«Don Quijote» and the Romances of Chivalry: The Need for a Reexamination

Daniel Einsenberg

The City College
City University of New York

Since Don Quijote was, above all things, a man who had read a great deal, it hardly seems possible to reach a satisfactory understanding of his personality without rereading some of his favorite books. Moreover, no satire can be adequately appreciated without a systematic study of the object it ridicules. Yet in recent years scholars have neglected the study of the Quijote in the light of the romances of chivalry that inspired Cervantes and his hero. Specialists in the romances of chivalry, such as Pascual de Gayangos and Sir Henry Thomas, have not considered themselves knowledgeable enough about Cervantes to attempt it. Cervantine scholars, on the other hand, have usually lacked access to the texts of the romances¹.

For our knowledge of the subject we must go back to the only study which claimed to be comprehensive, that of Diego Clemencín. Clemencín, who, for the moment, remains the person best acquainted with the romances of chivalry since the seventeenth century, began in 1833 the publication of his monumental edition of the Quijote, a project that was completed by friends after his death. In the notes which accompany his text is a vast amount of information about the romances of chivalry. As he believed one of the main functions of literary criticism to be the study of a work’s sources, he attempted to read as many as possible of the books Cervantes knew, including all those romances of chivalry he could obtain. Access to various private libraries made his project possible².

Francisco Rodríguez Marín has done much to deny Clemencín the position in Cervantine and chivalric criticism to which he is entitled. In his edition of Don Quijote, the most important one of this century, he criticizes in some detail, and sometimes with evident relish, Clemencín’s shortcomings, often those in the field of language³. This is not a serious fault; it is, after all, an essential part of criticism to note the mistakes of one’s predecessors. More disturbing, however, is that Rodríguez Marín not only does not add significantly to our knowledge of the romances of chivalry (which, as Director of the Biblioteca Nacional, he was well equipped to do), he takes, as it were, a step backward by not including in his notes many of Clemencín’s valuable comments. For example, when Don Quijote, upon taking his chivalric name Caballero de la Triste Figura, explains that he does so to be like the famous knights of old, who had similar names, «cuál se llamaba el de la Ardiente Espada, cuál, el del Unicornio, aquél, el de las Doncellas, aqueste, el del Ave Fénix, el otro, el Caballero del Grifo, estotro, el de la Muerte» (I, 19), Clemencín identifies the knights these names refer to⁴. This kind of help will not, however, be found in Rodríguez Marín’s notes, where there is only a comment on a change he introduced in the text. Twice in Don Quijote we find mention of Lirgandeo: in I, 43, where Don Quijote invokes him, along with Alquife, and in II, 34, where he is one of the figures in the procession at the ducal palace. Clemencín, but not Rodríguez Marín, identifies him as a sabio who appears in the Espejo de príncipes y caballeros. These are not isolated examples, but reflect a clear tendency on the part of Rodríguez Marín to give only the unavoidable minimum of treatment to chivalric material in his notes⁵.

To be able to evaluate Clemencín’s treatment of the romances of chivalry from a quantitative standpoint we need to establish how many romances of chivalry Cervantes was familiar with. At the same time we can discuss the extent of Cervantes’ acquaintance if we pause to consider how many romances of chivalry there were, a question that cannot be decided with certainty. Thomas’ chronological list at the beginning of Chapter V of Spanish and Portuguese Romances of Chivalry (Cambridge, 1920) includes 39, excluding Portuguese works and sequels of the same title. To this number we should add two works which Cervantes believed to be Spanish, although we know now they are not, Palmerín de Inglaterra and Tirante el Blanco⁶, and two works which Thomas was unaware of, Lidamante de Armenia, of Dámaso de Frías (1590)⁷, and Rosián de Castilla, of Joaquim Romero de Cepeda (Lisboa: Marcos Borges, 1586)⁸.

Of these, many are mentioned by name in the Quijote. A considerable number are discussed in the escrutinio de la librería: the founder of the genre in Spain, the Amadís de Gaula, as well as its offspring, the Sergas de Esplandián and Amadís de Grecia; Olivante de Laura, Lepolemo (El Caballero de la Cruz), Florismarte (for Felixmarte) de Hircania, the Espejo de caballerías, half Italian and half Spanish⁹, Palmerín de Olivia and his descendents Platir and Palmerín de Inglaterra, and Belianís de Grecia. Elsewhere in the Quijote we find reference to the Espejo de príncipes y caballeros (El Caballero del Febo [I, 1]), Cirongilio de Tracia (I, 32), Lisuarte de Grecia (II, 1), and the works of Feliciano de Silva (I, 1), by which we are to under stand the popular «dezeno» and «onzeno del Amadís», Florisel de Niquea and Rogel de Grecia¹⁰, not the earlier, less popular works which are today accepted as his¹¹.

