1. On Montaigne's reading practice and his method of quotation see A. Compagnon, La seconde main ou le travail de la citation (Paris, 1979), esp. 279-312; Mary B. McKinley, Words in a Corner: Studies in Montaigne's Latin Quotations (Lexington, Ky., 1981); Terence Cave, "Problems of Reading in the Essais ," in Montaigne: Essays in Memory of Richard Sayce , eds. I. D. McFarlane and Ian Maclean (Oxford, 1982), 133-66.

2. Terence Cave, "Problems of Reading in the Essais ," 153. For readings of the "Au lecteur" see also Starobinski, Montaigne en mouvement , chap. 1, sec. 7; Fausta Garavini, Monstres et chimères Montaigne, le texte, et le fantasme , trans. Isabel Picon (Paris, 1993), chap. 3; André Tournon, Montaigne: La glose et l'essai (Lyon, 1983); Michel Simonin, "Rhetorica ad lectorem: Lecture de l'avertissement des Essais," Montaigne Studies 1 (1989): 61-72.

3. Several critics have recognized the competing demands of both author (or text) and reader and have argued that they both must be respected. In On Deconstruction (Ithaca, 1982), 72-73, J. Culler speaks of the "divided quality of reading" that acknowledges textual completeness at the same time as it treats the text as something to be created in the process of reading. In his concept of the open text, which he elaborates in The Role of the Reader (Bloomington, 1984), U. Eco opens writing to numerous personal interventions on the part of the reader, but he maintains that in spite of reading the text remains the world intended by the author (p. 62). I place more stress on the tension between authorial intention and reading and argue that reading may (must?) in fact betray or misrepresent the world intended by the author. See also M. Charles, Rhétorique de la lecture (Paris, 1977), who posits the dynamic relationship between a text that both "permits" the reader's entry and constrains reading at the same time, between a reader who acts upon the text and who is acted on by it.

4. M. Foucault, Les mots et les choses (Paris, 1966), 50-59; Regosin, The Matter of My Book , 68-72; H. Friedrich, Montaigne , trans. Robert Rovini (Paris, 1968), 108-11.

5. For Ambrose's silent reading, see The Confessions of St. Augustine , book VI, 3.

6. R. Sebond, La théologie naturelle , trans. M. de Montaigne, in the Oeuvres complètes de Michel de Montaigne , ed. A. Armaingaud (Paris, 1932), 9:xi.

7. François Rabelais, Oeuvres complètes , ed. Pierre Jourda (Paris, 1962), vol. 1, p. 8. Translation is from The Complete Works of François Rabelais , trans. Donald M. Frame (Berkeley, 1991), 4.

8. See François Rigolot's chaps. 3 and 4 in Les métamorphoses de Montaigne on the thematic, structural, and stylistic significance of the letter in the Essais . Rigolot analyzes the letter as the expression of ideal friendship and contrasts it with the degraded essay written after the loss of the friend. My own observations on this subject reveal their debt to him.

9. On the effects of printing on the form and content of the Essais see Barry Lydgate, "Mortgaging One's Work to the World: Publication and the Structure of Montaigne's Essais," PMLA 96 (1981): 210-223.

10. In his analysis of this passage and its context in the essay in Rhétorique de la lecture (289-98), Michel Charles privileges the painter as the only artist whose work exceeds his intention and concludes that Montaigne's writing represents a pictorial rather than an exegetical model, one characterized by anamorphic presentation. He argues that the essayist's reluctance to affirm his intention cunningly allows him to recuperate all readings as "intended" and to profit from the role that fortune might have played in the production of the text. While the pictorial metaphor based on the play of different configurations on the same surface or plane provides a richly suggestive perspective on the Essais , I would argue that elements of surface and depth, outside and inside (elements of what Charles calls the exegetical model), also function in the text to complicate and undermine Montaigne's efforts to write (be) an unequivocal, transparent surface, however diversely configured. Montaigne is both painter and poet, as Charles's own quotations from the Essais reveal. I also think that the essayist's ability to recuperate readings depends in large part on the fact that he has already inscribed "all" readings and readers in his text. Cf. my "Conceptions of the Text and the Generation(s) of Meaning: Montaigne's Essais and the Place(s) of the Reader," Journal of Medieval and Renaissance Studies 15, no. 1 (Spring 1985): 101-14.

11. Cave, "Problems of Reading in the Essais ," 143. I am indebted to Cave's seminal essay, which touches on so many complex aspects of reading in the Essais both in terms of what he calls the moral and personal problems of reading (the text's status as a self-portrait) and intellectual and aesthetic problems concerning meaning and form in the work.

12. Montaigne's concept of man appears to include the "foreign" as an element of man's nature itself, that is, as something that he cannot escape without escaping from himself, no matter how close he comes to the ideal of knowing and being himself. To be human would thus not be to realize the dictum of the Delphic oracle but to strive toward it. Man is like his actions, which Montaigne characterizes in "De l'experience" (III, 13, 1,077) as "doubles et bigarrées à divers lustres." This is the doubleness that I have characterized as the monstrous. On the foreign as a psychological element in Montaigne, what Garavini calls "des monstres de l'inconscient," and on the essayist's efforts to neutralize it, see her Monstres et chimères .

