation marks, formally belonging to authorial speech but clearly distanced from the mouth of the author by ironic, parodic, polemical or some other pre-existing “qualified” intonation. [...] To relegate all these orchestrating and distancing discourses to the unitary vocabulary of a given author, to relegate the semantic and syntactic peculiarities of these orchestrating words and forms to the specific semantics and syntax of an author, that is, to perceive and describe everything as linguistic features belonging to some unitary authorial language, is just as absurd as blaming the language of the author for the grammatical mistakes he has employed to flesh out one of his characters. An authorial emphasis is present, of course, in all these orchestrating and distanced elements of language, and in the final analysis all these elements are determined by the author’s artistic will - they are totally the author's artistic responsibility - but they do not belong to the author’s language, nor do they occupy the same plane (BAKHTIN, 1981a, p.415).11

As it is not reduced to a quotation of other discourses, it can be inferred that the author’s consciousness also has a style of its own. It is necessary, however, to note that such a style, as it is supported by the compositional structure and themes of the discursive genre in which it is found, leads us to think that the author is both derived from the author-person and a product of the vicissitudes of the genre itself, in which specific discourse orientation dynamics (the author-creator is responsible for this) are determinant to its stable structure. As a consequence, the literary system of languages configured in diverse planes or the mere stylization of the languages of others receive, as a surface effect of the expressive intonation of the genre, the impression of a proper intonation from the author that makes perceptible the unity that we call style.

Lastly, there is one more question brought by Bakhtin’s (1981a, p.413)12 text: the parodical discourse in Cervantes’ narrative does not turn just to other social and literary discourses, but also to itself and its own existence as a novel. As evidenced by the polemic between the author of the second volume and the author of the fake sequel, there is in the novel a discourse about the novel itself. This is a characteristic that stands out in Don Quixote and already appears clearly developed in a well known passage of the first part: when, in a creative metalinguistic game, it is revealed that the author is, in fact, just editor of a text of someone else. Concluded the book - but unfinished the adventure of Don Quixote -, this editor looks for another volume that continues the story and then finds the one written by an Arabian historian called Cid Hamete Benengeli. He proposes to a Moorish that he encounters in Toledo to translate it:

I withdrew at once with the Morisco into the cloister of the cathedral, and begged him to turn all these pamphlets that related to Don Quixote into the Castilian tongue, without omitting or adding anything to them, offering him whatever payment he pleased. He was satisfied with two arrobas of raisins and two bushels of wheat, and promised to translate them faithfully and with all despatch; but to make the matter easier, and not to let such a precious find out of my hands, I took him to my house, where in little more than a month and a half he translated the whole just as it is set down here.


If against the present one any objection be raised on the score of its truth, it can only be that its author was an Arab, as lying is a very common propensity with those of that nation; though, as they are such enemies of ours, it is conceivable that there were omissions rather than additions made in the course of it. And this is my own opinion; for, where he could and should give freedom to his pen in praise of so worthy a knight, he seems to me deliberately to pass it over in silence; which is ill done and worse contrived, for it is the business and duty of historians to be exact, truthful, and wholly free from passion, and neither interest nor fear, hatred nor love, should make them swerve from the path of truth, whose mother is history, rival of time, storehouse of deeds, witness for the past, example and counsel for the present, and warning for the future. In this I know will be found all that can be desired in the pleasantest, and if it be wanting in any good quality, I maintain it is the fault of its hound of an author and not the fault of the subject (CERVANTES, 2004).13

There are, then, two narrative levels: to the adventure of the editor, which occurs in the reality, is opposed the adventure of Don Quixote himself, found in books. The way they unify themselves is not simple: there is a linguistic barrier (the editor does not know the idiom and asks someone to translate it) and an ethnical one (he does not trust Arabians). This brief passage in Chapter IX also attracts our attention for interrupting the narrative and introducing, in a few phrases, some sort of textual criticism: therefore, not only the editor tells his own story and the one from Benengeli’s book, but he also criticizes the form of the text of the Arabian historian. Alonso Quijano goes out from his library to the world: he is a reader before an adventurer; it is natural that the editor of his adventure expresses himself as a reader, too.

