Translation of Erich Auerbach's "Epilegomena to Mimesis"

MIMESIS THE REPRESENTATION OF REALITY IN WESTERN LITERATURE Fiftieth-Anniversary Edition BY ERICH AUERBACH TRANSLATED FROM THE GERMAN BY WILLARD R. TRASK With a new introduction by Edward W. Said PRINCETON UNIVERSITY PRESS PRINCETON AND OXFORD CONTENTS Introduction to the Fiftieth-Anniversary Edition IX Odysseus' Scar 2. Fortuna ta 3 24 3· The Arrest of Peter Valvomeres 50 1. 4· Sicharius and Chramnesindus 5· Roland Against Ganelon 6. The Knight Sets Forth 7· Adam and Eve 8. Farinata and Cavalcante 9· Frate Alberto 10. Madame Du Chaste! 11. The World in Pantagruel's Mouth 77 96 123 143 74 203 1 232 262 12. L'Humaine Condition 285 13· The Weary Prince 312 14. The Enchanted Dulcinea 334 15. The Faux Devot i6. The Interrupted Supper 359 17. Miller the Musician i8. In the Hotel de la Mole 395 434 454 i9. Germinie Lacerteux 20. The Brown Stocking 493 525 Epilogue 554 Appendix 559 575 Index APPENDIX: "EPILEGOMENA TO MIMESIS" * by Erich Auerbach Translated by Jan M. Ziolkowski SIN c E the publication of the book more than six years have passed; numerous reviews have appeared, among which many have been very extensive ..lt can be assumed that most of the critical ideas that Mimesis can prompt have been expressed in them. For that reason I would like now to say something in reply to some of these ideas. Only to some: it would be wholly impossible to give due recognition to everything instructive and interesting that has been presented by so many commentators, mostly well informed and understanding. I have selected topics that are especially close to my heart- either because I have something to concede or because I believe that I am obliged to defend my views against misunderstandings. I expected that the most serious objections against the train of thought in the book would come from the direction of classical philology, for ancient literature is treated in my book above all as a counterexample. According to the statement of the basic theme, I had to show what ancient literature does not possess. The one-sidedness that results from this can disturb and perhaps even offend a reader who is a philologist of ancient literature, and it could be moderated, but not altogether avoided. I was gratified that both of the reviews by classical philologists - Otto Regenbogen [i891-1966] and Ludwig Edelstein [i902-1965] formulated their objections with much understanding and consideration for the overall intent of the book. The two reviews have much in common: both try to contest or to weaken my view on the limits of ancient realism, both offer examples in opposition, and both engage in polemic against one passage (Mimesis, pp. 38-39) where I turn to speaking of the limits of ancient historiography. The much more extensive review by Regenbogen (which is most especially interesting for me) introduces, in addition, a criticism of my handling of Homer and Augustine. 3 2 1 - "' First published in Romanische Forschungen 65 (1953) 1-18. ' Mimesis. Eine Rezension. Den Mitgliedern von Svenska Klassikerforbundet .. . i.iberreicht. Uppsala: Privately Printed, 1949, 23 pp. Rept. in Otto Regenbogen , Kleine Schriften, ed. Franz Dirlmeier (Munich: C . H . Beck, 196i), pp. 600-617. Modem Language Notes 55 (1950) 426-43i. 1 1 I will not here delve into the questions relating to Augustine; they have been dealt with in the meantime in this journal [Romanische Forschungen ], 64 (1952) 309ff. 559 APPE DIX Here I have to admit, first of all, that the themes of Homer's lack of tension and "foregrounding" were emphasized all too strongly in the initial chapter, and that I am on the whole not entirely satisfied with my first chapter. On many points I am in agreement with what Regenbogen says, especially on pages 12 and 13 of his review; this first chapter might have dealt with other literary documents of archaic Greece (to bring into consideration ones from Homer himself, from Aeschylus, perhaps from Hesiod, and even perhaps the art of the sixth century) in which something very closely related to existential realism is sometimes expressed. However, that would have given the book an entirely new dimension, so to speak, into early antiquity, and I recoiled from that. I considered for a moment letting the Homer chapter fall entirely by the wayside. For my purposes it would have sufficed to begin with the time around the birth of Christ. But it proved not to be feasible to find an introduction that would have been able to measure up to the Homer chapter in clarity and effectiveness for presenting the problem, and so I let it stand, somewhat toned down in regard to the first draft. It seemed to me justified to stress themes that matter in the context of the book and that have been properly viewed, even if developed in a one-sided way. But I emphasize the one-sidedness of the presentation here expressly because, time and again, there are readers who especially praise the first chapter in particular. Perhaps I should also have refrained, in the second chapter, from the remarks about ancient historiography; perhaps then it would have been possible to avoid this complex of problems. For an undertaking like my book, it is advisable to confine oneself to what is absolutely necessary for the development of one's thoughts; and it is always difficult to choose one's expressions, in the treatment of a problem as broadly layered as is the problem of ancient historical writing, so that they characterize adequately the totality of the phenomena - as in this case, for example, Herodotus, Thucydides, Polybius, and the later historians. For instance, Herodotus is not comprehended in my terminology (rhetorically and moralistically), even if one understands the expression "moralistically" as it is meant in the context of my thought (in opposition to "historically"). But I can make no further concession. In the decade that has passed since the editing of the Petronius chapter, my views on the difference between ancient and modern historiography have become stronger in relation to the presentation of the problem and the shaping of the idea. To explain them again here would go beyond the framework of an essay; vita comite ("if I live so long"), as a Carolingian author was accustomed to say, I will return once 560 APPENDIX again to the topic. 