13. I am indebted in this section to Cave's "Problems of Reading" and to his analysis of these same quotations from the Essais , which lead him to suggest that "Perhaps the first sense was already a particular reading, to which the author brought his own momentary preoccupations, and which was dissipated as the author-reader shifted his perspective. There would thus be no moment of perfect unity of text and meaning, presided over by a godlike author, only a set of words that can be read in various different ways by different readers" (156-57). Where Cave qualifies his observation by adding, "No doubt Montaigne does not go quite as far as this," I am inclined to say that Montaigne's text does indeed allow us as readers to go as far as this, for "this" is precisely the nature of the obtrusive reader's text, which the Essais do go far enough to include.

14. Reacting to the intentional fallacy decried by American New Critics, Paul de Man seeks to restore the concept of structural intentionality to literary criticism. In terms of Heidegger's theory of hermeneutic circularity he explains literary interpretation as an effort at understanding that addresses the work of art as an intentional object rather than a natural or organic one and that produces a sense of literary "form" that is "the result of the dialectic interplay between the prefigurative structure of the foreknowledge [this foreknowledge is the text itself] and the intent at totality of the interpretative process" ("Form and Intent in the American New Criticism," in Blindness and Insight: Essays in the Rhetoric of Contemporary Criticism , rev. ed [Minneapolis, 1983], 31). De Man makes it clear that "form" is never absolute or absolutely or concretely in the work but is constituted in the mind of the interpreter as the work discloses itself to his endless questioning.

15. Charles recognizes this strategy in his epilogue on Montaigne in Rhétorique de la lecture , 297-98.

16. Georges Poulet, "Criticism and the Experience of Interiority," in The Structuralist Controversy: The Languages of Criticism and the Sciences of Man , eds. R. Macksey and E. Donato (Baltimore, 1970), 67. In his discussion of contemporary critics whose readings might be seen to realize the mysterious interrelationship among author, work, and reader that he has been describing, Poulet compares Jean Starobinski and his "optimism" to the Rousseau whom he characterizes in the terms I have quoted.

17. Ibid., 56-88. The quotation is from p. 63.

18. Georges Poulet, Studies in Human Time , trans. E. Coleman (Baltimore, 1956), 13. Poulet's image of Montaigne as detached yet still engaged with things strikes me as accurate, although I take issue with the way Poulet ultimately understates the resistance of things to the essayist's effort to reflect upon them, to integrate them within the mind, and, finally, to master them. Poulet recognizes the difficulty of Montaigne's task but insists on the ultimate triumph of the mind, a mind "forever free and forever faithful," and "faithful and identical to itself when it preserves and augments its power of apprehension" (48). My emphasis falls rather on the mind's occasional triumphs, on its equally frequent failures and lapses, and on its inability to succeed, to establish being and plenitude once and for all.

19. Thibaudet, Montaigne , 522.

20. W. Iser, The Implied Reader: Patterns of Communication in Prose Fiction from Bunyan to Beckett (Baltimore, 1974), 280 and passim; The Act of Reading: A Theory of Aesthetic Response (Baltimore, 1978), passim .

21. Iser, The Act of Reading , 34.

22. Iser, The Implied Reader , 279.

23. Ibid., 288.

24. Iser, The Act of Reading , 37.

25. Robert Holub sees this tension as reflecting inconsistencies in Iser's theory, his inability to decide how to weight the relationship between the freedom of the reader to produce meaning and the imposing determinacy of the text. See his Reception Theory: A Critical Introduction (London, 1984), 101-6. See also the exchange between Stanley Fish and Iser on the determinacy / indeterminacy question in Diacritics 11, nos. 1 and 3 (1981).

26. Iser, The Act of Reading , 167.

27. Cave, "Problems of Reading," 161.

28. See my "Conceptions of the Text and the Generation(s) of Meaning: Montaigne's Essais and the Place(s) of the Reader," where I explore the ways in which Montaigne's writing anticipates and authorizes its own readings.

29. See my "The Boundaries of Interpretation: Self, Text, Contexts in Montaigne's Essais ," in Renaissance Rereadings: Intertext and Context , eds. Maryanne Horowitz, Anne Cruz, Wendy Furman (Urbana, 1988), 18-32.

30. Michel Charles insists upon the double activity of reading, the text acting upon the reader and the reader upon the text. See Rhetorique de la lecture , 63. He also draws our attention to the pertinence of Plato's Ion to the discussion of the complex relationship between the demands of the text to be heard and the need for the reader to have his say. The Ion situates in the singularity of the rhapsode as the hermeneut, the interpreter, the dual task of both reciting the poem of Homer and commenting upon it, as if both the poet's meaning and the listener / reader's understanding had to be given their due. In a note (p. 64) citing the article of Jean Pépin, "L'herméneutique ancienne," Poétique 23 (1975): 291-300, Charles attributes to Pépin the conclusion that "l'herméneute peut être le porte-parole ou l'exégète, avec un jeu possible sur les deux sens comme dans l' Ion , précisément, qui est cité [by Pépin] comme exemple: 'les deux sens principaux du verbe hermeneuein ...s'entrelacent non sans subtilité avec les deux fonctions du rhapsode, sans que l'on sache toujours bien dans quel registre on se trouve' (p. 296)." In this figure of the rhapsode I find the simultaneous and competing demands of text and reader and the expression of their problematic relationship. Our discussion ultimately affirms Pépin's characterization of the rhapsode by suggesting how difficult it is to determine in any absolute way what finally belongs to the text and what to the reader.