SOURCE: Runacres, Charles. “Art and Ethics in the ‘Exempla’ of ‘Confessio Amantis’.” In Gower's Confessio Amantis: Responses and Reassessments, edited by A. J. Minnis, pp. 106-34. Cambridge: D. S. Brewer, 1983.

[In the following essay, Runacres investigates the “fruitful balance between pleasurable instruction and instructive pleasure” in the Confessio Amantis.]

At the beginning of the ‘Prologus’ to Confessio Amantis, Gower announced that he was relinquishing the single-minded pursuit of ‘wisdom’ (‘Prologus’ 13) which had characterised his earlier poems. His new work would mingle delight with the profit, in such a way that the matter of the poem would be both instructive and pleasurable. The explicit ethical analyses and statements of the ‘Prologus,’ Book VII and the end of Book VIII, and the great structure of vices and virtues expounded by Genius, coexist with the delights of the amatory situation of Amans and those of the multitude of stories. Gower's first idea, as he explained it in the poem itself, was that the ‘lust’ and the ‘lore’ of the work should be distinct, each appealing to a different audience.

And in such wise as I ferst tolde,
Whan I this bok began to make,
In som partie it mai be take
As for to lawhe and forto pleye;
And forto loke in other weye,
It mai be wisdom to the wise:
So that somdel for good aprise
And eek somdel for lust and game
I have it mad.

(VIII 3054-62*)

This view of the poem has proved appealing to many modern commentators, who have argued that there are clear divisions between kinds of subject matter in the Confessio.¹ In his most fully developed view of the poem, however, Gower insisted on the existence of a balance between ‘ernest’ and ‘game’ in the work. Profit and delight were not to be seen as discrete, co-existent parts of the work, but as interdependent aspects of a unity that combined instructive pleasure and pleasurable instruction.

Touchende that y undirtok
In englesch forto make a book
Which stant betwene ernest and game,
I have it made as thilke same.

(VIII 3107-10)

Such a balance is not inconceivable to a modern reader, but changes in literary taste have made it difficult to understand how Gower tried to achieve it, and therefore difficult to judge the extent to which he succeeded in his attempt. These difficulties have affected our understanding of many parts of the poem, and indeed of the whole, but few parts have suffered as much as the exemplary stories. The use of the stories to support Genius' ethical analysis means that profit and delight are intertwined in such a way that judgements about the nature of their relationship are crucial to any understanding of what is occurring in the poetry. I am therefore concentrating in this essay on the stories, in an attempt to see how Gower worked and thought, and to judge what success he had in his attempt to balance ‘ernest’ and ‘game’.

I. GOWER AND THE EXEMPLUM TRADITION

A great part of Confessio Amantis is organised by the confession of Amans and the structure of the Seven Deadly Sins. Within this structure of categories of vice and virtue, a number of stories are told. These stories are moralised, and made to serve as illustrations or explanations of ethical statements in the structure. The Middle Ages knew such moralised stories as exempla, and it is this term that Gower uses to describe his stories.

To wisse thee, my Sone, and rede,
Among the tales whiche I rede,
An old ensample therupon
Nou herkne, and I wol tellen on.

(IV 73-6)

Here the Confessor advances an exemplum against those who sin in love by delaying.

(IV 80 margin)

In using stories in this way, Gower was doing no more than turning to the most familiar and widely used didactic technique of his day.

Exempla have a long history in the Christian West, for they derive much of their authority from Christ's own use of parables. His practice of explaining Himself by means of stories was copied by the Fathers, who did much to establish the form and subject matter of exempla. St Gregory the Great was especially influential.² Prior to the middle of the thirteenth century, however, exempla were mainly used by inveterate story-tellers such as Helinand of Froidmont and James of Vitry, and were not in widespread favour.³ The large-scale recognition of the value of exempla and an increasingly universal employment of them in debate and instruction of all kinds came about in the course of the thirteenth century, prompted by developments at the end of the twelfth century which had led to an increased interest in popular preaching. The desire to communicate orthodoxy to a wider audience whose education fell short of the clerical had consequences in the Schools, on the form of sermons, and in the two newly-founded Orders, and led directly to a great interest in, and use of, exempla.⁴Exempla attracted so much attention because they were great aids to comprehension and because they made instruction more inviting, both characteristics which made them especially valuable for use with lay audiences. Humbert of Romans provides a typical assessment of exempla, which explains their popularity.

Seeing that, according to Gregory, exempla are more moving than words, are more easily grasped by the mind, are more deeply imprinted in the memory, and moreover are more readily listened to by many, and draw many to sermons because of their attractiveness, it is advantageous to men given the office of preaching to use a great number of such exempla.⁵

Exempla cause instruction to be more firmly grounded in experience and history, and thus make it easier to understand and remember. They also sweeten the didactic pill. By the fourteenth century, exempla were found in all didactic contexts, and were so important an instructional technique that they were being collected and organised for easy reference by teachers and preachers.⁶ For the next two hundred years, exempla were to dominate instruction.

Despite the very great importance of the form, exempla have never been defined very satisfactorily. Welter, the great historian of the form, offered a very general definition.