All this takes us to another object of our discussion, one that deals with another reader and his libraries: the Argentinean writer Jorge Luis Borges. One of his most famous short stories, Pierre Menard, Author of the Quixote, published originally in Ficciones (made famous in English language through a collection named Labyrinths), imagines a character called Pierre Menard, eccentric writer that decides to write, word by word, the Don Quixote, but with a caveat: he does not intend to do it by reading and copying the work, but by recreating it from zero, which he partially achieves by reproducing the chapters IX and XXXVIII, besides a fragment of XXII, all from the first part.

The choice of chapters is not accidental: in Cervantes’ book, all of them refer to literary problems. Respectively, they correspond to the criticism of Benengeli’s text, the discussion about the superiority of the man of arms above the man of letters and the dialogue with Gines de Pasamonte, who is writing a book about his own life but declares that the book is obviously unfinished as his life has not ended yet - a discussion that practically precognizes Bakhtin’s (1990b)14 considerations on the necessity of an exotopic perspective in autobiographical genres, the capacity to look at oneself from the surplus of vision given by the other as the vision of an I to itself is not capable of concluding this I in an aesthetically finished whole.

The reference to these almost theoretical moments from Cervantes’ novel, in which the narrative makes room for digressions about literature, coincides with the fact that the short story itself is also not narrative (for the most part): instead of just telling the adventure of the writer, Borges replaces the narrator for a literary critic who examines the version of Don Quixote created by Menard. Most importantly, there is the recurrence of the aforementioned Chapter IX, which is even quoted directly; however, while the moment of criticism is insubstantial in Cervantes’ book, Borges’ text focuses exactly on that: not on the plot of Don Quixote, but on the critical impressions of someone who reads the story. Therefore, it is not a short story about Don Quixote, but about its readings.

It is apt, then, to ask: what is the orientation of the discourse in a short story that, utilizing both epic-narrative forms and rhetorical forms of theoretical-philosophical genres, remains remarkably different from the novelist tradition that Cervantes helped to define and from the essayistic tradition that also serves as its model? An example may be found in the following excerpt:

It is a revelation to compare Menard’s Don Quixote with Cervantes'. The latter, for example, wrote (part one, chapter nine):

. . . truth, whose mother is history, rival of time, depository of deeds, witness of the past, exemplar and adviser to the present, and the future's counselor.

Written in the seventeenth century, written by the “lay genius” Cervantes, this enumeration is a mere rhetorical praise of history. Menard, on the other hand, writes:

. . . truth, whose mother is history, rival of time, depository of deeds, witness of the past, exemplar and adviser to the present, and the future's counselor.

History, the mother of truth: the idea is astounding. Menard, a contemporary of William James, does not define history as an inquiry into reality but as its origin. Historical truth, for him, is not what has happened; it is what we judge to have happened. The final phrases-exemplar and adviser to the present, and the future's counselor-are brazenly pragmatic (BORGES, 1964, p.55; emphasis in original).15

The direct discourse is employed in the text in which we see a phrase from Cervantes and another from Menard, but it can hardly be said that they are speeches from those characters. They are quotations, and this gives them a completely different aspect. Although they express positions of the characters in relation to the world, the fact that they were carefully chosen to be compared immediately stands out: they are there to prove a point of the narrator/critic of the short story, not the characters themselves. It is not unusual that a narrator in a novel offers a commentary about a character’s speech - Cervantes’ work itself does this -, but it is uncommon that a speech be chosen just to prove an argument, that instead of the sequence of lines be defined chronologically (one utterance comes after the other because it was spoken after it) or psychologically (the narrator’s memory organizes the order of the characters’ utterances), it is defined argumentatively: the critical and analytical purposes of the text are above any temporal perception.

It is then a rupture in the diegesis: narrator and characters are in distinct diegetic spaces; furthermore, the characters’ spaces are a diegesis to the narrator himself. If the discourse here is not as remarkably bound to its own references of internal structure, it is also not fully developed in the form of the discourse of the novel genre; in a way, the parody, working in Borges as a control of the stylistic nuance, constrains regressively the configuration of the discourse as a scenario of the dialogue: the fictional space in which the characters’ speeches are situated is not the same from the narrator’s/critic’s discourse, who converts the dialogue in edited quotations characteristic of a secondary genre less complex than the artistic form of the novel (the discourse of the literary criticism).

