It is plausible that these observations may have been set forth at some time and, perhaps, many times; a discussion of their novelty interests me less than one of their possible truth.

Compared with other classic books (the Iliad, the Aeneid, the Pharsalia, Dante's Commedia, Shakespeare's tragedies and comedies), the Quixote is a realistic work; its realism, however, differs essentially from that practiced by the nineteenth century. Joseph Conrad could write that he excluded the supernatural from his work because to include it would seem a denial that the everyday was marvelous; I do not know if Miguel de Cervantes shared that intuition, but I do know that the form of the Quixote made him counterpose a real prosaic world to an imaginary poetic world. Conrad and Henry James wrote novels of reality because they judged reality to be poetic; for Cervantes the real and the poetic were antinomies. To the vast and vague geographies of the Amadis, he opposes the dusty roads and sordid wayside inns of Castille; imagine a novelist of our time centering attention for purposes of parody on some filling stations. Cervantes has created for us the poetry of seventeenth-century Spain, but neither that century nor that Spain were poetic for him; men like Unamuno or Azorín or Antonio Machado, who were deeply moved by any evocation of La Mancha, would have been incomprehensible to him. The plan of his book precluded the marvelous; the latter, however, had to figure in the novel, at least indirectly, just as crimes and a mystery in a parody of a detective story. Cervantes could not resort to talismans or enchantments, but he insinuated the supernatural in a subtle—and therefore more effective—manner. In his intimate being, Cervantes loved the supernatural. Paul Groussac observed in 1924: "With a deleble coloring of Latin and Italian, Cervantes' literary production derived mostly from the pastoral novel and the novel of chivalry, soothing fables of captivity." The Quixote is less an antidote for those fictions than it is a secret, nostalgic farewell.

Every novel is an ideal plane inserted into the realm of reality; Cervantes takes pleasure in confusing the objective and the subjective, the world of the reader and the world of the book. In those chapters which argue whether the barber's basin is a helmet and the donkey's packsaddle a steed's fancy regalia, the problem is dealt with explicitly; other passages, as I have noted, insinuate this. In the sixth chapter of the first part, the priest and the barber inspect Don Quixote's library; astoundingly, one of the books examined is Cervantes' own Galatea and it turns out that the barber is a friend of the author and does not admire him very much, and says that he is more versed in misfortunes than in verses and that the book possesses some inventiveness, proposes a few ideas and concludes nothing. The barber, a dream or the form of a dream of Cervantes, passes judgment on Cervantes. . . It is also surprising to learn, at the beginning of the ninth chapter, that the entire novel has been translated from the Arabic and that Cervantes acquired the manuscript in the marketplace of Toledo and had it translated by a morisco whom he lodged in his house for more than a month and a half while the job was being finished. We think of Carlyle, who pretended that the Sartor Resartus was the fragmentary version of a work published in Germany by Doctor Diogenes Teufelsdroeckh; we think of the Spanish rabbi Moses of Leon, who composed the Zohar or Book of Splendor and divulged it as the work of a Palestinian rabbi of the second century.

This play of strange ambiguities culminates in the second part; the protagonists have read the first part, the protagonists of the Quixote are, at the same time, readers of the Quixote. Here it is inevitable to recall the case of Shakespeare, who includes on the stage of Hamlet another stage where a tragedy more or less like that of Hamlet is presented; the imperfect correspondence of the principal and secondary works lessens the efficacy of this inclusion. An artifice analogous to Cervantes', and even more astounding, figures in the Ramayana, the poem of Valmiki, which narrates the deeds of Rama and his war with the demons. In the last book, the sons of Rama, who do not know who their father is, seek shelter in a forest, where an ascetic teaches them to read. This teacher is, strangely enough, Valmiki; the book they study, the Ramayana. Rama orders a sacrifice of horses; Valmiki and his pupils attend this feast. The latter, accompanied by their lute, sing the Ramayana. Rama hears his own story, recognizes his own sons and then rewards the poet. . . Something similar is created by accident in the Thousand and One Nights. This collection of fantastic tales duplicates and reduplicates to the point of vertigo the ramifications of a central story in later and subordinate stories, but does not attempt to gradate its realities, and the effect (which should have been profound) is superficial, like a Persian carpet. The opening story of the series is well known: the terrible pledge of the king who every night marries a virgin who is then decapitated at dawn, and the resolution of Scheherazade, who distracts the king with her fables until a thousand and one nights have gone by and she shows him their son. The necessity of completing a thousand and one sections obliged the copyists of the work to make all manner of interpolations. None is more perturbing than that of the six hundred and second night, magical among all the nights. On that night, the king hears from the queen his own story. He hears the beginning of the story, which comprises all the others and also— monstrously—itself. Does the reader clearly grasp the vast possibility of this interpolation, the curious danger? That the queen may persist and the motionless king hear forever the truncated story of the Thousand and One Nights, now infinite and circular. . . The inventions of philosophy are no less fantastic than those of art: Josiah Royce, in the first volume of his work The World and the Individual (1899), has formulated the following: "Let us imagine that a portion of the soil of England has been levelled off perfectly and that on it a cartographer traces a map of England. The job is perfect; there is no detail of the soil of England, no matter how minute, that is not registered on the map; everything has there its correspondence. This map, in such a case, should contain a map of the map, which should contain a map of the map of the map, and so on to infinity."

Why does it disturb us that the map be included in the map and the thousand and one nights in the book of the Thousand and One Nights? Why does it disturb us that Don Quixote be a reader of the Quixote and Hamlet a spectator of Hamlet? I believe I have found the reason: these inversions suggest that if the characters of a fictional work can be readers or spectators, we, its readers or spectators, can be fictitious. In 1833, Carlyle observed that the history of the universe is an infinite sacred book that all men write and read and try to understand, and in which they are also written.