By the word exemplum they understood, in the broadest sense of the term, a narrative or a little story, a fable or a parable, a morality or description that could serve for proof to support a doctrinal, religious or moral exposition.⁷

Subsequent critics have been uneasy with this,⁸ but I intend to use the term in a similarly broad sense, because of a marked failure on the part of medieval commentators to draw careful distinctions or establish precise definitions.⁹ An exemplum, then, consists of a passage of narrative or descriptive material, a moral or religious lesson, and an explicit link between the two.

For greater clarity in describing the way in which exempla actually work, it is necessary to label the various elements that constitute the exemplum and its context. The section of narrative or descriptive material I call the narracio.¹⁰ Piero Boitani speaks of narraciones being ‘recounted between the statement of the theme and the conclusion’:¹¹ the theme and conclusion, the parts of the exemplum in which is stated the moral or religious lesson that is taught by the narracio, I call the moralitas. In Confessio Amantis, the narracio is the actual narrative, the actions and events that make up a story. Genius' comments, pointing out the ethical meaning of the narracio and directing Amans' attention to what he should understand from it, are the moralitates. To take the case of the ‘Tale of Acteon’ (I 333), the narracio runs from ‘Ovide telleth in his bok’ (I 333, the introduction to the narrative proper that establishes its authority), to:

This Hert his oghne houndes slowhe
And him for vengance al todrowhe.

(I 377-8)

This is all simple narrative. The moralitas is in two parts. One is the introduction of the theme:

Mi Sone, herkne now forthi
A tale, to be war therby
Thin yhe forto kepe and warde,
So that it passe noght his warde.

(I 329-32)

The other is the conclusion, that reiterates the theme and draws a direct lesson from the narracio:

Lo now, my Sone, what it is
A man to caste his yhe amis,
Which Acteon hath dere aboght;
Be war forthi and do it noght.
For ofte, who that hiede toke,
Betre is to winke than to loke.

(I 379-84)

The exemplum consists of just the narracio and the moralitas. A third element can, however, directly affect the significance and force of the exemplum. This is the immediate context of the exemplum, which in Confessio Amantis is that part of the lover's confession in which the exemplum is set. I call this the ‘exposition’. In the ‘Tale of Acteon’, the ‘exposition’ is the discussion of the vices in love that arise through looking, in which Genius explains the vice in general terms (I 309-28). Apart from Genius' explanations, the ‘expositions’ of the Confessio often include the confessions of Amans. It is in the ‘exposition’ that the subject under discussion is identified, and it is the argument of the ‘exposition’ that the exemplum illustrates.

The purpose of using an exemplum is so that some parallel may be pointed out between the situation in the narracio and the ideas in the ‘exposition’. In this way, greater clarity is achieved by providing a concrete example which is easier to understand than the more abstract subject of the ‘exposition’. It is the moralitas which is used to make explicit the parallels. To revert to the example of ‘Acteon’, Genius has stated in the ‘exposition’ that sight is the external sense most likely to lead to vice and corruption. Turning to sight in love, he has stated first that prying eyes often bring hindrance to lovers, and second that Love's arrow enters through the eyes. Sight thus often causes harm. This statement of the vice is fairly abstract, however, so the exemplum is used to help explain. The narracio tells of Acteon's hunting, his looking at Diana, and the punishment that follows. The moralitas focuses the audience's attention onto the part of the narracio where Acteon looks at the goddess, and onto the fact that he is punished, and states that this one instance of disaster through sight is typical of the dangers of looking, at all times. The particulars of Acteon's looking and his punishment are thus shown by the moralitas to have an exemplary value in the context of the ‘exposition’.

Gower was very careful to make his stories suitable for use as narraciones, an aspect of them that has not received much attention. The stories are predominantly pagan, and are mostly taken from Ovid and the two main Troy cycles.¹² The Old Testament supplies the bulk of the rest. The presentation of the stories themselves is designed to make them as useful as possible in an exemplary context. Gower often cites the source, be it ‘Ovide’, ‘Cronique’ or the ‘tale of Troie’, and mention is often made of Genius having found them in ‘olde bokes’. In this way, Gower established that his narraciones came from authoritative sources. The narratives too are suited to Gower's purpose. R. W. Frank has written suggestively about the popularity of abbreviated narratives in the late fourteenth century, and Gower's narraciones certainly have the rapid forward movement of narrative and the swift summaries of action and description necessary for short tales of this kind.¹³ To maintain the brevity, he goes out of his way to ensure that the narraciones unfold with the greatest comprehensibility and logic. This is especially evident in the strict chronology of events in the narraciones, to achieve which Gower would alter sources.¹⁴ He also succeeded in including a remarkable number of details in his narraciones by selective amplification. Physical backgrounds and spatial relationships are always specified, thoughts, motives and emotions are described, and narrative developments carefully prepared for by the inclusion in the narration of all the action and objects necessary to explain them. Narrative developments are never simply announced or juxtaposed, but every part of the narrative is carefully linked to those around it, deriving from what has gone before, and preparing for what follows. All is logical, with none of Chaucer's haphazard ‘aventure, or sort, or cas’.¹⁵ Much of the flavour of Gower's abbreviated narrative art can be found in the ‘Tale of Diogenes and Aristippus’ (VII 2217), with its constant reference to thought, motive, time, action, place and space, and its remarkable weaving together of all these elements (philosophical, spatial and vegetable) into its ethical significance. What is important in Gower's detail, logic and spareness, as far as exemplification is concerned, is that they make for very full and very clear illustration and explanation. It is always clear in Gower's narraciones what is happening, and why, and this gives the ‘exposition’ they support the most comprehensible foundation possible. At the same time, the compression of the narratives ensures that the details of the narracio do not simply obscure the lesson which the ‘exposition’ expounds and which the moralitas seeks to draw from the narracio.