This was, of course, already present in Cervantes, but it is exacerbated in the Borges’ short story: to the editor of Don Quixote, Alonso Quijano was a person as real as himself (so much that he reads a book written by a historian, not by a fictionist); the does not question so much Benengeli’s text as he questions his personality - to him, Benengeli is a person. However, in Borges’ story, Pierre Menard is seen, most of the time, more as an author-creator than as an author-person - what is discussed is Menard’s textual expression while the person itself stays in the background. In a narrative, the difference between narrator and characters tends to be temporal (the narrator in the present and the characters in the past), which is true even to Don Quixote. But the Borges’ short story is beyond that: the characters are in an axiological-temporal axis entirely different; the entire world of their values and themselves is treated as a fiction by the narrator. The characters’ world is not interpreted by the narrator as a world of life that will be transformed in an utterance through his voice: he already receives that world as a discursive one, as an utterance; he does not refer to lives, but to discourses. The characters’ speeches can only participate of his life on a conceptual basis.

What was said until now applies to any essayistic or theoretical text, but we must remember that Borges’ text is a fiction thus possessing narrative features. First, the narrator is a fictional consciousness different from the author himself. Such a difference does not exist, strictly speaking, in a real theoretical text (author-person, author-creator and the voice that enunciates the text are the same individual; we assume that the propositions presented in the text are in fact the propositions defended by the real individual who wrote it), but it would be absurd to consider the real Borges (1), the fictional consciousness who organizes the text (2) and the narrator/critic (3) to be the same individual. Thus, as convinced as the narrator appears in the short story, the text as a whole (that is, as an expression of an author-creator) treats his propositions as verisimilitude, not as reality. This, of course, is more a contextual information than something given by the text itself (someone who did not know that Pierre Menard, Author of the Quixote is a short story by Borges, published in a book called Ficciones [Fictions], could maybe think that it is a real essay in a literary criticism publication). Anyway, the consequence is that the short story can be better defined as conjectural (a term much utilized by Borges himself) than theoretical, philosophical or essayistic: the text imagines credible propositions that work in the fictional world of the story but not necessarily in our reality. After all, in the Borges’ short story it is possible that someone rewrites, letter by letter, part of Don Quixote without even having read it; therefore, while a theoretical text would go from a phenomenon from our reality to defend a set of ideas, Borges’ text suggests one that could not work in our reality, and that is why it is necessary to build a world that serves to support it - the narrative is necessary so the proposition of the narrator/critic can exist. The relation with reality is almost inverted from the one found in Don Quixote: instead of starting from the real heteroglossia to build a literary system that becomes its microcosm, Borges starts from the literary proposition to create an emulated reality.

Besides, returning to what we discussed previously about the author-creator as derived not only from the author-person but also from the genre in which he is expressed, the words are clearly removed from Borges’ lips as a parodical stylization of the discourse in literary criticism. The system of languages of the author-creator certainly takes on discursive elements of Borges himself and his living context, but its main feature in the short story is the stylization of elements from a discursive genre derived from literary criticism activity (among them, the analysis of literary quotations, the recourse to biographical data, the contestation of previous critical enunciates and the resorting to theoretical sources as a way to explain an aesthetic-literary phenomenon) and of the novel genre (the quotations of Cervantes’ text), which are then submitted to an ironic intonation that polemizes with them and to the conventions of another genre (the short story).

The comparison between absolutely equal quotations, analyzed in a completely distinct way (one of them is recognized for its supposed deepness and complexity, while another is treated as inferior - though they are typographically equal), expresses a specific attitude in relation to literary criticism - in a certain measure, that attitude tries to subvert the conventions of this genre. The main question is that this kind of authorial attitude should not be understood as an eccentricity by Borges, but as something that he accomplishes inside a specific genre - the short story - and that would have another aspect if it was realized inside any other genre.