4 Yet I wish to remind readers that I did not compare Thucydides or Tacitus with figures like them in modern times (it would be difficult even to find figures like them there!) but rather with a modern, albeit prominent, professor of history- just to show how the bases for the presentation of the problem and the shaping of the idea have changed. 5 I wish also to remind readers that the modern perspectival and "historicist" examination of history has been fully developed for just barely a century and a half. And finally I wish to state that the word "limits" of ancient historical writing contains no negative value judgment at all. Quite the contrary. The unity, drama, plasticity, and humanity, which come into being through the restriction to a narrow circle of people acting with and against one another, have become unattainable. In regard to the "stylistic differentiation" in later (thus, for example, "post-Socratic") ancient literature, I was sure of my facts from the beginning; but I still read with satisfaction which counterexamples were adduced by my reviewers who were classical philologists. Edelstein cites Aristophanes' assertion that he pursues serious intentions, as well as Plato's, Cratinus', and Cicero's remarks that tend in the same direction. He brings up Middle Comedy and Menander (is it justified to exclude the latter from an analysis of realism?); he speaks of Milesian fable, of Theocritus and Herondas (I never claimed that these were to be excluded from the category considered by me because they wrote in verse!), of mime and of epigrams; and he mentions later also Xenophon's Oeconomicus. People can adduce such counterexamples only if they have lost sight of the concept of realism as I meant it, and consequently assume that I intended to characterize the whole of ancient realism as "a vaudeville show" or as "poking fun." That, however, I neither intended nor did. I call the realism that is alien to antiquity serious, problematic, or tragic; I set it in express opposition to the "moralistic." Perhaps I would have done better to call it "existential realism," but I hesitated to use this all too contemporary term for phenomena of the distant past. And what I meant, it seemed to me, was to be inferred with unmistakable, even overpowering clarity from the passage about Peter and my analysis of it [Mimesis, pp. 40-49]. But Regenbogen, too, mentions Xenophon (Oeconomicus and Socratic Memoirs), where "the 4 [The phrases vita comes and vita comitante are frequent in Carolingian poetry: see Otto Schumann, Lateinisches Hexameter-Lexikon. Dichterisches Formelgut von Ennius bis zum Archipoeta, vol. 5 (Munich: Monumenta Germaniae Historica, i982), p. 680.] s [Auerbach compared the self-expression and thinking of the ancient authors with that of [Michael lvanovitch] Rostovtzeff [i870-1952] (Mimesis, p. 39).] 561 APPE DIX description of everyday life turns up in a sense not at all comic or idyllic." Does Regenbogen believe that these examples of "serious" realism have something to do with what the temptation of Peter contains, that they anticipate or even just announce the world-historical change in stylistic feeling that is proclaimed there? Edelstein writes toward the end of his review: ''Yet, in my opinion, it is not only the contrast, it is also the similarities (between ancient and modern concepts) that need to be emphasized." Of course. It is quite clear to me with what great justification, for example, early Christianity can be regarded as the product of late antiquity. I have read many significant investigations that have been written from this point of view, and I have learned from them. In Mimesis I also took account, partly expressly, partly implicitly, of tl1is approach. But the task that my theme imposed on me was a different one: I had to show not the transition but rather the complete change. Only much later, six years after the book had appeared, did E[rnst] R[obert] Curtius [1886-1956] publish his objections to it. He sees in the book a theoretical construct, from which he seeks to extract tl1eses in order to refute them. But the book is no theoretical construct; it aims to offer a view, and the very elastic thoughts or ideas that hold it together cannot be grasped and proven wrong in single, isolated phrases. I will return to this later. First it is necessary here to enter into the details of Curtius' refutations. He considers as the theses of the book the doctrine of stylistic differentiation and mingling (which for its part rests on the concept of the three ancient types of style) and the doctrine of the figural view of reality of Christian late antiquity and the Middle Ages. [Curtius] deals with the doctrine of the three styles in this periodical (Romanische Forschungen 64 [1952] 57ff.). 