II. QUESTIONING GOWER'S BALANCE

Mention in the previous paragraph of the relationship between narracio, moralitas and ‘exposition’, and of that between ethical framework and exemplary story, brings us back to the problem of understanding these relationships. The difficulties for modern readers are two-fold. First, it is felt that ethical analysis and artistic narrative are different kinds of discourse, which produce their meanings in opposite ways, and which seek different ends and effects. Second, it is felt that ethics inevitably dominate in exempla, not merely at the expense, but to the detriment, of the stories and their meanings and effects. The first point is a general assumption about ethics and art, which arises whenever the two come into conjunction. The second is a statement of what is felt inevitably to occur when art and ethics are yoked together.

The term ‘art’ is not accidentally contrasted to ‘ethics’ in the last two sentences, for distinctions about the nature of these seemingly opposed ways of thinking and writing are at the heart of modern discussions of Gower's ‘ernest’ and ‘game’. It is from modern assumptions about ethical analysis and literary art that modern oppositions, and modern judgements about the likelihood of Gower succeeding in balancing ‘ernest’ and ‘game’, have grown. Derek Pearsall bears witness to these assumptions when he says:

It is in the stories that Gower's imagination receives its final and fullest release. It is an imagination in which moral discrimination continues to operate, but operates in an artistically integrated manner. The truth about human nature and human behaviour which Gower tells in the tales is not one that he could have isolated or spoken of in expository fashion, for it is one which depends on the initial response of imaginative sympathy to the human situation.¹⁶

The stories are contrasted to the parts of the Confessio where Gower writes and thinks ‘in expository fashion’: literary art and ethical analysis are separated, and in this Pearsall's remark is typical. Art is felt to have two qualities lacking in ethics: art establishes its significance through a ‘response of imaginative sympathy’, and art is able to do this because it expresses its significance through a response to particulars. Since the late eighteenth century, particularity has been especially valued as the most accurate, the most informative, and the most desirable source of truth.¹⁷ Particulars are present in Gower's stories, and are felt to be lacking in his ethical analysis, and this makes the latter less interesting and less valuable. It is not only a question of the presence or absence of particulars, however, but of the way in which the truth is conveyed to a reader. The ethical ‘expositions’ consist primarily of statements about ethical categories of vice and virtue. These categories are presented as authoritative statements of objective truth, general principles of morality by which the complexities of experience may be ordered. The categories appear not to derive from a study of the particulars of human experience, but seem to require the assumption that the particulars of that experience can be made to conform to the principles stated in the categories. The categories are information, to be accepted, understood, and acted upon, without reference to experience. None of all this process of statement allows for a ‘response of imaginative sympathy’. The art of the stories, by contrast, does not state its significance. Instead, a story presents a number of particular details of location, time, character, motion, action, and so forth. The meaning of the story is arrived at by the reader's observation of the particulars and the way in which they interact to form sequences of actions. The meaning is a compound of the meaning of the particulars and the meaning of the sequences, the two interacting to produce a significance which is apprehended by the reader using his (modern) imagination. The stories all concern people and human behaviour and experience. The imagination of the reader is thus stimulated to understand the events of the stories in terms of the reader's own experience, using the particulars of the narration to guide their response. Art is thus opposed to ethics in being suggestive and subjective, and more accurately truthful. Through its appeal to the (modern) imagination, art is also more pleasurable. Overall, then, particulars and the ‘response of imaginative art sympathy’ make the art of the stories very different from the less informative statements of ethical analysis. In trying to balance the two, it is felt, Gower was attempting to reconcile irreconcilably different kinds of discourse, and inevitably failed.

The second main argument advanced by modern readers against the possibility of Gower having succeeded in balancing art and ethics is, as I said, that ethics inevitably dominate in exempla. This argument is based on the correct identification of the importance of Gower's ethical purpose in the Confessio. The exempla do exist to help illustrate, explain and support Genius' assertions in the ethical framework of the poem, and the narraciones are indeed parts of the ethical analysis. Because Gower's main purpose in telling the narraciones is to add to his ‘exposition’, modern readers have felt that he invariably subordinated the stories to the needs of his ethics, and deprived the narraciones of their intrinsic (and particular) significances and effects. Art, in effect, seems to many modern readers to have been thrown overboard solely to add buoyancy to the ethics.

