When in 1814 John C. Dunlop published his History of Prose Fiction,' he appears to have been unaware of any relation between Heliodorus and Cervantes. The omission is, however, supplied

by Felix Liebrecht, who translated Dunlop's work into German with the addition of numerous valuable notes.2 The former saw fit, nevertheless, to repeat merely the unqualified statement that the Persiles is an imitation of Heliodorus, which he took, perhaps, as much from Ticknor as from Bouterwek. In 1822 the same idea had emanated from the pen of the noted Calderon scholar, Friederich W. V. Schmidt, which is all the more remarkable since he was an enthusiastic admirer of the Persiles, and must have recognized in it something more than a mere imitation of Heliodorus. Whereas we have extravagant praise in his Beiträge referred to above (p. 5, n. 3), we are now told merely that "die berühmteste Nachahmung [des Heliodor] bei den Spaniern ist die nordische Geschichte Persiles und Sigismunda von Cervantes." In 1857 Schmidt's early studies on Calderon's plays were incorporated in his important work on that poet, so we have the same idea unchanged, after a lapse of thirty-five years.*

1 This work, of the utmost importance for a study of the genre to which the Persiles belongs, was entitled: The History of Fiction: Being a Critical Account of the Most Celebrated Prose Works of Fiction from the Earliest Greek Romances to the Novels of the Present Day (Edinburgh, 1814; 3 vols., 8vo; 4th Engl. ed., 2 vols., London, 1888, from which I shall quote from time to time).

2 The title reads: J. Dunlop's Geschichte der Prosadichtungen oder Geschichte der Romane, Novellen, Märchen . . . . aus dem Englischen übertragen. . . . vermehrt. ... mit Anmerkungen versehen (Berlin, 1851; cf. pp. 458 and 511). Liebrecht's notes were incorporated into the fourth English edition. The remark referred to is on p. 404, Vol. II, n. 3, of latter work. Erwin Rohde, in his excellent work, Der griechische Roman und seine Vorläufer (2d ed., Leipzig, 1900), cites Liebrecht's note without comment (p. 472, n. 1). In the English edition of Dunlop's work the Persiles is called by the peculiar title of The Sorrows of Persiles and Sigismunda, and in German Die Leidensgeschichte des Persiles und der Sigismunda, a title which Liebrecht may have taken from Dorothea Tieck's translation called Die Leiden des Persiles und der Sigismunda (cf. p. 5, n. 3). A better rendition of Trabajos would be "Wanderings," since the plural Trabajos is used in this connection to signify the hardships of adventure.

3 Wiener Jahrbücher der Litteratur, Vol. XVIII, 1822. Cf. Anzeige-Blatt für Wissenschaft und Kunst, No. XVIII, p. 8.

Die Schauspiele Calderon's dargestellt und erläutert von Fried. Wilh. Val. Schmidt (Elberfeld, 1857), p. 290. Even Gervinus, in his Geschichte der poetischen National-Litteratur der Deutschen (2d ed., Leipzig, 1840), left the opinion of his predecessors unchallenged. He says (Vol. I, p. 263): "Es ist aber zu vermuthen, dass, wie später Tasso den Heliodor benutzte, wie den italienischen und spanischen Schäferdichtern Longus vorschwebt, wie Cervantes' ernster Roman [i. e., Persiles y Sigismunda] den ganzen Zuschnitt der griechischen Romane trägt, so auch in früherer Zeit vielerlei Griechisches in die neue romanische Poesie Eingang gefunden haben mag." This view was modified in the fifth edition, entitled

I have dwelt thus far only upon the appreciation which the Persiles met in Germany, where scientific research and scholarly criticism in the field of Spanish made practically the only progress achieved during the first half of the nineteenth century.' We come now to the judgment passed upon the Persiles by George Ticknor, which is the most important of all, inasmuch as it has been unhesitatingly accepted and repeated up to the present time. Ticknor's criticism is, as usual, a thoroughly independent one, and will to a large extent at least, where common-sense or what is rational forms the only criterion-remain irrefutable. But while, generally speaking, it is impossible for a historian who covers a nation's whole literature to do justice to every important work, it will also be admitted, in the particular case of Ticknor, that, great as is his history as a whole, he was temperamentally less fitted to judge some works than he was others. Among those which suffered in his clear, unemotional treatment we must place the Persiles; whose importance lies in the fact that it is a characteristic production of its epoch, a creation not only typical of Spanish temperament, but one indispensable in any final word on the genius of Cervantes. This neither Ticknor nor any critic who followed him has duly recognized.