6 He begins with an enumeration of the expert opinions on the types of style that have been preserved to our day, from the Rhetoric to Herennius down to Meinhard of Bamberg (eleventh century), so as to reach the conclusion "that the ancient rule of stylistic differentiation is neither so unified nor so absolute as it might seem according to Auerbach." The compilation of expert opinions is useful,7 but it contributes nothing to the criticism of Mimesis. 6 What is said there occasionally about the questions, which were also dealt with by Edelstein and Regenbogen, will not be discussed again here. That the "rustic, serious, and sober" Hesiod has nothing to do with the realism of the Gospels is perfectly clear. And that alone matters, not words that have been taken out of context and that can be interpreted variously. 7 Some observations about this: I do not find a mention of tl1e especially poetic theory of three styles of Heracleides of Pontos ([as transmitted by] Philodemos), which is to be APPE DIX Mimesis is an attempt at the history of the matter itself, not of the expert opinions on it; to write the latter with the resources that were at my disposal in Istanbul would have been altogether impossible. The conceptual pair "stylistic differentiation/stylistic mixing" is one of the themes of my book and always has the same significance throughout the twenty chapters, from Genesis all the way to Virginia Woolf. Thus it does not conform to changes in expert opinions. It has to do with a version of the thought, which was formed by me around i940. In particular, the idea of realism, which is present in Mimesis, was dealt with previously only rarely- and even then in another context. It has nothing to do with "Jest and Earnest in the Middle Ages" or "Kitchen Humor." 8 Incidentally, in the extant ancient opinions on the three styles (most of which relate to oratory) very little is said about realism. It is an entirely different question, whether my pair of concepts covers the tradition adequately, whether it is applicable, whether, therefore, stylistic differentiation was really a characteristic element of ancient stylistic feeling. On the last two pages of his essay (and already in note 3 on page 60) Curtius tries to contest this also, and specifically he thinks: L I advocated the thesis that ancient comedy had been classed in the humble style. 9 Presumably I let myself be misled to this end by Dante's regarded as a source of Horace, as presented by Christian Jensen ("Herakleides vom Pontos bei Philodem und Horaz," Sitzungsberichte der Preussischen Akademie der Wissenschaft.en [Berlin: Akademie der Wissenschaften , i936], Phil.-Hist. Klasse 23, pp. 292ff.; pp. 304ff. on the three styles)-a significant work, of which I was made aware by Curtius himself (European Literature and the Latin Middle Ages, p. 439, note i4). The concept of the prepon appears for the first time not in Theophrastus but instead already in Aristotle (Rhetoric p.1404 b). "Good taste" and "neatness" are however probably too general translations of ele[!,antia and munditia in Cicero, Orator 2'}.-79; it has to do with linguistic purity in a punstic sense, as emerges from what follows: 'sermo purus erit et latinus" ("the language will be pure Latin"). Compare Quintilian 8.3.87 and also many passages in Cicero himself, for example the one cited by Curtius later, De optima genere oratorum 4. On the significance of elegantia, see George Lincoln Hendrickson, "The Origin and Meaning of the Ancient Characters of Style," American Journal of Philology 26 (1qo5) 249-290; on elegantia, pp. 263-264. I hope Tater to return to some medieval material. It might also be noted as far as method is concerned that an assembly of rhetorical exrert opinions is a somewhat single-track way to clarify the situation. A picture of the truly living ideas can be gained only through the study of terms, as for example altus, gravis, sublimis, suavis, dulcis, subtilis, planus, tenuis, privatus, humilis, remissus, pedester, comicus, and so forth; this can now be done with the help of the Thesaurus Linguae Latinae [Leipzig: B. G. Teubner, i900- ). 8 [These two phrases refer pointedly to the title and subtitle of an excursus in European Literature and the Latin Middle Ages. In the handling of tl1ese two phrases and in all page references, I follow the English translation as found in European Literature and the Latin Middle Ages, trans. Wilfard R. Trask, Bollingen Series 36 (Princeton, ew Jersey: Princeton University Press, i953; rept. i990).) 9 As witness (so says Curtius ironically, but inaccurately) I cited Montaigne. It would APPE DIX statements in De vulgari eloquentia 2-4 and Epistola 10.10. But [according to Curtius] this theory surfaces for the first time, as Paget Toynbee [1855-1932] has demonstrated (Dante Studies and Researches [London: Methuen and Co. , 1902] p. 103), in Uguccione of Pisa [often known by his Latin name, Hugutio] (ca. 1200). 2. I seemed generally to accept that for ancient theory a conscious correspondence existed between the types of style and the genres of poetry. This is [according to Curtius] false. At the beginning of De optima genere oratorum Cicero denies the equation of the types of style with the genres of poetry. While there should be no transition between the genres of poetry, there must be between the types of style. Curtius cites verbatim: "oratorem genere non divido, optimum enim quaero" ("I do not divide up the orator by class, for I seek the best"). This would be an express denial of stylistic differentiation . In Institutio oratoria 10.2.22 Quintilian reproduces Cicero's train of thought in a way true to its meanmg. I advocated absolutely no "thesis" on the ascription of genres of poetry to set levels of stylistic elevation. But to be sure, tragedy is always assigned to the high style,1° comedy- ever