Two main explanations are given of the way in which ethics dominate, both of which are held to be observations of processes inevitable in exemplification. The first is that when a narracio is incorporated into the meaning of an ‘exposition’, the attention of the reader is diverted away from the particulars of the narracio. Instead of allowing the reader to engage with the meaning and emphasis of the particulars (the intrinsic meaning of the narracio), the moralitas insists that account is taken only of those aspects of the narracio which are typical or representative, and which exemplify the argument of the ‘exposition’. In Gower's exemplum of Acteon, for instance, the moralitas defines the significance of the narracio by requiring the reader to ignore the particular facts of Acteon's actions, and only to notice the fact that he looked and was punished for it. Only what is typical of the evils of sight is to be of importance. In an exemplum, it is argued, the expression of the particulars is limited to that required by the ethical analysis. This restriction means that the particulars of the narracio must be understood as representative, and not as having their own significance for the reader. Any possibility of the narracio expressing its intrinsic meaning, or evoking a ‘response of imaginative sympathy’ to itself, is dismissed. It is to this general feature of exempla that John Burrow refers here:

In a fiction which merely exemplifies an ethical concept (‘patience’, ‘gluttony’) or an accepted truth (‘Women are fickle’, ‘Radix malorum est cupiditas’), literature condemns itself to an ancillary role as the servant of the moral or political or religious beliefs of its age … In the literal mode of ‘exemplification’, the story may do no more than illustrate slavishly idées reçues.¹⁸

Exempla do not respect the importance and integrity of stories, and are thus inevitably destructive of what is most important for modern readers in literary works of art.

The second reason given for the dislike of exempla and for the assumption that they cannot allow the balance of art and ethics is an extension of the first, and is that many exempla are so presented that the expression of the particulars of the narracio is not only implicitly but explicitly restricted. In order to ensure that the meaning expressed by the narracio is congruent to that of the ‘exposition’, and useful in exemplifying it, two techniques were actively used to prevent the particulars of the narracio expressing their own significance. One involved choosing a narracio that was already fixed in its significance. In some restricted narraciones, there are insufficient particulars to describe and explain fully the action of the narrative, so that it is necessary for the reader to turn to the moralitas or ‘exposition’ for an explanation of the significance of the narracio. In others, the significance of the narracio is not something that the reader perceives through his understanding of the particulars of the narrative, but is contained in a statement within the narracio. The second technique involves using the moralitas to assert that only some parts of the narracio are of importance to the reader. This deprives the narracio of any possibility of independent expression, and is felt to mean that balance is impossible within exempla. John Burrow again illustrates the view:

[In the Confessio Amantis] each story is told to illustrate a particular sin or species of sin, or a corresponding virtue; and the reader's attention is thereby directed towards a specific aspect of the behaviour of the characters in the story; an aspect formally defined in terms of the scheme of the Seven Deadly Sins, ‘their dependencies, circumstances and species’. The drawback of this arrangement is that it tends to narrow our response to a chosen episode.¹⁹

It is not possible to read the stories whole, and for their own significance. Instead, exempla force the reader to accept the stated meanings of the ethical ‘exposition’, if necessary by the active restriction of the narracio. Modern assumptions hold out little hope that Gower could have done what he announced himself as doing.

To accept these views of ethical and exemplary processes is to accept that Gower failed, and that his poem is lacking in unity. On one side are ethics, on the other the art of the stories, and neither can be linked to the other, for if such a link is effected the art inevitably suffers. Above all, it is difficult to see from a modern viewpoint how Gower could have hoped to balance his interests in art and ethics, and how he could have expressed both simultaneously. Yet the fact remains that he felt this to be possible, and it is therefore necessary to find a means by which he could have arrived at this end. It is my argument that he found this means in two related medieval theories: Aristotelian ethics, and medieval exemplum theory. Aristotelian ethics insist that ethical knowledge is based on, and derived from, a study of the particulars of human behaviour and experience. Theories of exemplification stressed the importance of particulars in human understanding through the (medieval) imagination. In the view of Gower and his contemporaries, exempla laid great stress on the particulars of the narracio and their importance in an exemplum. In addition, the pleasure of narraciones was a most important aspect of exempla, for the Middle Ages. Finally, exemplum theory reveals that, far from being inevitable, restriction is something that must be imposed on narraciones, for the form naturally encourages their independence. Exempla do not work by fixed processes, but present possibilities that Gower could exploit. It is in the light of these theories that Confessio Amantis must be viewed, and that an assessment of Gower must be made. The results turn out to be far more flattering than many modern readers have suggested.

III. ARISTOTELIAN ETHICS AND THE MEDIEVAL THEORY OF EXEMPLA

Gower's ethical purpose in Confessio Amantis is central. His intention was to write a large-scale analysis of human behaviour and values, and of the ways in which these could be ordered for the greatest good. This intention need not imply, however, that his poetic art was a minor consideration for him, or even a secondary one. His stories were capable of providing clear and full illustrations of human experience, provided only that they could be incorporated into the ethical analysis without having their meanings impaired. An acknowledgement of Gower's dependence on an Aristotelian view of ethics is the first step towards an understanding of how Gower attempted to combine complex story-telling with explicit ethical analysis.

For Gower, Aristotle was ‘the Philosophre’ (VII 1645). Aristotle is credited with the authorship of all the views on kingship expressed in Book VII, views that are clearly central to the poem's ethical analysis. We do not have to look for proof that Gower had read Aristotle himself, however, for two of the works known to Gower, and used in the composition of Book VII, made explicit use of the philosopher's ideas. Brunetto Latini's Li Livres dou Trésor is in part made up of an annotated translation of the Nichomachean Ethics, while Giles of Rome's De Regimine Principum, a work of great popularity, contained a full discussion of Aristotle's ethical principles, again from the Nichomachean Ethics.²⁰ Gower does not quote Giles' discussion of principles, but the discussion is clearly a crucial indication of the way in which Gower thought.

