The second edition of this book incorporates some minor alterations, corrections and additions of material. Moreover, I have updated the notes and the bibliography to take account of several important books which were not in print when the first edition went to press, most notably Judson B. Allen’s The Ethical Poetic of the Later Middle Ages (Toronto, 1982) and Glending Olson’s Literature as Recreation in the Later Middle Ages (Ithaca and London, 1982). Writing in 1980 (the year in which the first edition was actually completed) I complained that ‘The study of late medieval literary theory is still in its infancy’ (see p. 3 below). There has been considerable growth since then, thanks in large measure to the two books just mentioned, but the subject has still a long way to go before it reaches full maturity.
My main objective in writing was to demonstrate the considerable importance of scholasticism for the development of literary theory, a phenomenon which earlier writers had either ignored (leaping from twelfth-century ‘humanistic’ literary theory to that produced in the early Renaissance) or belittled and misunderstood. In particular, scholastic Scriptural exegesis was a central force in the re-shaping of literary values in the later Middle Ages. The central event, from my point of view, was the emergence (in Bible commentaries) of the view that the human author possessed a high status and respected didactic/stylistic strategies of his very own—in short, auctoritas moved from the divine realm to the human.
The ramifications of this event were many and varied, and so I limited myself, for the most part, to two areas of inquiry. The first was, the main implications thereof for Scriptural exegesis from the time of the Summa theologiae associated with Alexander of Hales until the time of Nicholas of Lyre and the ‘classicizing friars’ (as they were dubbed by the late Beryl Smalley). The second chosen area was the influence of a few aspects of scholastic literary theory on vernacular writers of the later Middle Ages, with special reference to two major English authors, Chaucer and Gower. The fact that my final chapter focuses on those two writers should not be taken as a tacit statement that scholastic literary theory is more important for the study of literature in Middle English than for that in any other vernacular—for this is certainly not the case. Italy and Spain yield far more obvious data than England does. Moreover, medieval Latin literature and scholastic culture influenced all the emergent vernacular literatures of the later Middle Ages, and so to pick on one may give a false impression. But one must start somewhere, and I chose to begin with Middle English literature, even though I recognized then (and my subsequent research has overwhelmingly confirmed) that the richest vein to mine was the outstanding literary theory and criticism of Dante, Petrarch and Boccaccio, so innovative and yet so traditional in many ways. Anyone interested in that area of enquiry can now turn to the texts and annotation in the relevant sections of A. J. Minnis and A. B. Scott (eds.), Medieval Literary Theory and Criticism c. 1100–c. 1375: The Commentary Tradition (Oxford, 1987). When, in literary discourse, auctoritas moved from the divine realm to the human, and (as an intriguing concomitant) in some measure from the past to the present, certain ‘moderns’ found in it a means not only of describing recent literary enterprises (including their own) but also of elevating them. This translatio auctoritatis, to coin a phrase on the model of the term translatio imperii, was one of the most significant movements in the history of vernacular literature. And nowhere was the sense of change felt more keenly than in late-medieval Italy.
Since it is the implications of scholastic literary theory for vernacular literature which seem to have been the main source of interest to readers of the first edition of this book, it seems appropriate, in this the Preface to the second edition, to offer a few comments on the kinds of relationship which may be posited between that theory and the literary practice of some of the moderni. Here we are approaching, of course, the ultimate question: to what extent should twentieth-century critics take medieval literary theory and criticism into account in the contemporary appreciation of medieval literature? It is hoped that the following categories and examples will serve as pointers for further investigation, despite the summary form in which they must necessarily be presented here.