in accord with its character-to the middle or humbler style," as Boileau still does (and, by the way, Dante, too, in De vulgari eloquentia , loc. cit.). Paget Toynbee takes care not to claim that Dante's characterization of comedy surfaces first in Uguccione. He contents himself with the reference to Uguccione as Dante's direct source. A. Philip McMahon, whom Curtius, for reasons beyond my comprehension, likewise cites, references even older sources of Uguccione: Papias and Isidore ("Seven Questions on Aristotelian Definitions of Tragedy and Comedy," Harvard Studies in Classical Philology 40 [1929] 97-198, here: 140). It is hard to understand how the author of the book European Literature and the Latin Middle Ages can believe that Uguccione, or another medieval author whom Uguccione could have used, created a new definition of comedy. Uguccione's and Dante's definition derives, in the final analysis, from one of the oldest not be so wrong, if I had done it. Montaigne, after all , was traveling on the road from Rome . '° Curtius maybe still remembers the passage cited by him (European Literature and the Latin Middle Ag_es, p. 417) from Wilnelm Schmjd, Geschichte der griechischen Literature, vol. ih (Muruch: C. H . Beck, i934), p. 85. " The special situation of Old Comedy (Aristophanic comedy) in ancient theory, for which Curtius cites Quintilian io.i.65, is a subject that has concerned me for a long time because it plays a role in Dante criticism from the sixteenth up to the eighteentn century, in Vico too . But there was no room for it in Mimesis. APPENDIX definitions we know, and the one that became far and away the most influential: that of Theophrastus. It developed as follows in the glosses of Placidus (fifth-sixth century: Placidus fiber glossarum. Glossaria reliqua, ed. Georg Goetz, Corpus glossariorum Latinorum 5 [Leipzig: B. G. Teubner, 1894], p. 56): "Comoedia est quae res privatarum et humilium personarum comprehendit, non tam alto ut tragoedia stilo, sed mediocri et dulci" ("Comedy is that which comprehends the affairs of ordinary and humble individuals, not in so high a style as tragedy but in a humble and engaging one"). Although the words privatarum et humilium point rather to the lower style, this gloss rates comedy as belonging to the middle style, to which Menander and Terence particularly gave impetus. The contrast to the level of tragedy is essential and permanent. A scholium on Terence (Scholia Terentiana, ed. Friedrich H. Schlee [Leipzig: B. G. Teubner, 1893], p. 163, I. 12) counts comedy as belonging to the lower style: "comoedia villanus cantus, ut qui sit affinis cotidianae locutioni" ("Comedy is lower-class song, the kind related to everyday speech"). Note the agreement with Uguccione and Dante: it is a late-antique topos!' And why does Curtius reject the Horace passage, Ars poetica 93-98, Dante's true source - and one cited by him? Because it has to do with the connection of Lexis ["language"] to the prepon ["the apt (the virtue of parts that fit harmoniously into a whole: decorum)"]. But the doctrine of the types of style is nothing other than the expression of that will to style that connects Lexis to the prepon. From its earliest beginnings, ever since Aristotle, the prepon has been the basis for the doctrine of the types of style. That is the heart of the argument. I never claimed a precise connection of the genres of poetry to the types of style; except for tragedy, the epic in Virgil's or Lucan's style, and on the other hand for the various forms of humbler realism, the classification is uncertain. But I claim the differentiation of style, which is based on the prepon; a hierarchy of forms of expression corresponds to a hierarchy of topics. Every offense against it is cacozelia ["affectation of style"] ("aut magnarum rerum humilis dictio aut minimarum oratio tumens" ["either humble diction for great topics or bombastic speech for the least important"]: Marius Plotius Sacerdos, Artes grammaticae, in Heinrich Keil, ed. Grammatici 2 " A selection of additional, infrequently cited testimonies: Seneca, Epist. 100.10; Donatus, Commentum Terenti, ed. Paul Wessner (Leipzig: B. G. Teubner, 1902-1908), passim (e.g., Adelphoe 638, Hecyra 611); Anthologia Latina, ed. Franz Buecheler and Alexander Riese, vol. iii (Leipzig: B. G. Teubner, 1894-1926), nos. 664 and 661a; quite similarly Ausonius, Opuscula, ed. Rudolf Peiper (Leipzig: B. G. Teubner, 1886), p. 412, no. 367, II. 2-3. APPE DIX latini, vol. 6 [Leipzig: B. G. Teubner, 1874], p. 455, ll. 12-13). Curtius' polemic against this truism of classical philology rests upon a misunderstanding of the texts. He mistakes the mingling of the types of styles or the levels of stylistic elevation with Cicero's challenge that the ideal orator must command them all. The latter appears in the passages of Cicero and Quintilian that he adduces, but there is nothing about a rejection of stylistic differentiation. That Cicero demands the command of all levels of stylistic elevation only from the orator and not from the poet has only a very remote connection with the subject treated in Mimesis (of the differentiation between the high style and everyday realism), but I would like nonetheless to present here briefly the thinking of Cicero. He thinks - and this corresponded to the actual situation [that] there are genres of poetry in which, under all circumstances, a loftiness prevails, namely, tragedy or grand epic on the one hand, comedy on the other.' 