In Giles, Gower would have found a complete definition of the process by which truth is arrived at in ethics. In defining his work as ethical, Giles explains that its method of proceeding is figural and concrete: ethical analysis must work through types.²¹ The rest of Giles' prologue is devoted to explaining this remark, and the need for figurality. He argues, closely following Aristotle, that ethics is not a demonstrative science, like mathematics, in which proof may be obtained by logical demonstration and by subtle and inward contemplation. The study of ethics, a persuasive discipline, does not have that kind of certainty, for ethics ‘is about human behaviour’ and thus ‘about particular events, which (as has been declared in the second Book of the Ethics) have great uncertainty because of their variability’.²² It is not possible to know directly in ethics, because of the complexity of the particulars of human behaviour, and their variability. In this approach to ethics, then, particulars are not incidental, but are an integral part of ethical study. As Giles goes on to show, they are still more important in ethics than this, for they are the foundation on which ethical knowledge is based. Ethical study is not pursued for the sake of knowledge, but so that we may act well. As Giles' translator, John Trevisa, put it, ‘we taketh moral work … nouȝt by cause of contemplation nother for to be konnyng bote for to be good’.²³ The end sought in ethics is not knowledge but works, not truth but the good. Ethical knowledge is not intellectual, but affective.²⁴ The affection, which governs people's actions, is not enlightened by pure speculative reasoning, but is moved by material evidence. It is the particulars of human behaviour and experience which supply this figural evidence. ‘The sothe is ischewed boystousliche and by liknes by soche preues and of dedes that falleth ofte tyme.’²⁵ Genius' ethical analysis seeks to move Amans, and the audience, to good deeds and to a practical knowledge of the good. To do this, Genius cannot use abstract ethical theory and theoretical categories, but must instruct by means of ‘euydens and figures and liknes’,²⁶ that is, by means of the particulars of human behaviour and experience as they are recorded in his narraciones. Particulars are not incidental to an Aristotelian ethical analysis, nor are they a separate interest. They are the heart of the analysis, making it comprehensible in concrete terms, and giving it its affective force. Gower's understanding of ethics is not that of his modern readers, and does give a role to particulars. With such an understanding, a balance of art and ethics becomes a possibility once more.

There are particulars in Confessio Amantis that are not in the narraciones, and Gower's Aristotelian approach to ethics clearly makes these of importance, as well as the particulars of the narraciones. Although predominantly abstract, Genius' ‘exposition’ is not uniformly so. Many vices are presented conceptually, and not in terms of the behaviour characteristically associated with them. For instance, Inobedience (I 1235-51) is first presented through a mixture of simile and personification that describes the mental attitudes of the inobedient, without specifying the behaviour that derives from these attitudes. In a number of cases, however, Genius chooses to describe a vice in action, as in the case of Stealth (V 6493-554). Sometimes the two kinds of description are combined, as when the definition of Falssemblant (II 1879-1956) is amplified at Amans' request by a description of the vice in action in the context of love. In these cases, a very full statement of the nature of the vice is provided. Finally, Amans himself contributes greatly to the particularity of the ‘exposition’ through his confession. He is careful throughout to distinguish vices of intention and vices of action, but his confession is usually in terms of his behaviour, often describing it in detail, as in his confession to Somnolence (IV 2746-916). The Aristotelian approach does not enable us to distinguish between these particulars, which are not necessarily ‘artful’, and those of the narraciones. The particulars in the ‘expositions’ are very clearly subordinate to the ethical analysis, and the question of their relationship to the ethics is neither so complex nor so important to our understanding of the poem as the problems raised by the narraciones. Giles' Aristotelian approach helps to clear the way for a view of art and ethics close to Gower's, and is of great value in suggesting that the two activities might be balanced. Actually to understand how the balance is achieved, however, it is necessary to look at the writings of medieval theorists of exemplification. These theorists provide no systematic body of analysis of exempla, but their practical concerns do lead them to explain many points which throw much light on Gower's understanding and use of the form.

The writings of the theorists reveal two main facts which govern exempla, and which directly affect the question of how the narracio serves the ethical purpose of the ‘exposition’. One, which takes us one step beyond Giles, is that the particulars of the narracio are crucial not simply to ethical understanding, but to the whole process by which the mind was seen to understand, through the imagination as this faculty was then defined. This argument means that an exemplum is only instructive when its particulars are understood as closely as possible, as closely as the moralisation will permit. In this way, the exemplum form places special emphasis on the narracio and on the precise significance of its particulars. The second revelation provided by the theorists is that although it may be desirable in many contexts to restrict the expression of the particulars of the narracio, the restriction is not inevitably inherent in the way exempla work. If a writer wished to allow his narraciones to supply all the instruction in his ‘exposition’, he could do so, by a careful choice of narracio, moralitas and ‘exposition’. Gower, seeking to have his audience enjoy his narraciones and to understand them to their fullest extent, could still use the exemplum form to combine these aims with his ethical purpose.

The first aspect is found in its most general form in the remarks of Stephen of Bourbon, who discusses the value of exempla in terms of the way in which the mind understands ideas. He starts by citing Christ's use of concrete examples, and compares these to the Incarnation, whose meaning is understood in a way analogous to that of exempla. He then quotes Dionysus on the way in which exempla affect the imagination.