As for the text itself, the characters’ speeches that are interpreted by the narrator as coming from the world of culture are interpolated with those that come from his world of life (though fictional): there is the Pierre Menard author-creator, criticized/analyzed/interpreted by the narrator, but there is also the Pierre Menard author-person, of whom the narrator remembers, about whom he reports and with whom he shares the same world of values. “The Quixote-Menard told me-was, above all, an entertaining book; now it is the occasion for patriotic toasts, grammatical insolence and obscene deluxe editions” (BORGES, 1964, p.55),16 says the narrator. The discourse here is also direct, but expresses a much closer relation between narrator and character: in this case the difference is temporal, they are both inserted in the same axis, the same diegetic field, and none of them is abstractly reduced to a discourse. The same could be said about the paragraphs that narrate Menard’s work to create his Quixote, his project of becoming Cervantes and the following abandonment of such a project. There is, then, a contact that is not just conceptual, but also emotional: the images of the critic and the narrator switch themselves in the short story’s utterances; concept and memory are both tools to build the discourse. The criticism of Menard’s text walks along his biographical narrative.

Most importantly, however, is that neither in Menard’s biography (the narrative about him) nor in the criticism of his Don Quixote (the critical-essayistic element) we can fully find the meaning of the short story. Both dimensions are submitted to conjecture, that is, the imaginative extrapolations that the narrator-critic accomplishes from the essay and the narrative - thus his discourse is neither in the narrative past nor in the abstract atemporality of cognition, but truly in the fictionally possible future of conjecture: the theoretical conclusions and emotional impressions flow into the opening of new worlds only possible through thought. The analysis of the work and biography of Pierre Menard ultimately extends itself to other authors and possible works:

Menard (perhaps without wanting to) has enriched, by means of a new technique, the halting and rudimentary art of reading: this new technique is that of the deliberate anachronism and the erroneous attribution. This technique, whose applications are infinite, prompts us to go through the Odyssey as if it were posterior to the Aeneid and the book Le jardin du Centaure of Madame Henri Bachelier as if it were by Madame Henri Bachelier (BORGES, 1964, p.56).17

Departing entirely from the author, such a fictional time is fully submitted to his consciousness: his relation to conjecture cannot be polemic in the way Bakhtin considered the set of relations in the novel genre. Or, more specifically, the author-creator’s discourse polemizes with extraliterary discourses (like the literary criticism that it parodies), but not with his own text. The relations between narrator and characters or among the characters themselves may even be marked by a polemical aspect, as it can be noted at the very beginning of the story: “Unpardonable, therefore, are the omissions and additions perpetrated by Madame Henri Bachelier in a fallacious catalogue which a certain daily, whose Protestant tendency is no secret, has had the inconsideration to inflict upon its deplorable readers [...]” (BORGES, 1964, p.49).18 It is concerning the author, however, that such a polemic is not developed, for the author has become the center of the meaning in the short story (instead of the hero, as it usually is the case in the novel). The narrator and characters are the expression of his conjecture, that is, they are themselves a materialization of his set of credible propositions. It should be noted, then, that the main problem are not the propositions given by the characters - it is the fact that the characters themselves, as a whole, are an expression of those own propositions: the possibility that Pierre Menard exists in that world is, after all, the fictional materialization of an anachronic reading of literary texts. Pierre Menard is a vessel for a conjecture made by the author.

All this is made more obvious when compared to that Chapter IX of Don Quixote: the authorial perspectives of Cervantes in relation to his culture, to history and to Arabians are profoundly problematic and expressed through conflicting discourses coming from the most varied socio-cultural spheres; it chooses to represent not the deepness of the conjecture that comes from the author, but the complexity of the reality that is beyond him. It should be considered what Bakhtin said about novelistic heteroglossia while thinking about humoristic novels, which is the case of Cervantes’ work:

As we have said above, the narrator’s story or the story of the posited author is structured against the-background of normal literary language, the expected literary horizon. Every moment of the story has a conscious relationship with this normal language and its belief system, is in fact set against them, and set against them dialogically: one point of view opposed to another, one evaluation opposed to another, one accent opposed to another (i.e., they are not contrasted as two abstractly linguistic phenomena). This interaction, this dialogic tension between two languages and two belief systems, permits authorial intentions to be realized in such a way that we can acutely sense their presence at every point in the work. The author is not to be found in the language of the narrator, not in the normal literary language to which the story opposes itself (although a given story may be closer to a given language) - but rather, the author utilizes now one language, now another, in order to avoid giving himself wholly to either of them; he makes use of this verbal give-and-take, this dialogue of languages at every point in his work in order that he himself might remain as it were neutral with regard to language, a third party in a quarrel between two people (even though he might be a biased third party) (BAKHTIN, 1981a, p.314).19