3 In each of them, individual poets (Cicero names Homer and Menander) distinguished themselves as specialists, as it were. In contrast, there is, for the most part, in one and the same judicial or political speech a motivation for the application of many levels; yet that does not mean such application must happen all at once, but rather in alternation according to an intention (docere "to show," delectare "to delight," commovere "to move") . Cicero and Quintilian never taught that one ought to present (docere) the facts in the high style or excite and rouse the audience in the matter-of-fact, lower one. That would amount to a rejection of stylistic differentiation for oratory; but it would appear to them as cacozelia ["affectation of style") or tapeinosis ["lowness of style"). A greater authority than Cicero and Quintilian demanded, by the way, or so it seems, the same from the poet as they did from the orator. At the end of Plato's Symposium it is related how at daybreak, among the many people sleeping, Socrates explained to Agathon and Aristophanes, who were still drinking with him but who were also already half asleep, that one and the same person must know how to compose tragedies and comedies. I believe that one can have confidence in my idea of ancient stylistic differentiation without fear of being misled. The idea is not incautious. The second of my "theses," that of the figuralism of the Christian view of reality, Curtius briefly "repudiated," as he puts it, in another place. ' 3 In the process Cicero (and likewise Quintilian) gives an exact formulation of stylistic differentiation: "in tragoedia comicum vitiosum est, et in comoedia turpe tragicum" ["in tragedy the comic is faulty, and in comedy the tragic is indecent"] . APPENDIX The repudiation, which is directed against my essay on figura (first in Archivum Romanicum 22 [1938], reprinted in Neue Dantestudien, Istanbuler Schriften, no. 5 [Istanbul, 1944)), 14 is found somewhat irrelevantly in a footnote of his work on [Gustav] Grober [1844-1911] (Zeitschrift fiir romanische Philologie 67 (1951) 276-277) and consists substantially of an enumeration of book and essay titles. [He charges] that I had not taken into consideration the research results contained in these writings [and that] if I had done so, my thesis would have become worth discussing. Among the works enumerated are found only two (by [Jean] Danielou [1905-1974] and [Rudolf Karl] Bultmann [1884-1976]) of the specialized theological investigations of typology, which have recently become very numerous. All of these works appeared long after figura; the two specialized works did not appear until four years after Mimesis. What is more, they would not have been accessible to me in Istanbul.' 5 It is also astounding that Curtius mentions among the witnesses against me the work of Bultmann, which makes reference to my work.1 6 More important, however, is that the theological writings on typology- both those named by Curtius and other more recent ones -give me no cause to change anything substantive in my views.' 7 This is because, among ﾷｾ＠ Now reprinted ｢ｾ＠ Francke in Bern. ["Figura" is available in English, translated by Ralph Manheim, in Erich Auerbach, Scenes from the Drama of European Literature: Six Essays (New York: Meridian Books, Inc., 1959; rept. Gloucester, Mass.: Peter Smith, 1973), PP· n-76.] 15 Compare Mimesis, p. 5F I was able to write the works on figura and passio because an entire set of Migne s Patrologia was located in an athc-levef library room of the Dominican monastery of San Pietro di Galata. The monastery library was not public, but the apostolic delegate, Monsignor Roncalli (now papal nuncio in Paris and a cardinal ), had the kindness to grant me use of it. [Born Angelo Giuseppe Roncalli in 1881, the man to whom Auerbach referred as Monsignor Roncalli was apostolic delegate in Turkey and Greece from 1934 to 1944, when he was promoted to new -and difficult duties as papal nuncio to occupied Paris. In 1953 he was created a cardinal. Later, in 1958, Roncalli was elected to the papacy, as Pope John XXIII. The Second Vatican Council (1962-1965) was the achievement for which he is best known. He died on 3 June 1963.] 16 "Ursprung und Sinn der Typologie als hermeneutischer Method," in Pro regno, pro sanctuario: een bundel studies en bijdragen van vrienden en vereerders bij de zestsigste verjaardag van Prof Dr. G. Van der Leeuw, ed. W. J. Kooiman and J. M. Van Veen (Ni jkerk: G. F. Callenbach, 1950), pp. 89-100, also in Theologische Literaturzeitung, (1950) 205ff. I have been in contact with Bultmann for over two decades-contact that was interrupted only by the war; I owe much to his counsel, as well as recently to that of Erich Oinkler. So as to mention a Catholic point of view on the topic as well, compare William F. Lynch in Thought (New York) 25 (1951) 44-47. 17 I will take this occasion to communicate to the readers of this periodical, most of whom are not theologians, those points in my presentation that could give rise to controversies. Both of them relate to the early Christian period. I. In my presentation of the beginnings of Christian typology, the role of Paul is APPENDIX other things, far and away most of them concern themselves with individual questions of sources and with restricted segments of time, whereas my efforts rest upon collections of motifs that I began seventeen years ago and that extend from Paul up into the seventeenth century. What Curtius understands by typological allegoresis, about which I am supposed to have refrained from informing myself, is unfathomable; typological allegoresis is, after all, the subject of my investigations. Whether one calls it so or calls it figural explication is irrelevant. My original terminology came into being naturally through the fact that I proceeded from a semantic history of the word figura. I spoke there extensively about the fluctuation of the terminology in late antiquity and the Middle Ages. The terminology that I first preferred is so practical, and was alive so long in the tradition, that a separate Index figurarum and Index de allegoriis are still found in the second volume of indices in Migne's Patrologia.' 