Wise men body forth their speech by clothing it with similitudes and exempla. For corporeal speech moves more easily from the senses to the imagination, and from the imagination to the memory.²⁷

The precise meaning of this argument about imagination is made clear in the writing of Thomas Waleys.

It also appears that speaking by similitudes, if they are well chosen, is more pleasurable and is more firmly fixed in the memory than simple statement or reasoning, for while we are in this life the understanding of our intellect is naturally by conversion to images (phantasmata), which are the likenesses of things; which, if they are well chosen, please more.²⁸

The theory of imagination current when exempla were popular maintained that the mind did not directly understand concepts and ideas. As Alastair Minnis has explained,

Objects perceived by the five exterior senses were believed to cause separate sensations, these sensations met in the sensus communis, and the imagination, stimulated by these sensations, formed the mental pictures (imagines or phantasmata) necessary for thought. Images thus produced were handed over to the reason, which employed them in the formation of ideas. These ideas were in turn handed over to the memory, the store of ideas of past things.²⁹

Both the reason and the memory could compare and arrange images so as to arrive at an understanding of complex thoughts.

The great value of exempla, as this theory stresses, is that they provide particulars of behaviour and experience which make it possible for the mind to understand ethical truths. The emphasis of the theory is very much on the importance of particulars. Abstract thought is possible, but all thought begins with particulars. The advantage of exempla is that they supply, in their narraciones, records of particulars. These particulars greatly facilitate the creation of mental images, which are the basis of thought. That is why exempla are particularly useful for simpler audiences,

to whom things are to be set forth as though corporal and palpable and such as they have frequently known by experience.³⁰

The medieval theory of the imagination, as it is applied to exempla, grants particulars the greatest importance. The narraciones, through their effect on the imagination, become the main reason for using exempla, and also the source of the meaning of exempla.

When we are speaking in Latin, in a convent and to a congregation of the wise, then we can say many things, and do not need to descend to particulars: to laymen, however, it is necessary to demonstrate everything as though to the eye, and in a way perceptible to the senses.³¹

Since the particulars of the narracio are so important to thought, and to ethical analysis (as Giles has shown), it is peculiarly important for an audience to have as clear, as close and as complete an understanding of those particulars as possible, if the ethical instruction is to be understood as fully as possible. The importance of particulars in the understanding of ethics, especially in exempla, thus means that the first effect of an ethical structure is to throw great emphasis onto the particulars that support it. What the theorists of exempla first reveal, then, is that an exemplum works by stressing the importance to an understanding of the ‘exposition’ of a close reading of the narracio, and ensures that this close reading takes place by directing the audience's attention firmly to the particulars of the narracio.

As John Burrow argued, and as I indicated earlier, however, the exemplum form does not always permit the full understanding of all the particulars of a narracio. Restriction does play a part in the presentation of exempla, and if it is a regular occurrence it will inevitably negate the freedom of expression which the theory just described grants to narraciones. It is no good being directed to the narracio if its meaning is prescribed by a limiting moralitas. The answer to this problem lies in the second main revelation made by the theorists.

The theorists are all clerics, as are the great majority of users of exempla. Their ‘expositions’ are concerned with the presentation and assertion of orthodox Christian ethical and spiritual truths. Exempla are very useful in making these truths more comprehensible, but it is crucial for the clergy that the exempla do not distract audiences from those stated truths, or present behaviour or experiences that contradict the truths. Preachers and moralists are thus concerned always to use exempla whose significance is strictly consonant with that of their ‘exposition’, and whose interest and attractiveness are limited. This end is most easily achieved by presenting the exemplum in such a way that the expression of the narracio is restricted. Such restriction also has the effect of ensuring that the ethical purpose of the exemplum is dominant, since the narracio cannot assert any significance to rival the ethical one. What the theorists stress in their writings, however, is that restrictions of this kind do not occur automatically in exempla, but must be made to occur. It is possible to restrict the meaning of an exemplum by restricting the expression of its narracio, but this requires a careful choice and manipulation of the parts of the exemplum. Exempla do not work in fixed ways that are inevitably restrictive. Rather, the exemplary process is a subjective one, that requires careful control.

Out of many those exempla are to be chosen which are more necessary and contain useful evidence and are short; and if the narrative is long, the useless or less useful parts are to be removed, and only what bears on the matter to be narrated.³²

Two problems arise for the exemplum user who wishes to present a limited argument. One, as implied in the remark just quoted, is that the narraciones of exempla can, through their particulars, express meanings different from those of the ‘exposition’. The second, particularly worrying for the clergy, is that the narracio is capable of distracting the audience, because the narracio is more attractive or more interesting than the ‘exposition’. This is especially a danger if the narracio is derived from a literary source.

We ought to banish unfruitful fables and the curious songs of the poets from our sermons.³³

The chagrin of preachers is captured in Cesarius of Hesterbach's story of a preacher who shouted out to his sleeping congregation that he was about to tell an Arthurian tale as an exemplum. He goes on to exclaim:

See, my brothers, a great unhappiness. When I spoke of God you slept; but as soon as I insert levities, you are all wide awake, listening with pricked ears.³⁴

Exemplification is not the simple, restrictive, process that John Burrow once suggested it was. The narraciones and their particulars are given great prominence, and are crucial to the audience's understanding of the ‘exposition’. The narraciones are valuable partly because they are pleasurable, which makes them more useful. Finally, it is clear that the particulars of the narraciones are capable of producing their own meaning, independent of that of the ‘exposition’, and that this meaning can radically affect the meaning of the exemplum, and hence of the ‘exposition’.