It does not necessarily follow, of course, that because the name of a book or a character is cited, that that book has been read. In many cases, however, together with the names of the chivalric romances there is additional information which shows that Cervantes had at least a superficial, and in some cases a substantial, acquaintance with the book in question. For example, it is certain that Cervantes knew more of the Espejo de príncipes y caballeros than the name of its protagonist, since in the introductory sonnet from the Caballero del Febo he refers to several incidents in the work. The innkeeper in I, 32 tells several details about his books; Cervantes knew enough of Belianís de Grecia to know how fiery its protagonist was, and how miraculous the cures he received. Cervantes’ knowledge of Tirante el Blanco was so thorough that he remembered the insignificant character Fonseca¹².

The other alternative -if one assumes that Cervantes’ acquaintance with the romances of chivalry was slight- is to believe that he picked as the subject for a satirical work a type of literature he knew little or nothing about, and that to find the points he was going to make fun of he inquired among all his friends as to what struck them as particularly ridiculous about the romances of chivalry. Humorists do not work this way, at least not great ones; besides which, near the end of the sixteenth century, if one were to learn anything about the romances of chivalry, one had to read them for oneself. In literary circles, few people in Spain paid them the slightest attention.

It is also clear, even from those titles that are explicitly mentioned in the Quijote, that Cervantes’ interest in the romances of chivalry led him to investigate the subject seriously, and not to be satisfied with perusing those books that were easily obtainable. We see that he was acquainted with the later romances, such as Olivante de Laura, as well as the classics of the genre. That he knew Tirante el Blanco is, however, quite surprising, for the work was not popular in Castile, was never reprinted, and was soon forgotten¹³. Platir -as the priest remarks, an «antiguo libro»- reposed in similar oblivion.

We may well pause a moment to wonder how and where Cervantes was able to read these books, since he was of modest means, and they were not cheap; Don Quijote had to sell «muchas hanegas de tierra de sembradura» to obtain the money to support his addiction. As Don Quijote must have had trouble obtaining these books in La Mancha, no more a cultural center then than it is now, so Cervantes, even had he had the money, would have found it difficult to buy these obscure books of several generations before. All this leads to the suggestion that Cervantes might not have purchased the books himself, but rather read them in some collection, accumulated when the romances of chivalry were in their heyday. This would be even more likely if it is true that Cervantes «discovered» the romances of chivalry not, as many readers did, when young, to despise them in maturity, but when he was already middle-aged, and further removed from the height of the genre’s popularity.

Of the romances of chivalry mentioned by name in the Quijote, and which must therefore be examined first as possible Cervantine sources, there are no less than four which Clemencín was unable to examine. One of these, Platir, is exceedingly rare. The other three are also rare, but no more so than the other romances; they are Felixmarte de Hircania, Cirongilio de Tracia, and Feliciano de Silva’s Florisel de Niquea, Book X of the Amadís series. He does not conceal the fact that he was unable to locate copies of these¹⁴. He attempted to make up for it by reading many works which are not mentioned by name. He found evidence that Cervantes knew at least one romance of chivalry not referred to by name, and Rodríguez Marín found similar evidence for another. In the discussion which the canon from Toledo has with Don Quijote about the romances of chivalry, he states: «¿Qué ingenio, si no es del todo bárbaro e inculto, podrá contentarse leyendo que una gran torre llena de caballeros va por la mar adelante como nave con próspero viento, y hoy anochece en Lombardía y mañana amanezca en tierras del Preste Juan de las Indias, o en otras que ni las describió Tolomeo, ni las vio Marco Polo?» (I, 47). This is, as Clemencín correctly noted, an explicit reference to the romance Florambel de Lucea, first published in 1532 and reprinted in 1548.