Commentary as source. In many cases, medieval commentaries on classical and Scriptural auctores were the direct sources of certain statements, attitudes, images, etc., as found in the works of major vernacular writers. Late-medieval vernacular versions of the Bible, in part or whole, clearly reveal the influence of Latin Scriptural exegesis; very often a gloss would provide a translator with details not found in the text itself, or determine his preference for a particular word, phrase or idiom, in rendering a certain passage. For instance, Peter Lombard’s commentary on the Psalter is reflected by Richard Rolle’s Middle English psalm versions, while the Glossa ordinaria on the Psalter and especially Nicholas of Lyre’s superb exposition thereof lie behind the Lollard Psalter in its ‘second version’ (as Harry Hargreaves has demonstrated). The influence of Latin commentaries on secular authorities was just as pervasive. The anonymous author of the Old French Les faits des Romains used a glossed copy of Lucan’s Pharsalia; the French translation of Ovid’s Ars amatoria which recently has been edited by Bruno Roy was influenced by, and is accompanied by, glosses, which are rendered in the vernacular to provide a running commentary on the text. Jean de Meun consulted William of Conches’s popular commentary on the De consolatione philosophiae when writing the Boethian passages in Le Roman de la Rose and subsequently in making his complete French translation of the Consolatio. Nicholas Trevet’s commentary on Boethius was especially popular among English writers: its influence is writ large in Chaucer’s Boece and it has also been detected in Boethian passages in such works as Troilus and Criseyde and The Monk’s Tale. Trevet on Boethius was used also by John Walton in making his all-verse English Boethius (completed 1410), and by Robert Henryson, grammar master at Dunfermline in the early fifteenth century, who found in Trevet’s exposition of the story of Orpheus in the underworld the inspiration, and much of the information, for his Middle Scots poem Orpheus and Eurydice. Glosses on, and medieval adaptations of, Aesop suggested to Henryson touches which appear in both the narrationes and the moralizationes in his Moral Fables of Aesop. Then again, clear evidence of the influence of glosses on Ovid’s Heroides may be found in certain stories included in Chaucer’s Legend of Good Women and John Gower’s Confessio amantis. The class of commentary known to Chaucer has even been identified (by M. C. Edwards).
This list of examples could be extended indefinitely, but enough has been said, we trust, to make the point that it is quite crucial for modern scholars to be aware of the manuscript contexts of the Latin sources of vernacular literature. One might go so far as to say that it is the original text together with its accompanying commentary (often, it must be remembered, written around the actual text in the manuscript margins) that should be regarded as the source. At any rate, it is obvious that some effort should be made to divest oneself of assumptions engendered by the reading of texts in modern editions, apart from their medieval apparatus of glosses (the ‘Loeb Classics’ syndrome). How can one possibly begin to ascertain what a major writer like Dante or Chaucer is doing to his source-text unless one is aware of how that text had been expounded and elaborated in medieval scholarship of a kind readily available to (and often demonstrably consulted by) the writer concerned?
Academic prologue as literary model. It is obvious that in many cases writers modelled their own prefaces on one or other of the types of prologue which they had encountered during their grammar-school education (wherein standard prologue headings provided the basis of the introductory lecture on each and every ‘set text’) and in subsequent readings of commentaries in manuscripts.
Some of the vernacular prologues of this type are, as one would expect, either straight translations or faithful adaptations of their Latin equivalents, as is the case with the following Middle English examples: the prologue to the translation, attributed to John Trevisa, of Giles of Rome’s De regimine principum, the English version of the standard ‘Paris Bible’ prologue to the Apocalypse, the prologue to Rolle’s English Psalter (see pp. 190–91 below), and the accessus in an English commentary on the first book of the Consolatio philosophiae as translated by Chaucer (preserved in Oxford, Bodleian Library, MS Auct. F.3.5), which owes its basic form to the ‘type C’ prologue as described in our first chapter. Others are far more original in purpose and application, good examples being the prologue to the Old French Livre du Chevalier de la Tour, the Latin prose preface (in some measure modelled on twelfth-century type accessus to Ovid) to Juan Ruiz’ Libro de buen amor, the introductions to such reference-books as Vincent of Beauvais’s Speculum maius and Pierre Bersuire’s Reductorium morale, all the prologues used by John Gower in his Latin poem Vox clamantis (on which see pp. 168–77 below), the Messenger’s introductory speech in the English morality play Everyman (cf. p. 272 n. 14), and many of William Caxton’s prologues and epilogues, including his preface to Malory’s Morte Darthur (cf. pp. 206, 279 n. 117). For the various and varied works introduced by ‘Aristotelian prologues’ see pp. 161–2. Sometimes the prolegomena may be of a more complicated type, a good example being afforded by the beginning of Gower’s Confessio amantis, on which see pp. 177–81.