The consequences of these facts are very great. Above all, they make it possible to argue that Gower could, by deliberate and careful choice, have presented his narraciones in such a way that attention was drawn to their particulars and the significance evolved by them, that his audience would be required to understand the particulars and their meaning as fully as possible, and that the ethical significance of Genius' ‘expositions’ would be directly affected by, and occasionally dependent on, the audience's close understanding of those particulars. The crucial point is that Gower, although treating of ethics and with the ethical purpose of his poem strongly in mind, was a poet, not a cleric. He wished to present an ethical analysis of mankind and the world, but he was also concerned in the Confessio to display and use his poetic skills. As I have suggested in my discussion of medieval theories of ethics and exempla, there was nothing to prevent him combining the two interests by using existing structures to link his art and his ethics. Poetic predecessors and contemporaries had done the same. Boccaccio and Guillaume de Machaut had both manipulated the exemplum form to raise wider human issues in their poetry, and Chaucer was to use the form to question the ways in which human experience might be assessed and understood.³⁵ Gower's aim was to strike a balance, with his stories retaining their significances and their intrinsic importance, while still contributing to the ethical analysis. What the use of exempla enabled him to do in theory was to enhance the importance of a close reading of his stories, and to direct his audience's attention to the stories and their particulars and significances. At the same time, the exemplum form, and his approach to ethics, made it possible for him to enhance his ethical analysis by including in it the information derived from this enhanced ‘response of imaginative sympathy to the human situation’, if he wished to. The question that remains to be investigated is whether Gower did avail himself of these opportunities in Confessio Amantis, and of the extent to which he succeeded in his aim.

IV. THE EXEMPLA OF CONFESSIO AMANTIS

Although he was more concerned with arguing that the stories are a more informative source of ethical knowledge than Genius' ‘expositions’, Derek Pearsall did hit, in his remark which I have already quoted [on pp.111-12 above], upon the most important aspect of Gower's stories. What makes them so informative is that they are concerned with human behaviour and experience, that is, with ethical matters. The particulars of the narratives are ethical particulars, and the significances of the stories are ethical significances. The particulars of Gower's narraciones are thus concerned with the same area of human existence and knowledge as his ‘expositions’, and are therefore able to contribute directly to the ‘expositions’. Elsewhere, Derek Pearsall has given some indication of this connection.

In this poem Gower found, as if by chance, his natural vocation as a polished and fluent verse narrator, and it is this story-teller's gift which is our chief delight in reading Gower, and his chief claim on our attention. Yet it is not his only claim, nor is it separable from his role as a moralist, for Gower's narratives … are staked out in a world of consistent social and moral values, a world of civilised feelings, which Gower can now define all the more effectively because he is writing out of imaginative sympathy and not out of admonitory purpose [as he was in his earlier works].³⁶

The ethical nature of Gower's stories does more than make them better able to support his ‘expositions’, however. Above all, it means that the stories can, if appropriately presented, contribute directly to Gower's ethical analysis, without being subordinated to the ‘expositions’ and without having their ethical significance explained or prescribed by a moralitas. The ethical concerns of the stories thus make it possible for Gower to benefit from the inclusion in his ‘expositions’ of much detailed investigation of human values and behaviour. At the same time, these concerns ensure that the independent value and importance of the stories are recognised, and that the independence of the stories' intrinsic meanings is preserved. It is by virtue of these latter qualities of the stories that Gower is enabled, as I shall argue, to strike an effective balance in his exempla between art and ethics.

Before moving on to this argument, however, it is necessary to look more closely at Gower's actual practice in using exempla. He did not always seek, nor did he always succeed in striking, a balance. Some of his narraciones were not suited to such a balance, and in some of the more explicitly didactic parts of his poem, like Book VII, he wished to retain the dominance of his ethical statements over the particulars of the narraciones. In these cases, restricted narraciones and restricting moralitates do appear. In other cases, however, he did wish to effect a balance, as is clear from the importance he accords to the intrinsic significances of his narraciones. In a number of these instances, however, the significance of the narracio is at variance with that of the ‘exposition’, sometimes accidentally, sometimes with important consequences for the meaning of the ‘exposition’. These examples add to the general balance of art and ethics within the whole Confessio, but there is no balance within the individual exemplum. Finally, and most characteristic, are the exempla in which narracio and ‘exposition’ complement each other, and in which the balance is achieved. It is not possible to approach this last category of exempla, however, without having first investigated the others.

Most frequently in Confessio Amantis, the restrictions on the expression of the significances of the particulars of the narraciones come about because of the moralitates. There are two kinds of moralitates in the Confessio: those in the Latin commentary, and those in the English text. The Latin ones would have been available only to actual, and educated, readers of a manuscript; the English ones are an integral part of any reception of the poem. The Latin commentary always stresses the ethical importance of the material of the poem, and does not take into account the effects of the particulars of the narraciones. Its moralitates are thus all restrictive to some extent. Because they are remote from the text, however, their effect is less marked than that of the English ones. What the Latin moralitates do is to emphasise the typical and representative nature of the events of the narraciones. Some of the moralitates are simple:

Here he advances an exemplum against those who are delicate.