Ticknor begins by saying that the purpose of Cervantes seems to have been to write a serious novel when he undertook the PerGeschichte der deutschen Dichtung, Vol. III (Leipzig, 1872), p. 206: "In Persiles und Sigismunda ging er [Cervantes] bis auf die Quelle der ernsten Ritterdichtungen zurück, auf den alexandrinischen Roman, schildert uns gleichsam zur Erkenntniss den Typus dieser ganzen Litteratur, in dem er uns ein liebendes Paar, das durch ein stetiges Gefühl aneinander geknüpft ist, von dem wunderlichsten Wechsel der Dinge ergriffen und als Spielball einer günstigen Göttin, Fortuna, zeigt." The latter idea is importaut and will be considered in connection with Cervantes' theory of fiction. O. L. B. Wolff, Allgemeine Geschichte des Romans (Jena; 2d ed. 1850, p. 119), adds nothing to our knowledge. J. L. Klein, Geschichte des spanischen Dramas (Vol. IX of Geschichte des Dramas; Leipzig, 1872; p. 274), sees no saving qualities whatsoever in the Persiles.

1 To be convinced of the interest and activity in behalf of Spanish literature in Germany at this time, one need but consult the notes in Ferd. Wolf's work on Spanish and Portuguese literature (1859), or such works as Schack's history of the Spanish drama, or Lemcke's Handbuch der spanischen Litteratur; and as regards the interest taken in Cervantes alone, the long list of translations as well as of editions in the original Spanish printed in Germany (given by Rius, Bibliografía, Vol. I) is an ample testimony.

2 History of Spanish Literature, by George Ticknor (3 vols.; London, 1863), Vol. II, pp. 133 ff. The edition from which I quote differs but little from the German version of Julius, or the Spanish edition by Gayangos. Ticknor himself said, referring to all the scholars who completed his work: "From the results of their labors, carefully prosecuted.... I have taken.... everything that, as it has seemed to me, could add value, interest, or completeness to the present revised edition." (Preface, p. x.)

siles, and then he casts about to see what models Cervantes could have found for serious romantic fiction. All that the latter says, however, is that he hopes to produce an excellent libro de entretenimiento,' and nothing could have been farther from his thoughts than Ticknor's "serious"-that is, "modern"-conception of fiction. What Cervantes meant to produce was simply a tale of adventure extended beyond the ordinary length of the current novela. That this is all he implied can be seen from the common meaning of entretenimiento in his day. Near the beginning of the novela, Las fortunas de Diana, written shortly after the death of Cervantes, Lope de Vega tells of his hesitancy in undertaking this genre in literature, which he had left untried up to that time, and which seemed to him more at home in Italy and France than in Spain. He admits the success of Cervantes in this field, and then adds:

....

Confieso que son libros de grande entretenimiento, y que podrian ser ejemplares, como algunas de las historias de Bandelo. .. Y habiendo hallado tantas invenciones para mil comedias . . . . servirè a vuestra merced con esta.

This, however, was addressed to his mistress, who was probably not expecting any serious psychological treatment in a tale written for her pleasure and entertainment. Moreover, the large majority of the reading public, especially the women, considered a book of fiction as a pleasant means of passing an hour of leisure, and not even a limited circle of the educated classes was trained to look upon a novela or a comedia as an accurate reproduction of society and its environment. All that the public demanded of a libro de entretenimiento is voiced in the desire so often expressed, namely, that the events described therein be verosimiles or credible. Characters and sentiments were not subjected to scrutiny, provided they were pleasing or amusing. Therefore, even such produc

1 Cf. "Dedicatoria al Conde de Lemos," Don Quixote, Part II. "Con esto me despido, ofreciendo a V. Ex. los Trabajos de Persiles y Sigismunda, libro a quien daré fin dentro de quatro meses, Deo volente; el qual ha de ser, o el mas malo, o el mejor que en nuestra lengua se haya compuesto: quiero dezir de los de entretenimiento; y digo que me arrepiento de auer dicho el mas malo, porque segun la opinion de mis amigos, ha de llegar al estremo de bondad possible."

2 Printed in La Filemena, con otras diversas Rimas, Prosas y Versos, de Lope de Vega Carpio (Madrid, 1621); accessible in "Biblioteca de Autores Españoles" (Rivadeneyra), obras no dramaticas de Lope de Vega (Madrid, 1872), p. 1.