There is, then, plenty of evidence to suggest that major writers of the later Middle Ages found in academic prologues idioms which they regarded as being sophisticated enough (and of course distinguished enough) to provide a basis for the description and justification of their own writings—although it must be recognized, of course, that often they made quite unusual use of traditional materials. Knowledge of the traditions of the academic prologue is, therefore, quite indispensable in our own attempts to comprehend and assess those same descriptions and justifications.
‘Modern’ commentary and ‘self-commentary’. Thus far we have been considering how certain medieval writers responded to the elements of medieval commentary, i.e. the particular glosses and the general prologues. But the influence of the commentary tradition did not stop there, since it was often drawn upon by those who wished to provide vernacular texts with an apparatus which at once described certain aspects of those texts and tacitly claimed a degree of prestige for them (because that apparatus was of the type which conventionally had accompanied the works of the revered auctores). A good case in point is the mise-en-page of the most popular of all the vernacular translations of the Consolatio of Boethius, the ‘Anonymous Verse-Prose Version’ or ‘Revised Mixed Version’ as it is variously called (which has been studied intensively, and is being edited, by Glynnis Cropp). In some manuscripts the text is divided into short sections, each of which is followed by a section of French commentary, often marked ‘glose’ (the information coming, it would seem, from some version of the William of Conches commentary on Boethius). Here features of layout which had been developed in generations of scholastic manuscripts are being used in presenting a translation of an auctor to an audience of wealthy layfolk.
Of even greater significance for the history of literary theory and criticism are the changes of attitude revealed by the treatment of vernacular works which, while in many instances still heavily dependent on Latin models, were beginning to demand attention of a kind which hitherto had been afforded only to established Latin authorities. Academic commentary became a precedent and source for ‘new’ commentary and even ‘self-commentary’: certain moderni set about the business of producing commentaries on ‘new’ works, works written by their medieval contemporaries and indeed by themselves.
Some of these appropriations of method and matter were more daring than others. Certain moral and didactic vernacular works received, quite naturally, moral and didactic glossing, the latter serving to enhance the prestige of the former with little sense of disjunction existing between text and gloss. A good example of this is provided by what is, to the best of my knowledge, the most heavily glossed original work written in Middle English: two copies of the mid-fifteenth-century Court of Sapience contain an extensive and erudite apparatus of Latin glosses, probably provided by the writer himself. The Echecs amoureux, by contrast, was something of a risqué poem, yet the person responsible for the French commentary on it—probably the first commentary to be written on any original poem in Medieval French—is hardly self-conscious about his hermeneutic activity, since he limits his justification to the claim that he is bringing out the profit and delight offered by the text under scrutiny. In La Vita nuova Dante applied to his own poetry certain scholastic techniques of textual exposition (most notably a type of divisio textus which grinds exceedingly fine) with apparent ease and freshness, without drawing attention to the pedagogic contexts which were their normal environment.
The glosses (in Italian) which Boccaccio appended to his Teseida are, however, a very different case, and a special one. This Italian epic was studiously equipped with apparatus of a kind which accompanied its Latin models in manuscript (e.g. the scholia on the Thebaid, the Aeneid and that highly popular Medieval Latin facsimile of a classical epic, Walter of Châtillon’s Alexandreis), an apparatus calculated to dispose the discerning reader in favour of the poem and underline for his benefit the superlative literary criteria in accordance with which it had to be judged and esteemed. Here, techniques of exposition traditionally used in interpreting ‘ancient’ authorities are being used to indicate and announce the literary authority of a ‘modern’ work. No criticism is value-free (to cite a commonplace of contemporary literary theory), the truth of which is borne out by the glossing of the ‘modern’ commentators and ‘self-commentators’ of the later Middle Ages, who sought to aggrandize vernacular literature by applying to it a value-bestowing type of criticism.
The most articulate of all the self-commentators was certainly Dante. Il Convivio constitutes at once a commentary on three of his own canzoni and on the methods of commentary themselves. Dante gives a fascinating explanation of the reasons why he decided to write an Italian, rather than a Latin, commentary on Italian texts, and his famous explanation of the two types of allegorical exposition, the ‘allegory of the poets’ and the ‘allegory of the theologians’, is found at the beginning of his exposition of the first poem, Voi che’ntendendo. Literally, this canzone is supposed to record the conflict in Dante’s mind between the memory of the lost Beatrice and a gentle lady—usually identified as the ‘lady of the window’ of La Vita nuova—whose pity sought to console him. Allegorically, this new love is Lady Philosophy (here Dante is, as he himself admits, influenced by the female personification created by Boethius); no rival to Beatrice, but rather a means towards her now-glorified self. Twelfth-century Bible commentators had used allegorical interpretation to establish the auctoritas of holy Scripture (as is explained in Chapter 2 below): here Dante, with a similar purpose and result, is applying the very same method to Voi che’ntendendo. Once again, appropriation of hermeneutic idioms and techniques may be identified as a strategy for the translatio of literary auctoritatis from Latin into the vernacular.