In the way Bakhtin comprehends the novel configuration, the author does not own a proper language: it should be remembered that his tongue is a “system of languages” mimetized from social heteroglossia, acting as a response to it. The author’s inclination can only be recognized by his emotional-volitional attitude in relation to this heteroglossia, which has, by its turn, its own resistances to the authorial intentions (after all, the social languages themselves are responsive). Consequently, the author’s voice does not appear directly in the text (it cannot be confused with the figure of the narrator): it can be recognized by how the various discourses are selected and presented, but per se it is “neutral with regard to language” and is enshrouded by the plurality of discourses that make up the literary work.

Let us think, now, about Borges’ short story. The author-creator in this case is, evidently, a philosophically erudite writer of fictions and the short story is his proposition: the whole intention recognized behind the story is the expression of a philosophical-literary conjecture directed to a responsive receiver who shares the same references as the author himself. The narrator is almost exactly like him - the philosophically erudite writer of an essay about literature that casts his own propositions after his commentary about the life and works of Pierre Menard. The characters keep reproducing this model: Pierre Menard and Madame Henri Bachelier are philosophically erudite writers that cast their own propositions. This kind of reproduction is consistent in Borges’ works - we are reminded of the poem Chess, in which the kings at war are moved by players that are moved by a God that may be moved by a higher God. What it means in this case is that, in its whole or in details, the short story propagates in different levels the same authorial discourse - an erudite discourse that walks a line between literature and philosophy. There is not an author invisibly expressing himself through various conflicting social discourses - what is there is an author visibly and carefully building a literary and conceptual world through a language adequate to such a task, a language that is reproduced at various levels in the short story. The novel’s polemical heteroglossia that creates an ambiguous image of the author is replaced by the constant reproduction of an essayistic language that serves literary purposes.

It is not the case of considering the presence of a monologic discourse such as the one Bakhtin saw in the strictly poetic genres (in which the voice of the author and the lyrical subject coincide and supplant any opposite voices). In Borges’ short story, unlike those, the narrator and characters do have differing voices and they dialogue with each other. The question is that they are all oriented a priori by the same ethical initiative in relation to the world: one that, starting from the discourses about the world, conjectures situations in which those discourses can be extrapolated - therefore, they are all reproductions of the author-creator’s image. Furthermore, the fact that the narrator can situate them in another axiological-temporal axis, that he can treat their speeches as texts - as discourses that, once apprehended by cognition, are no longer part of those characters’ own world of life - make room for turning those speeches into a vessel for the narrator’s propositions. After all, that is how the same line (present in Cervantes and Menard) can unfold itself and get two entirely distinct interpretations: it is after being apprehended by the narrator and examined by his consciousness that those lines differentiate themselves and are transformed in distinct discourses in the short story. The necessity for such a comprehension through the narrator’s consciousness indicates that it is him who fulfills the meaning of the voices of those characters - this is what prevents them, in certain measure, from polemizing against him.

None of this diminishes, evidently, Borges’ work in relation to Cervantes’; it is just about comprehending that, belonging to another genre (the short story), his work has a different ethical initiative, another attitude regarding reality and another mode of discourse stylization. While Cervantes’ novel tried to parodically represent the reality of the world of life, multifaceted and conflicted, as a joyous game, Borges’ short story turns itself to the more stable reality of the world of culture, particularly of the essayistic-philosophical universe of the written word from which it tries to project other realities, fictional ones (attitude that we have been naming conjectural).

We have emphasized, until now, the contribution of Bakhtin’s theory to the comprehension of two expressions of Don Quixote, in Cervantes and Borges. We could not finish this discussion without mentioning the disconcerting fact that, while Borges is entirely missing in the writings of the Russian philosopher, Cervantes’ Don Quixote itself had only a very restrict presence in them - even though that novel coincides so well with Bakhtin’s theoretical propositions and is situated right in the period that he considered determinant in his studies (the European Renaissance). How should be understood the almost absence of such a seminal work from the history of the novel in Bakhtin’s so elaborate and rich theory of the novel?