tions in Spanish literature as may be said to give a good picture of contemporary life must be carefully examined, if definite results regarding the customs and culture of the times are to be reached. This is especially true in the case of the theater of Cervantes' day. To be sure, the comedia is one of the most important sources that we have for the study of Spanish culture, but its value is frequently vitiated by the playwright's failure to differentiate sufficiently the spirit of fiction in comedy from that of the novela. In the latter, absence of psychological truthfulness and an excess of romantic or imaginative elements are pardonable and even logical; but the farther a comedia gets from that which is simply natural and actually representative, the less it can be used as a reliable document on contemporary life. The power of appreciating the distinctions between fact and fiction, however, is a matter of training, and playwrights were indifferent to them even when they were ostensibly walking upon the solid ground of history. Not infrequently do we find the claim of a historia verdadera' made for a comedia which, though drawn from a germ of truth lodged in some chronicle or popular ballad, is in its ultimate form, for the most part, an imaginary creation. Such being the spirit of every kind of fiction, a novelist would not feel tempted to look for "serious" models for his work; he would be guided by the spirit and practice of contemporary writers. It is therefore plain that Cervantes was merely in need of some framework which would enable him to draw out indefinitely the manner of the novela, and thereby create a book for general entertainment,' longer than the ordinary tale. That was all he could have intended to do. But Ticknor is troubled to find a guide for the Persiles, and all that he can hit upon is "the imaginary travels of Lucian, three or four Greek romances, and the romances of chivalry." I have been

1 For a full discussion of the term historia verdadera in connection with the comedia cf. Max Krenkel, Klassische Bühnendichtungen der Spanier, Vol. III (Leipzig, 1887), pp. 21 ff.

2 The term libro de entretenimiento or libros entretenidos (cf. p. 3, n. 1) had come to include all prose creations of fiction, just as the term comedia included both tragedy and comedy. It was applied to trifles like patrañas, and diálogos (cf. those de apacible entretenimiento, by Gaspar Lucas Hidalgo), as well as to a long history like that of Persiles (the aprobacion of the Spanish version of Tatius [cf. p. 14, n. 1] says it was worthy of being printed "para apacible entretenimiento y exemplo de artificiosas y utiles ficciones"). Or we find it replaced by pasatiempo and recreo (cf. El Patrañuelo, by Timoneda, epistola al amantisimo lector), or by apacible recreacion, as in Valdiviesso's aprobacion, cited above.

able to discover no evidence from the Persiles itself that Cervantes ever saw Lucian's True History. Moreover, it would be a difficult task to prove either from his life or his writings that he could read Greek or had the time to do it. I hope to show in what follows later that the knowledge which he had of Latin authors could have been obtained through the medium of translations; and I see no reason to believe that he could read French. On the other hand, both his long sojourn in Italy as well as the testimony derived from his works justify the conclusion that he was thoroughly acquainted with Italian.' I have been unable to find any mention of a complete Spanish translation of Lucian' printed within the lifetime of Cervantes, but at least seven editions in Italian appeared in the first half of the sixteenth century. One of the latter he could therefore have seen during his sojourn in Italy. But the idea of Ticknor is at bottom somewhat illogical. The True History of Lucian is a wild extravaganza,' a satire on previous books of travel; and, notwithstanding this fact, Cervantes, who had planned a "serious romance," according to Ticknor, is supposed to have had it among the few books which served as a guide for the Persiles. Lucian may therefore be dismissed without

further thought.

The influence "of three or four Greek romances," as Ticknor rather vaguely puts it, is, on the other hand, worthy of the most careful consideration. In the absence of any specific names, we

1 It is possible that Cervantes knew the works of Teofilo Folengo (1491–1544), which may have suggested to him the origin of Don Quixote's madness. The first impulse to write his great work would thus have come from Italy. Cf. B. Zumbini, Studi di Letteratura Italiana (Firenze, 1894), p. 165.

2 Salva's catalogue No. 1879 mentions a Historia verdadera de Luziano traduzida de Griego en lengua Castellana (Argentina, 1551); but this contains only Book I. Lucian's Dialogues, however, appeared in Spanish in 1550 (anonymously), and again in 1621, translated by Franc. de Herrera Maldonado. Both are mentioned by Salvá (Nos. 3934, 3935 of his catalogue), and by Graesse, Trésor de livres rares et précieux (Dresden, 1863; under Lucian, Vol. IV, p. 277). Lucian's works were first translated into French in 1583 (Paris); cf. Graesse; another edition, 1634 (Paris), is mentioned in Fabricius, Bibliotheca Graeca, Vol. III, p. 507 (Hamburg, 1726).

3 Cf. Graesse, Trésor de livres rares et précieux.

4 It will be remembered that among the various experiences through which Lucian and his companions go in their travels, are shipwrecks upon islands where the rivers are of wine and the trees women from the waist upward; a trip to the moon, where they meet men carried by great vultures; a battle between the hosts of the Sun and the Moon, in which the soldiers from the Great Bear are mounted on fleas as large as elephants; a sojourn in the belly of a whale large enough to hold forests and great cities, etc. Cf. Rohde, Der griechische Roman, op. cit., pp. 204 ff.