8 But the terminology does not matter, so perhaps emphasized too exclusively. Of course people in the Middle N!.,es were of my view. Compare, for example, such representations as the one of "St. Paul Grinding the Com of the Doctrine of the Prophets in His Mill" on a capital in Vezelay, in Joan Evans, Cluniac Art of the Romanesque Period (Cambridge: Cambridge University Press, i950), fig. i75b. [Abbot] Suger had the same representation painted on a window of Saint-Denis and had the following verses placed there (Erwin Panofsky, Abbot Suger on the Abbey Church of St.-Denis and Its Art Treasures [Princeton, New Jersey: Princeton University Press, i946]; 2d. ed. by Gerda Panofsky-Soergel [Princeton, New Jersey: Princeton University Press, i979J, pp. 74-75): Tallis agenda molam de furfure, Paule, farinam. Mosaicae legis intima nota facis. Fit de tot grams verus sine furfure panis, Perpetuusque cibus noster et angelicus. [By working the mill, you, Paul, take the flour out of the bran. You make known the inmost meaning of the Law of Moses. From so many grains is made the true bread without bran, our and the angels' perpetual food.] On the same window is found a representation, on which the veil has been taken from Moses' face, with this distich: Quod Moyses velat, Christi doctrina revelat; Denudant legem, qui spoilant Moysen. [What Moses veils the doctrine of Christ unveils; They who despoil Moses bare the Law.] 2. Furthermore, there a_ppears sometimes in more recent, specialized works the tendency to ascribe to Origen a significant role for typology, whereas I counted him among the abstract-allegorical interpreters. That is a decisive problem for the conception of typology. I believe that I am right without any alteration of my view, but I must leave the clarification of the question to theologians. 8 ' [The reference is to Patrologiae cursus completus; series Latina, ed. J.-P. Migne, 221 vols. (Paris: J.-P. Migne, i844-1864).] APPE DIX long as one distinguishes clearly between abstract/allegorical and real/ prophetic methods of explication. I have often heard the reproach that I generalize the meaning of the figural or typological principle more than is appropriate, but even so, I had never heard it yet from a medievalist or a historical theologian, apart from Curtius. Unfortunately Curtius has occupied himself little with the subject; earlier researchers of the Middle Ages of the same level possessed great experience in it-not only liturgists and hymnologists, but also figures such as [Konrad] Burdach [1859-1936] or [Karl] Strecker [1861-1945], whose notes on the poems of Walter of Cha till on are a trove of typological information. Among contemporaries, apart from some art historians, it would be fitting to name as an example Ernst H. Kantorowicz [1895-1963].'9 The effect of typology is most certainly just as important and permanent a phenomenon for the medieval structure of expression as is the survival of ancient rhetorical topoi of form and content; it has strengthened in me ever more the conviction, and that conviction has been confirmed through discussions with experts in the material, that typology is the real vital element of Bible poetry and hymns, or, even more, of almost the whole Christian literature of late antiquity and the Middle Ages, as also of Christian art, from _ the sarcophagi down to the end of the Middle Ages - and sometimes beyond. Politically, too, to establish or deny claims to power, it played a significant role over many centuries. Curtius certainly has the right, in his synthetic researches into the Middle Ages, in so monumental a subject, to limit himself to the points of view that interest him particularly; but the neglect and inadequate treatment of the problem of allegory (in the broadest sense) must be emphasized. In this context I have stated that Curtius' claim is misleading and substantially incorrect that "in his study of the sacred text [Augustine] persisted in the antiquarianizing and allegorizing method which Macrobius had applied to Cicero and Virgil" (European Literature and the Latin Middle Ages, p. 74). In opposition to that, Curtius refers me to the fifth chapter of the third part of [Henri Irenee] Marrou's book on Augustine. This chapter bears the heading "La Bible et les lettres de la decadence." But a person can get better instruction about Augustine's exegesis from Augustine than from 20 ' 9 [For a study, see Yakov Malkiel, "Ernst H. Kantorowicz," in On Four Modem Humanists: Hofmannsthal, Gundolf, Curtius, Kantorowicz, ed. Arthur R. Evans, Jr. (Princeton , New Jersey: Princeton University Press, i970), pp. i46-219.] [By translating the German phrase "antiquarisch spielenden" (which Auerbach quotes) as "antiquarianizing,'' Trask loses a nuance of playfulness.] 