Medieval literary theory as interpretive aid. It has already been suggested that, with regard to textual details, knowledge of how a writer’s source-text was understood by medieval scholars can be of considerable importance in our own attempt to grasp the significance of what he wrote. On many occasions it is possible to go much further than this, and claim that knowledge of how a given source was integrally interpreted and evaluated in medieval scholarship (or ‘medievalized’, as some would call it) can provide vital clues to the plans and purposes of some of the writers who drew on that source.
For example, the last book of the De amore (1184–6) of Andreas Capellanus, in which the addressee Walter is advised to refrain from the love of women, has been described by P. G. Walsh, the work’s most recent editor, as an utter volte-face which has no real precedent in Ovid: ‘The chasm between Andreas’ first two books and the third is thus much more profound than is that between Ovid’s Ars and his Remedia, in both of which the mock-didactic tone is uniform throughout’. However, as the accessus to Ovid make perfectly clear, medieval readers were taught to consider the Remedia as a genuine retraction, an attempt to make amends for the Ars which (so the story goes in the vitae Ovidii) had misled many men and women into reprehensible amours. Those same medieval readers, it may be inferred, would have regarded Andreas’s transition as more essentially Ovidian, and hence more conventional and precedented, than his modern editor would allow. I endorse Walsh’s remark that Andreas is proclaiming himself ‘the medieval magister amoris, a twelfth-century Ovid’, but would wish to add that the Ovid whom Andreas was imitating was the twelfth-century Ovid, i.e. Ovid as interpreted in that century.
Knowledge of the way in which Ovid was commonly interpreted is of considerable value in our approach to another work which owes much to that auctor, namely John Gower’s Confessio amantis. How can a work which professes to treat of human love admit consideration of such political matters as the qualities required in a good ruler and the current sorry plight of the ‘three estates’ in medieval society? Gower’s. basic rationale becomes clearer when it is remembered that, in medieval literary criticism, Ovid’s love-poetry was described as ‘pertaining to ethics’ (ethice subponitur). As one branch of practical philosophy as defined by Aristotle, ethics is related to politics, which is another branch of the same subject (cf. pp. 177–90). In sum, the Confessio amantis seems to operate through an assimilation of materials which, although they may appear heterogeneous to the modern reader, would have been regarded as quite compatible by the medieval reader who had learned his Ovid in grammar school.
These two cases may be regarded as manifestations of the late-medieval attitude to intentio auctoris, the intention of the author, in this case Ovid—by which was understood not any highly personal preoccupations but rather the author’s pedagogic purpose, whether explicit or implicit; what was meant in ultimate terms. Medieval commentators were more interested in Truth than in the personal and subjective areas wherein we individualistic moderns are accustomed to seek the true intention of an author. Should we, then, moralize medieval texts in this manner, on the ground that, as an authentically medieval interpretive method, it must reveal the intended meaning of those texts? Our answer must surely be in the negative. We can treat a medieval work in the way in which medieval commentators treated their authorities only if the work in question clearly demands such treatment. Otherwise we are simply perpetuating the medieval interpretive system in an unthinking and highly reductive kind of way, with little regard for that rich diversity of style and sense which is such a marked feature of medieval textuality.