Rodríguez Marín’s discovery is particularly striking because he made it by accident. While arranging books for a Cervantine exhibition, he happened to open at random a copy of Book IV of Clarián de Landanís, also a work never mentioned by Cervantes, and found in it no less than a Caballero de la Triste Figura, as well as a Caballero de los Espejos (one of the names used by Sansón Carrasco). Who knows what he would have found if he had read the book all the way through! As it was, he contented himself with perusing «una buena parte»¹⁵.

Although other romances of chivalry not mentioned in the Quijote might not yield such surprises, certainly the time has come to fill the holes in Clemencín’s work, and to make as complete a survey as possible of the entire corpus of Spanish romances, as we now know it¹⁶. It is, however, equally important to realize that much of the work which Clemencín did do cannot be said to be adequate by modern standards; little criticism of the early nineteenth century is. In many cases he was working under a handicap, in that he had to refer to books he had read and made notes on many years before, and which he could not readily consult again. Thus, we find notes like the following: «De la amistad de Alquife con Urganda, con quien vino a casar en segundas nupcias, se habla largamente, no me acuerdo bien si en la historia de Esplandián o en la de Amadís de Grecia»¹⁷. Clemencín also lacked many critical tools which we take for granted. He was unaware of problems of style, oral and written, so that we still know only through intuition the extent to which Cervantes (and Don Quijote) used chivalric language¹⁸. He was aware of plot only in the gross sense of the adventures Don Quijote undertook or underwent; minor episodes and exchanges, and the sources for them, he often did not discuss.

We should also note that Clemencín did not particularly like the romances of chivalry, and read them only out of dedication to the Quijote. He believed that Cervantes wrote the Quijote to banish the romances of chivalry, and comments at length on Cervantes’ apparent justification in doing so in the prologue to his commentary. It is not without significance that one of Clemencín’s longest notes elaborates on the «desaforados disparates» which the Toledan canon had said the romances of chivalry were full of. It begins: «¿De qué género los quiere el lector? ¿históricos, geográficos, cronológicos? ¿ponderaciones monstruosas, relaciones absurdas, desatinos contrarios a la razón, y al sentido común? De todo hay con abundancia en los libros caballerescos...» (note 34 to I, 47). This point of view led him to make a number of unfavorable and much-repeated comments on the romances of chivalry, such as his succinct condemnation of the Espejo de príncipes y caballeros as «pesado» and «fastidioso» (note 16d to I, 1), or his note on the many wounds which Belianís de Grecia received: «Sólo en los dos primeros libros de los cuatro de que consta, se cuentan ciento y una heridas graves, y probablemente son más las de los dos libros que siguen; pero no me ha alcanzado la paciencia para contarlas, y no ha sido menester poca para hacerlo en los dos primeros» (note 11 to I, 1). It probably affected his commentary in ways much more profound than this.

A number of discoveries about the Quijote made in the course of a preliminary sampling of the romances of chivalry provide further evidence of the need for a methodical study. One of the funniest adventures in the book, that in which Maritornes leaves Don Quijote dangling by one arm at the inn, might well have been inspired by a similar episode in Cirongilio de Tracia¹⁹. This romance (as stated above, one which Clemencín was unable to obtain) is no more than named by Gayangos²⁰ and Menéndez Pelayo²¹; Thomas speaks of it only to subject it to his usual ridicule²². Although we need not agree with the author of the book’s colophon, who claimed that the language of the work could be said to surpass Ciceronian Latin, the book is not devoid of merit, and the author makes, at times, a distinct effort to attain a refined style.

The episode in question is the following: in Cirongilio, a certain knight delights in playing tricks on others, and is named the Metabolic Knight, the author (confusing the word with «metamorphic») tells us, because of the disguises he uses in carrying out his tricks (III, 12). Dressed as a girl, he succeeds in robbing the horses of two knights, by means of a series of deceptions (III, 13). They have no choice but to buy their own horses back from him, and outside his castle offer to do so. The Metabolic Knight refuses to open the castle doors, but from a tower lowers a basket on a rope for a squire to be pulled up, along with the money. Having pulled the squire halfway up, he ties the rope securely and goes off and leaves him (III, 14). The squire manages to escape by using the money to bribe one of the castle servants to lower him. The same servant lets the knights into the castle, and they, with considerable amusement, take their revenge on the Metabolic Knight by suspending him with ropes by his wrists²³.