0 ' APPE DIX Marrou. His posing of the problem motivated him to an overly sharp development of the influence on Augustine of late-antique erudition, an influence that is incontestable in itself. Yet even in his significant, but one-sided and not always insightful book, Marrou did not and would not ever have used a formulation such as that of Curtius. Later he published an addition to the book, entitled Saint Augustin et la fin de la culture antique: Retractatio (Paris: E. De Boccard, i949). There one can read on page 646: "If tl1ere is a chapter the inadequacy of which I deplore today, it is in fact the one I dared to entitle 'La Bible et les lettres de la decadence.'" Many reviewers have ascribed to the book, in praise or blame, tendencies that were far removed from me: that the method of the book is sociological, even that the tendency was socialist; that it is focused all too much on the Middle Ages, but also the opposite: it is antimedieval and anti-Christian; that it is wholly pro-Romance, especially pro-French, neglects German, [and] is unjust toward German literature. But there have also been patriotic readers who have congratulated me on the observation that the tragic in the Hildebrandslied and in the Nibelungenlied is deeper than that of Roland. One reviewer concluded on the basis of the first paragraph of the Roland chapter that I am an enlightened pacifist. Here I will go into only one of these questions, namely, tl1e relation of the book to German literature and culture, and in fact chiefly because in tl1at context one misunderstanding can be dispelled. World history has made it so that one in my situation can scarcely speak on this topic without hurting someone's feelings. I will take the risk anyway. The preponderance of Romance material in Mimesis is to be explained not only because of the fact that I am a Romanist, but rather above all because in most periods the Romance literatures are more representative of Europe than are, for example, the German. In the twelfth and tl1irteenth centuries France took unquestionably the leading role; in the fourteenth and fifteenth centuries Italy took it over; it fell again to France in the seventeenth, remained there also during the greater part of the eighteenth, partly still in the nineteenth, and precisely for the origin and development of modern realism (just as for painting). It would be erroneous to read between the lines of my selection [any] preferences or aversions of a fundamental kind - and equally wrong to see estrangement or aversion in the regret or criticism that APPE DIX occasionally comes to be expressed about certain limitations of outlook in German literature of the nineteenth century. The opposite would be more accurate. The criticism comes out of sorrow over missed possibilities to give a different direction to European history. The great French novelists are, for the posing of the problem in Mimesis, of crucial significance; my admiration for them is great. But for pleasure and relaxation I prefer to read Goethe, Stifter, and Keller. It has been said that I acquired my category of stylistic mingling from modern French realism, and indeed one could deduce that from the epilogue of Mimesis. However, the arrangement of this is chronologically misleading. The motif of a stylistic break became apparent to me first in the story of Christ, during my Dante studies in the i92os; one finds it in Dante als Dichter der irdischen Welt (which appeared at the end of i928), pages 18-23. Shortly after the appearance of this book I began to teach in Marburg, and the teaching activity led me back to French, which I had rather neglected during my years as a librarian, in which I worked on Vico and Dante. While preparing a course of lectures in Marburg, the thought came to me that one could present the principle of modern realism in corresponding fashion; it was then published in two essays that appeared in 1933 and 1937-'3 There is yet another side to the matter: Mimesis attempts to comprehend Europe, but it is a German book not only on account of its language. Anyone who is a little familiar witl1 tl1e structure of the humanities in various countries sees that at once. It arose from the themes and methods of German intellectual history and philology; it would be conceivable in no other tradition than in that of German romanticism and Hegel. It would have never been written without the influences that I experienced in my youth in Germany. 24 21 22 " [Translated into English by Ralph Manheim, Dante, Poet of the Secular World (Chicago: University of Chicago Press, 1961).] [After passing the Staatsexamen in 1922, Auerbach acquired training in library science. From 1924 to 1929 he was salaried as a librarian at the Prussian State Library in Berlin. After completing an abridged translation of Giambattista Vico's The New Science in 1924, a collaborative translation of Benedetto Croce's introductory study of Vico in i927, and a book of his own on Dante in i929, Auerbach was transferred to the University Library in Marburg. In i920 he was appointed to a professorship in Romance philology at the University of Marburg.] '' "Romantik und Realismus," in Neue Jahrbilcher fi.ir Wissenschaft und Jugendbildung 9 (1933) i43ff., and "Ober die ernste ｎ｡｣ｨｾｭｴｧ＠ ､･ｾ＠ A1ltaglic!1en," in Travaux du Seminaire de Philologie Romane, vol. 1 (Istanbul : Istanbul Oniversitesi Edebiyat Faki.iltesi , 19?7), 2foff. "' An unfnendly and also unpleasant review begins with the claim that Mimesis has been greatly discussed and praised especially abroad [outside German-speaking countries]. That gives a false impression. Of the reviews and oilier extensive assessments that 12 57 1 APPE DIX It has often been said that my conceptualization is not unambiguous and that the expressions that I use for organizational categories required a sharper definition. It is true that I do not define these terms, in fact even that I am not consistent throughout in using them. That happened intentionally and methodically. My effort for exactitude relates to the individual and the concrete. In contrast, the general, which compares, compiles, or differentiates