Since this point is of the first importance, we must dwell on it for a moment. The last major attempt to relate medieval hermeneutics (particularly Scriptural exegesis) to vernacular literature, led by D. W. Robertson, produced an interpretive determinism, and there seems to be little scholarly advantage in replacing one kind of interpretive determinism with another, by making pan-allegorization give way to panmoralization. And that is why I have reservations about one aspect of the late Judson Allen’s stimulating book The Ethical Poetic of the Later Middle Ages. ‘For the Middle Ages’, Professor Allen wrote, ‘meanings tended to be multiple but not at all random or unpredictable; they existed in an array of possibilities’. Once this ‘normative array’ is defined—and we are assured that relatively few levels of meaning are possible—the critic is supposed to possess the definitive interpretive programme for revelation of the meaning of complex medieval poems. In a far more controversial book, which Allen wrote in collaboration with T. A. Moritz, these principles are pursued relentlessly in an interpretation of Chaucer’s Canterbury Tales based on what medieval commentators had to say about the Metamorphoses; in particular, the practice of classifying Ovid’s stories according to whether they were natural, moral, magical or spiritual is deemed to provide the system with which Chaucer’s stories must also be classified. In this regard it is necessary, I feel, to enter some caveats.
To claim or imply that late-medieval moral interpretation should become a critical ‘skeleton key’, able to open any literary lock, is going too far. The truth seems to be far more complex and pluralistic, and therefore less reducible to rule or ‘normative array’. What in the hands of medieval commentators was an evaluative method became in the hands of certain practising poets a possible literary strategy. By ‘possible strategy’ is meant a certain well-defined literary procedure which medieval poets could use but did not feel obliged to use. Sometimes that strategy is dominant, as for example in the attempts by John of Garland (Morale scolarium) and John Gower (Vox clamantis) to produce satire according to the medieval understanding of the genre, which involved the commendation of virtue and the castigation of vice. Sometimes the strategy is one among several, perhaps the best example being Juan Ruiz’ Libro de buen amor, in which critical commonplaces relating to the ethical function of love-stories (derived, most likely, from accessus to Ovid’s Ars amatoria and Remedia amoris) are interspersed with anecdotes of a lascivious and naturalistic kind; here the impeccably pious is juxtaposed with the energetically profane. Ruiz, it would seem, is providing instant commentary on his own text; the ‘good men’ in the audience (to echo Chaucer’s Nun’s Priest) can take the morality if they want to. But only if they want to.
The problem becomes even more acute when we consider late-medieval self-commentary. In Il Convivio Dante provides us with a learned philosophical and in part allegorical interpretation of a text (Voi che’ntendendo) which seems originally to have been written as a poem of human love. In the Latin glosses on his Confessio amantis, Gower provides us with the basis of an exclusively moral interpretation of his lover’s confession. Neither writer seems to have envisaged his commentary as revealing the determinate meaning of his text, but rather one possible meaning thereof. The fact that the moral understanding is the one which any wise man would adopt does much to recommend it but does not enshrine it as the one and only ‘real’ understanding. A single text can have different kinds of meaning, depending on the kind of reader, or the kind of reading in which the reader is engaged at a given time. Conversely, in the hands of different writers a single story can take on different significations (as is obvious from medieval manipulations of exempla, for instance), and some writers imposed more semantic restrictions than others. This is, to be sure, at variance with the spirit of certain forms of modern ‘New Criticism’, wherein interpretation of a text is envisaged as being rather like many people climbing a single mountain, confident that, on its sun-kissed summit, they will all meet, shake hands and reconcile their differences, discussing (often wryly, for some took wrong turnings) the relative merits of the several routes to the top, but now united in the conviction that they share the secret of F6 (or wherever). But medieval textuality seems to offer a veritable range of mountains which are very different in shape and size, and hence demand different strategies: a method which ensures success with one may spell disaster with another. Any one of these mountains, moreover, may have several peaks, of varying or equal importance as the case may be—and certain climbers will turn out to be blind to some of them. Indeed—and here our metaphor breaks down completely—one often perceives a medieval reader/author making his own mountain in the very act of climbing.
Sometimes medieval poets thought and wrote like medieval commentators, and sometimes they did not; on occasion they assumed what the Epistle to Can Grande della Scala called the ‘office of the commentator’ (lectoris officio) but elsewhere quite different stances were adopted. Certainly, the ethical justification and moral interpretation of poetry were of crucial importance in the later Middle Ages, and every major poet took them seriously (a claim which quite definitely cannot be made for the ‘allegory of the theologians’). But how they informed poetic purpose and plan was a matter for the individual poet. In short, much medieval textual meaning was pluralistic in nature. And such pluralism of meaning makes interpretive determinism impossible.