Another discovery has to do with the Cave of Montesinos, a central episode of Part II of the Quijote. It casts additional doubt on Clemencín’s reliability since its source is found in a work which he supposedly studied. Among other examples of caves, Clemencín cites one in the Espejo de príncipes y caballeros (last note to Don Quijote II, 22), but for his main illustration of this adventure he cites an episode in the Sergas de Esplandián (note 41 to Don Quijote II, 23). María Rosa Lida developed this parallel²⁴. But the similarities between the Cave of Montesinos adventure in Don Quijote and the Cave of Artidón adventure in the Espejo de príncipes are so striking that they suggest that the Espejo de príncipes is, if not the only, at least the primary source for this important adventure²⁵.

Whereas in the Sergas de Esplandián XCIX, it is the author Montalvo who, by accident, falls into a nameless well, in both Don Quijote II, 22 and the Espejo de príncipes II, 4 and 5 it is a protagonist who enters a famous cave in search of adventures. Both Rosicler, who carries out this adventure in the Espejo de príncipes, and Don Quijote are concerned about their ladies, which Montalvo is not. Don Quijote «sees» her, a fact of great importance to him; Rosicler learns about her. In both the caves of Artidón and of Montesinos, we meet a dead lover, with his heart in the one case exposed, in the other removed; they both talk when there is need, but sparingly. In both cases the desired lady is enchanted there as well.

Clemencín also did not note in his reading of the Espejo de príncipes y caballeros that Lirgandeo, one of the two «authors» of the work, comments on the story in a manner surprisingly similar to Cide Hamete in his «marginal notes». When the author, Diego Ortúñez de Calahorra, speaks in the first person, as he does on infrequent occasions, his tone is similar to that of Cervantes when we hear him speaking²⁶.

There are also episodes in Don Quijote which stand out as clearly inspired by the romances of chivalry, though not by a particular one. For example, the description in I, 9 of Don Quijote’s battle with thevizcaíno is a delightful parody of the clichés used in describing duels in the romances of chivalry: the fierce appearance, the blow stopped by fortune, the blow which carves off part of the armor. Also, the fact that Don Quijote steals away from home to begin his adventures has no profound psychological significance, such as Madariaga gives it²⁷. It was, in fact, customary for knights-errant to start off on their adventures secretly. Generally, their families and friends were interested in seeing them remain at home, believing them, for one reason or another -often their youth- unready to practice the de manding profession of knight-errantry. Thus, the only way they could begin their adventures was secretly.

In the realm of style, Hatzfeld has seen in Cervantes’ use of contrary-to-fact conditional sentences «la gran idea de la condicionalidad del ideal»²⁸. In fact, this sentence structure is a common feature of the romances of chivalry, which Cervantes has picked up, at least in some places deliberately. Three examples, found without much effort, should demonstrate this:

Frequently what is not discovered in an investigation is just as enlightening as what is discovered in it. A case in point is the role of magic in the romances of chivalry. Although it is usually present, it is more often benign than evil. Virtually every knight had a sabio, among whose abilities was that of working magic, to protect him; rarely do we see evil enchanters at work, and certainly beautiful women are not changed into ugly ones. Thus Don Quijote’s paranoia stands out even more clearly: he is not explaining the world in terms of the romances of chivalry, but in terms of his psychological needs.

Finally, it becomes obvious, if it were not so already, that Don Quijote’s favorite book was far and away the Amadís de Gaula. This romance, decidedly un-Spanish in nature, no matter what its original language, is sentimental to a degree far surpassing the other Spanish romances of chivalry, in which action is, more than love, the main interest. Don Quijote’s devotion to Dulcinea, which is a constant motivating force throughout the book, could have been modeled only on that of Amadís to Oriana.

In conclusion, a thorough study of the chivalric sources of the Quijote, preliminary to one of Cervantes’ humor²⁹, is in order. Enough private collections of the romances of chivalry have passed into public libraries that the works are accessible to all: on microfilm, one can assemble all the works which presumably were in Don Quijote’s library, until now an unrealizable but common dream of Cervantine bibliophiles. Spanish bibliography has progressed to the point that we now know the location of at least one copy of almost every romance of chivalry³⁰. Based on a modern interpretation of all aspects of the Quijote, and without the nineteenth-century bias against the romances of chivalry, such a study could not help but prove immensely rewarding, both in terms of our understanding of the Quijote and of the romances which gave it birth. All we need to do is begin.