phenomena, ought to be elastic and flexible; to the utmost that is possible, it ought to fall into line with what is feasible from case to case, and it is to be understood from case to case only from the context. There is not in intellectual history identity and strict conformity to laws, and abstract, reductive concepts falsify or destroy the phenomena. The arranging must happen in such a way that it allows the individual phenomenon to live and unfold freely. Were it possible, I would not have used any generalizing expressions at all, but instead I would have suggested the thought to the reader purely by presenting a sequence of particulars. That is not possible; accordingly I used some much-used terms, like realism and moralism, and, compelled by my subject, I even introduced two little-used ones: stylistic differentiation and stylistic mingling. That they all, but especially the much-used words, signify all and nothing was perfectly clear to me; they should acquire their meaning only from the context, and in fact from the particular context. That has obviously not always worked out. Almost all misunderstandings have arisen because, all the same, the reader has precisely the possibility to release the schema of a concept from the context and to hold fast pedantically to it; and thus, to give an example that has not been mentioned yet, he can hold against me that he finds Phedre more realistic than Madame Bovary. A good writer must write in such a way that one infers from the text what he intended to express. That is not easy. Earlier I believed that one could devise words and collocations that comprehend the general in tl1e historically intellectual more exactly than do the usual ones, and I tried it with "popular spiritualism," "dialectic of feeling" ((Karl] Vossler (1872-1949]), and "serious imitation of the everyday." But that only leads to new misunderstandings and, what is more, sounds pretentious and pedantic. It is in I have seen up until now, over half appeared in Germany or in German-speaking Switzerland; of the remaining foreign ones, once again almost half (mostly in the U.S.A.) were composed by such individuals who had spent their youth in German}' and received their education there. The rest are distributed among Scandinavia, Holland, Belgium, the Spanish-speaking world, and Turkey. Only a few remarks have come to my attention from France, not a single one from England. 57 2 APPE DIX the nature of our subject that our general concepts are poorly differentiable and are undefinable. Their worth - the worth of concepts such as classic, Renaissance, mannerism, baroque, enlightenment, Romanticism, realism, symbolism, and so forth, most of which originally designate literary epochs or groups, but which are also applicable far beyond those - accordingly, their worth consists in that they elicit in readers or hearers a series of ideas that facilitate for them an understanding of what is meant in the particular context. They are not exact. The attempts to define them, or even only to collect completely and without contradiction those characteristics that compose them, can never lead to the desired result- even though they are often interesting, for the reason that someone produces in the discussion a new point of view and thereby assists in the enrichment of our ideas. One must beware, it seems to me, of regarding the exact sciences as our model; our precision relates to the particular. The progress of the historical arts in the last two centuries consists above all, apart from the opening up of new material and in a great refinement of methods in individual research, in a perspectival formation of judgment, which makes it possible to accord the various epochs and cultures their own presuppositions and views, to strive to the utmost toward the discovery of those, and to dismiss as unhistorical and dilettantish every absolute assessment of the phenomena that is brought in from outside. This historical perspectivism was founded by the pre-Romantic and Romantic critics; since then it has turned out to be very refined and ever more complicated, through insight into a great number of previously unknown or unheeded developments, influences, and relationships. A person with a classificatory taxonomy that works with exact and set conceptions of order cannot succeed in drawing together the aspects that intersect multiply into a synthesis that does justice to the subjects. Another objection that people have made is this: that my presentation is all too time-bound and all too much determined by the present. That is also intentional. I tried to make myself thoroughly conversant with the many subjects and periods that are treated in Mimesis. With a deliberate extravagance of time I studied not only the phenomena that had direct significance for the aim of the book, but I read around widely in the various periods. But in the end I asked: How do matters look in the European context? No one today can see such a context from anywhere else today than precisely from the present, and specifically from the present that is determined by the personal origin, history, and education of the viewer. It is better to be consciously than unconsciously time- 573 APPENDIX bound. In many learned writings one finds a kind of objectivity in which, entirely unbeknownst to the composer, modern judgments and prejudices (often not even today's but instead yesterday's or those of the day before yesterday) cry out from every word, every rhetorical flourish, every phrase. Mimesis is quite consciously a book that a particular person, in a particular situation, wrote at the beginning of the i94os. 574