As the overlap of Danae and Andromeda in the Perseus story shows, in myth the hero's relationship to a consort is often not clearly distinguished from that which he has with his mother. «The Christian theological position that identifies Jesus with God makes Mary's relationship to the two not unlike that of Isis, Horus and Osiris. Joseph Campbell discusses a number of deities who are at once the consorts and sons of the Great Goddess of the Universe...». (El Saffar 189, n. 35). (N. from the A.)

On the avoidance of triangles by Periandro in the Girardian sense see El Saffar, p. 157. (N. from the A.)

Although not in relation to the myth of Perseus, El Saffar has stressed that throughout P and S one can observe examples of «the literally transforming effects of compassion and forgiveness» (155). (N. from the A.)

Forcione traces the multiple appearances of this symbol in P and S (87). (N. from the A.)

I should point out that in recent studies of the Goddess controversy has arisen regarding Athena's personification of the feminine, some feminists feeling she was co-opted by the structures of patriarchy. For a sympathetic treatment of her as a personification of the feminine see Karóly Kerényi's Athene (Zurich, 1952), translated by Murray Stein as Athene: Virgin and Mother (Irving, Texas, 1978). (N. from the A.)

I am indebted to the comments and suggestions of my friends and colleagues Frederick de Armas, Helena Percas de Ponseti and Thomas A. O'Connor, who had occasion to read this article at various stages of its development. Remaining errors and weaknesses are my own. (N. from the A.)

On closure (as ending) in the novel see Frank Kermode, Maria Torgovnick, D.A. Miller; Nineteenth Century Fiction (33); Yale French Studies, No. 67. Salvador J. Fajardo deals with aspects of the end of D.Q.I., but without touching on the formation of the canon as a reader or with other readerly issues in the book's final chapters. Julio Rodríguez-Luis also addresses questions of closure in Part I but not those examined here. Armine Kotin Mortimer has a nicely compact description of narrative closure: «La conception de la clôture narrative dépend souvent d'un sentiment satisfaisant que toutes les données du récit ont abouti à leur fin plus ou moins nécessaire, que les problèmes posés par la narration sont résolus, qu'aucun bout du fil narratif ne reste flottant, que les signes composant l'univers narratif sont épuisés, en somme, que ce qui à été ouvert est clos» (15). This description applies principally to the classical realistic novel, but it sets a possible norm from which we can gauge the deviations notable in Don Quixote. (N. from the A.)

The open form was one important characteristic of the Baroque in general, and Cervantes showed a special fondness for it. Giancarlo Maiorino comments: «At its paradoxical peak, boundary art marked yet another threshold, for the inherent dynamics of growth had to sacrifice product to process. Michelangelo had taught (Florence and Rondanini Pietás...) that creation could come out of incompleteness, and baroque artists steered this lesson toward parody and denial» (131). Besides Don Quixote I, El coloquio de los perros and Rinconete y Cortadillo, for instance, are open in form. In Don Quixote I this openness receives special emphasis in line with the text's own ludic/parodic character. (N. from the A.)

On this topic see Daniel Eisenberg and Edwin Williamson. (N. from the A.)

Cf. Parr, Ch. 2. (N. from the A.)