Critics have long held that, even if Cervantes was at least somewhat aware that his work would be successful, this was only because he knew it was funny, and hoped that, in reading it, as he famously wrote in his first preface to Don Quixote, "the melancholy would be moved to laughter, and the merry made merrier still."
The concept of fiction that existed at the time, which stems from the Latin fingere, to fake or pretend, referred simply to any discourse that was not true. Literate Spaniards in the early seventeenth century knew about Aristotle's distinction between history and poetry mainly through Alonso López Pinciano's commentary, Philosophia antigua poetica, published in Madrid in 1595. In that book, which take the form of a dialogue, the philosopher explains the distinction in these terms, using the Spanish equivalent of the Latin fingere: "it is clear that if he created it [fingio] and wrote what he imagined, then it would be a poem, regardless if the thing was really happening, just as he is lying who says something is the case without knowing it, even if it really is true, because he presents it as true without knowing it."
Lopez Pinciano used the verb fingir to refer to the poet who makes something up even while it is true, or the liar who tells a lie not knowing that it is true, and in so doing his version of fiction remains aligned with Aristotle's poetry. Cervantes, however, does something entirely different with it. Where poetry remains imitation in general, and tragedy the imitation of an action that is serious, complete, and of a certain magnitude, Cervantes' fiction is the imitation of an imitation of an action. By thus doubling back on itself, his writing is able to cross the divide between history and poetry, as well as open up the space of individual perspective and interiority that we now associate with fiction.
Well versed in Aristotelian theory almost certainly through his familiarity with Lopez Pinciano, Cervantes makes frequent allusions to the problem of poetry and history in the many debates about literature he stages in his books. He grabs the bull by the horn, though, in the preface he penned for the 1613 publication of his collection of twelve Exemplary Novellas, in which he expounds on what he sees as his innovation, and on the very notion of exemplarity that his novellas are meant to embody: "these stories are my own, neither imitated nor plagiarized: my wit engendered them, my pen gave birth to them, and they are growing up in the arms of the printing press."
The previous novellas translated from foreign languages were, for the most part, those written by Boccaccio around the middle of the fourteenth century, along with his many imitators. Boccaccio certainly innovated the form of the novella as an extended narrative relating for entertainment's sake a series of events focused on a particular theme or example, endowing them with an unprecedented degree of attention to atmospheric detail. Nevertheless, Cervantes' take on the genre more than 200 years later would revolutionize it. Where Boccaccio's characters exhibit a rich pictorial realism in their actions, environments, and the behaviors they engage in, Cervantes' narratives function by constantly begging the question of the intent behind the description. We settle on a "realistic" description of giants as windmills or wineskins only as a function of offsetting or contrasting the "illusory" description of the same as giants. Likewise the overall tale of Quixote's adventures can be "realistic" only when we knowingly agree that such a provisional reality is framed by a ludicrous farce, namely, that the story we are reading is a true history recounted by an Arab historian and translated for our narrator by a market scribe.
Asserting their complete originality is tantamount, of course, to placing them squarely on the side of poetry in Aristotelian circles, and thus depriving them of the attribute of historical truth. To pass the scrutiny of the censors, though, they would then have to make a claim to some other kind of truth, namely moral truth, which Cervantes then asserts in the form of their exemplarity. As he writes, "if you consider it carefully, there is not one from which you cannot derive some profitable example, and did I not wish to avoid discussing the subject at too great length, perhaps I might show you the tasty and morally beneficial fruit that could be derived from all together as well as from each by itself." The novellas, Cervantes seems to be claiming, are both entirely fantastic, in that they were born exclusively of his own imagination, and yet have real pertinence for his readers' lives today. The key to unlocking this, as he calls it, "mystery" at the heart of the novellas is, he tells his readers, their careful consideration. They must read these completely original fruits of his imagination with a mind to how they could apply to their own lives; in order to see what lessons they can glean from them, they should read the stories as if they were the protagonists.
In the autobiographical poem he published around the same time, The Journey to Parnassus, Cervantes puts in a nutshell the relation between reality and fantasy that his writing inaugurates: "With my Novellas I opened the way for the Castilian tongue to reproduce a bit of fantasy with perfect naturalness." Natural fantasy. With these two words Cervantes shows that, contrary to the arguments generations of critics, he did indeed grasp the nature of his innovation. Fiction presents the untrue in the form of truth, poetry in the form of history, and in doing so, allows access to a different kind of truth. Fictions can be exemplary because the readers are invited into the fictional world as potential characters; they identify with the protagonists and judge their actions as if they were their own. But they are also exemplary in the way they reveal the illusions that blanket our lives, and the appeal that binds us to them.