The article The Problem of Cervantes in Bakhtin’s Poetics, by Walter Reed, suggests a few solutions. At the time it was published in 1987, Reed (1987, p.29) noted that, in more or less 1200 pages of Bakhtin’s texts translated to English until then, only twelve were dedicated to the study of Don Quixote. However, that should not be interpreted necessarily as an absence of the book. To Reed, the question is that Don Quixote is an important work not only because it is prototypical of the entire novel genre, but also prototypical of the theories of the novel themselves. That is why he considers that Ian Watt’s studies on the novel’s formal realism are lineal descendants of Sancho Panza’s attempts to counter-balance Don Quixote’s romancing idealism; or that the triangular desire theory by René Girard is derived from the antagonistic mimesis of the Bachelor Sampson Carrasco; or that Wayne Booth’s prescriptive rhetoric of fiction has its origins in the inquisitorial proceedings of the priest and the barber at the beginning of the first part of the novel; or yet that the historical-philosophical theory of Lukács finds its prototype in the speech of the Canon of Toledo at the Chapters XLVII and XLVIII of the first volume (REED, 1987, p.35).

But Bakhtin’s case is different from the others because it is reflected in the activity of Don Quixote himself, hero of the novel, especially in the way he theorizes his practice and Bakhtin practices his theory: as the chivalric ideal expands and envelops everything in the mind of Don Quixote, the Bakhtin’s concept of novel also expands vastly and absorbs genres that, to other critics, would seem completely distinct - in fact, Bakhtin (1981b, p.6)20 himself recognized this phenomenon in his theory and named it “novelization.”

Like Cervantes’ hero, Bakhtin thought literature and life in combative terms too - we must remember the tension between letters and arms in the Chapter XXXVIII of Don Quixote, first part, and the tension between “art and answerability” present since the very first texts of Bakhtin (1990a, pp.1-3).21 Reed (1987, p.36) considers especially important the way that Bakhtin presented problems related to representation and truth as problems of aesthetic wholeness and ethic answerability, echoing the way that Don Quixote himself would build his world (in contempt of his critics, like the Canon of Toledo) in a singular mixture of truth and fiction. The thinking of Don Quixote would work, then, as a horizon for Bakhtin’s theory - involved in the interior of that thought, the theory would be incapable of contemplating it externally, of containing it (as it lacks the necessary surplus of vision).

This takes us back to Borges. Reed (1987, p.37) writes that “[l]ike that other eccentric cervantista of the twentieth century, Borges’ Pierre Menard, Bakhtin demonstrates the centrality of Cervantes’ great fiction more by existential homage than by essential reflection.” Similarly to the impossible task of Pierre Menard, Bakhtin’s theory did not talk enough about Don Quixote so it could, in a certain sense, become himself.

1BAKHTIN, M. Author and Hero in Aesthetic Activity. In: BAKHTIN, M. Art and Answerability: Early Philosophical Essays. Edited by Michael Holquist and Vadim Liapunov; Translation and Notes by Vadim Liapunov; Supplement translated by Kenneth Brostrom. Austin: University of Texas Press, 1990b, pp.4-255.

2LUKÁCS, G. The Theory of the Novel: A Historical-philosophical Essay on the Forms of the Great Epic Literature. Translated by Anna Bostock. London: The Merlin Press, 1971.

3AUERBACH, E. Mimesis: The Representation of Reality in Western Literature. Fiftieth-Anniversary Edition. Translated by Willard R. Trask; introduction by Edward W. Said. Princeton: Princeton University Press, 2003.

4BAKHTIN, M. Epic and Novel. In: BAKHTIN, M. The Dialogic Imagination: Four Essays by M. M. Bakhtin. Edited by Michael Holquist; translated by Caryl Emerson and Michael Holquist. Austin, TX: University of Texas Press, 1981b, pp.3-40.

5For reference, see footnote 4.

6For reference, see footnote 4.

7BAKHTIN, M. Discourse in the Novel. In: BAKHTIN, M. The Dialogic Imagination: Four Essays by M. M. Bakhtin. Edited by Michael Holquist; translated by Caryl Emerson and Michael Holquist. Austin: University of Texas Press, 1981a, pp.259-422.