Medieval literary theory and criticism, of which medieval theory of authorship forms just one part, should not, therefore, be reduced to a rigid interpretive programme which brooks no exceptions to its rules. That attitude cannot be justified either diachronically or synchronically: it is indefensible in historical terms and also in terms of the complexities presented by so many of the most interesting texts. There is no skeleton key. What is offered are bunches of specific keys which fit specific locks: to refuse to use them would be surely perverse. Of course, literature is not firmly controlled by the literary theory contemporaneous with it (to think otherwise is, in my view, to be very naive about the nature of literary theory), yet, particularly in the case of writers whose works were demonstrably influenced thereby, such theory can assist us in identifying the criteria in accordance with which those works should be judged. Thus, theory becomes part of the process of contextualization which is necessary for the meaning of a text to be grasped. There is little credit to be gained in assessing texts with reference to an alien system of value, in punishing them
under a foreign code of conscience.
The words of a dead man
Are modified in the guts of the living.
(W. H. Auden, ‘In Memory of W. B. Yeats’)
Cultural change is one thing: cultural imperialism is something else. One can only hope that greater awareness of medieval literary theory and criticism will help us to go back to the texts and their contexts with the desire to listen and learn, not to shout down and dominate.
It is a great pleasure to acknowledge those scholars who have helped me to pursue my research. My greatest debt, personal as much as academic, is to Mr M. B. Parkes. To Malcolm I owe (among other things) hours of stimulating and companionable discussion, precise reading of every draft of first my thesis and then my book, and constant encouragement. Those who know the late Beryl Smalley’s work well will recognise her as the ‘primary efficient cause’ of the present book. Indeed, she managed—a feat unprecedented in my scholastic prologues—to combine that office with the role of instrumental cause, in so far as she provided invaluable information and advice, and carefully perused my material in an earlier form.
Generous advice and information on specific matters was provided by the following scholars: the late R. W. Hunt, Richard H. Rouse, the late Judson B. Allen, James J. Murphy, Éamonn Ó Carragáin, Nigel Palmer, Pamela R. Robinson, Pamela de Wit, the late J. A. W. Bennett, John Burrow, Paul Miller, the late Elizabeth Salter and the late William A. Pantin. I am very grateful to Dr Brian Scott for his considerable and meticulous help with passages of medieval Latin, and to Dr Alan Press and Miss Evelyn Mullally for discussion of some passages of Old French. To Dr Scott and Miss Robinson I owe careful proof-reading.
I was particularly fortunate, during my time at The Queen’s University of Belfast, in being able to draw on the resources of a unique department, the Department of Scholastic Philosophy. Professor Theodore Crowley helped me to attain that familiarity with medieval philosophy and theology which was the main basis of my research, and his successor, Professor James McEvoy, answered a host of technical queries which arose in the course of my work.
Generous research grants from The Queen’s University of Belfast made much of my research possible. An award from the British Academy enabled me to study in Paris libraries during the Summer of 1978.
Some of the material included in this book has appeared, in different forms and in different contexts, in the following periodicals: Beiträge zur Geschichte der deutschen Sprache und Literatur, Medium Ævum, Mediaeval Studies, and New Literary History.
Photographs are reproduced here by kind permission of the Trustees of the British Library and the Curators of the Bodleian Library, Oxford. For permission to print brief extracts from manuscripts I am indebted to the following: the Conservateur en chef of the Bibliothèque Nationale, Paris, the British Library Board, the Curators of the Bodleian Library, Oxford, the President and Fellows of Trinity College, Oxford, the President and Fellows of Corpus Christi College, Oxford, the Warden and Fellows of New College, Oxford, the Master and Fellows of Balliol College, Oxford, the Warden and Fellows of Merton College, Oxford, the Syndics of the University Library, Cambridge, the Master and Fellows of Corpus Christi College, Cambridge, the Master and Fellows of Christ’s College, Cambridge.
Finally, some very personal acknowledgements. My wife Florence spent many hours discussing ideas, reading and improving drafts, compiling the index, and in general hastening the progress of this book. I have dedicated my work to my parents in grateful recognition of their constant understanding and support. It is impossible to describe my debt to my mother, who died in 1972, for the intellectual companionship which I enjoyed with her since I was old enough to think.
Bristol, Winter 1986
A.J.M.