8For reference, see footnote 7.

9For reference, see footnote 7.

10KRISTEVA, J. Word, Dialog and Novel. In: KRISTEVA, J. The Kristeva Reader. Edited by Toril Moi. New York: Columbia University Press, 1986, pp.34-61.

11For reference, see footnote 7.

12For reference, see footnote 7.

13CERVANTES, M. Don Quixote de La Mancha: Vol. I. Translated by John Ormsby. Urbana: Project Gutenberg, 2004. Available at: Accessed on: Feb. 25, 2019.

14For reference, see footnote 1.

15BORGES, J. Pierre Menard, Author of the Quixote. In: BORGES, J. Labyrinths: Selected Stories and Other Writings. Edited by Donald A. Yates and James E. Irby; preface by André Maurois. 2. ed. New York: New Directions Publishing Corporation, 1964, pp.49-56.

16For reference, see footnote 15.

17For reference, see footnote 15.

18For reference, see footnote 15.

19For reference, see footnote 7.

20For reference, see footnote 4.

21BAKHTIN, M. Art and Answerability. In: BAKHTIN, M. Art and Answerability: Early Philosophical Essays. Edited by Michael Holquist and Vadim Liapunov; Translation and Notes by Vadim Liapunov; Supplement translated by Kenneth Brostrom. Austin: University of Texas Press, 1990a, pp.1-3.

Statement of Authorship and Responsibility for Published Content

We declare that both authors had access to the research corpus, participated actively in the discussion of the results and conducted the review and approval process of the paper's final version.

Translated by Newton de Castro Pontes -;


AUERBACH, E. Mimesis: a representação da realidade na literatura ocidental. Tradução George Bernard Sperber 5. ed. São Paulo: Perspectiva, 2004. [ Links ]

BAKHTIN, M. Epos e romance (sobre a metodologia de estudo do romance). In: BAKHTIN, M. Questões de literatura e de estética (a teoria do romance). Tradução Aurora Fornoni Bernardini et al. 6. ed. São Paulo: Editora UNESP, 2010, p.397-428. [ Links ]

BAKHTIN, M. Arte e responsabilidade. In: BAKHTIN, M. Estética da criação verbal. Tradução do russo e introdução de Paulo Bezerra. 5. ed. São Paulo: WMF Martins Fontes, 2011a, p.33-34. [ Links ]

BAKHTIN, M. O autor e a personagem na atividade estética. In: BAKHTIN, M. Estética da criação verbal. Tradução do russo e introdução de Paulo Bezerra. 5. ed. São Paulo: WMF Martins Fontes, 2011b, p.3-192. [ Links ]

BAKHTIN, M. Teoria do romance I: a estilística. Tradução, prefácio, notas e glossário de Paulo Bezerra; organização da edição russa de Serguei Botcharov e Vadim Kójinov. São Paulo: Editora 34, 2015. [ Links ]

BORGES, J. L. Pierre Menard, autor do Quixote. In: BORGES, J. L. Ficções. Tradução Carlos Nejar. 3. ed. São Paulo: Globo, 2001, p.53-63. [ Links ]

CERVANTES, M. Dom Quixote de la Mancha I. Tradução Aquilino Ribeiro. São Paulo: Difusão Europeia do Livro, 1963. [ Links ]

KRISTEVA, J. Introdução à semanálise. Tradução Lúcia Helena França Ferraz. 2. ed. São Paulo: Perspectiva, 2005. [ Links ]

LUKÁCS, G. A teoria do romance: um ensaio histórico-filosófico sobre as formas da grande épica. Tradução, posfácio de notas de José Marcos Mariani de Macedo. 34. ed. São Paulo: Duas Cidades, 2000. [ Links ]

REED, W. L. The Problem of Cervantes in Bakhtin’s Poetics. Bulletin of the Cervantes Society of America, v. 7, n. 2, pp.29-37, 1987. [ Links ]

Received: October 21, 2018; Accepted: September 13, 2019

Creative Commons License Este é um artigo publicado em acesso aberto (Open Access) sob a licença Creative Commons Attribution, que permite uso, distribuição e reprodução em qualquer meio, sem restrições desde que o trabalho original seja corretamente citado.