(VI 976 margin)

Others provide a full explanation of the meaning of the particulars of the narracio:

Here the Confessor gives an exemplum against those who, when they cannot openly assuage their hatred by their ire, pursue their vengeance secretly by counterfeit dissimulation.

(III 973 margin)

The more complex the moralitas, clearly, the more it restricts the significance of the particulars whose meaning it prescribes. However, these moralitates are not closely linked to the English texts of the narraciones, and the restrictive effects they have are therefore not great. The Latin moralitates mainly serve as a constant reminder of the importance of the ethical purpose of the poem.

The English moralitates, by contrast, have a direct effect on the reader's understanding of the significance of the narraciones, and a number of these moralitates are restrictive. The most characteristic of these relate the narracio very directly to the ‘exposition’, using conjunctions to require the audience to understand the narracio solely in terms of the preceeding ‘exposition’:

And that a man mai riht wel frede,
These olde bokes who so rede.

(IV 3511-12)

Whereof a tale amonges alle …
I thenke forto tellen hiere,
That thou such moerdre miht withstonde.

(III 1880-3)

‘That’, ‘whereof’ and ‘such’ refer directly back to the ‘exposition’, and leave no possibility of the narracio asserting its own meaning within that ‘exposition’. The narracio can only be a subordinate part of the ethical analysis in exempla of this kind. If such a moralitas is combined with an ‘exposition’ in which a detailed and particular definition of the ethical category is provided, the effect on the narracio is very restrictive, as is demonstrated by the ‘Tale of Virgil's Mirror’ (V 2031). The exemplum is told ‘in ensample of alle tho’ (V 2027), a direct reference back to the comprehensive ‘exposition’. The lesson of the moralitas is that

Whan Covoitise hath lost the stiere
Of resonable governance,
Ther falleth ofte gret vengance.

(V 2226-8)

The narracio is thus attempting a very restricted exemplification of a vice that is already well-defined. Moreover, the moralitas directs attention only to the part of the story in which Covoitise is punished. Any particulars of behaviour associated with the vice itself are thus of an even more limited importance in the reader's understanding of the narracio. Some attempt is made in the narracio to define the vice more fully through the behaviour of Crassus, but this is incidental. What matters in the exemplum is that the narracio should provide simple corroboration of the analysis in the ‘exposition’. This fact is reflected in the great simplicity of the definition of the vice in the narracio itself, and the slight use made of emotional detail to expand this basic statement. Crassus

… was so coveitous,
That he was evere desirous
Of gold to gete the pilage.

(V 2069-71)

He ‘redily’ (V 2112) has the philosophers lodged near him when they first appear, and is ‘glad’ (V 2119) when told of the first dream. His eagerness is suggested in his visit to the site, and in his immediate agreement to the digging up of the mirror. The purpose of the exemplum is entirely ethical, however, and the reader's attention is firmly directed to its representative meaning, which is mostly evident in the part of the narracio describing the devastation of the city and the aural punishment of Crassus. Only a partial response is possible, with this narracio.

‘Virgil's Mirror’ is typical of a number of narraciones in the Confessio in having its significance restricted by the ethical machinery of the exempla. It is also typical in having a narracio that lacks sufficient particulars to evolve an independent significance. In particular, the narracio does not establish its own definition of the vice under discussion, but is reliant for this on the definition stated in the ‘exposition’. Such a lack of particulars and a reliance on stated definitions is characteristic of a number of narraciones in the Confessio. Such narraciones are restricted both internally and externally, by their own lack of particulars, and by their subjugation to the statements of the ethical analysis.

The most restricted narraciones are those in which the ethical meaning of the narracio is directly stated or argued in a speech by one of the characters. In these narraciones, the particulars of the narration are simply a redundant background to the speeches. The brief ‘Tale of Antigonus and Cinichus’ (VII 2115) describes a simple exchange of demands and responses. There is some evaluation of the exchanges, as when the first demand is described as ‘over myht’ (VII 2120), but the significance of the narracio is contained in Antigonus' two answers: one gift would be too great for his estate, the other small for his dignity. There is no elaboration through particulars of a situation in which an ethical significance might emerge. Rather, the significance is stated, in the baldest manner. Even when the narracio is more complex, the inclusion within the narracio of such a statement of significance inevitably restricts the meaning of the narracio to the stated one. As I mentioned earlier, the ‘Tale of Diogenes and Aristippus’ (VII 2217) contains a great number of inter-linked particulars that are not redundant in the expression of the significance of the narracio. Diogenes' frugality and self-sufficiency are stressed, and coupled to his devotion to philosophy. They stand in marked contrast to the pursuit of wealth by flattery which is the life of Aristippus. The significance of the narracio, however, is stated in Diogenes' rebuke to Aristippus. Every particular in the narracio supports this significance, and gives it liveliness and force, but the particulars have no part in the creation of the significance. The statement denies them expression and importance in the ‘exposition’.

Other factors also restrict some narraciones. One common lack in the narraciones is, as I have suggested, an inadequate definition in the narracio of the nature of the vice. This often takes the form of an assumed definition, on which the narracio is based but which requires fuller definition elsewhere if the vice is to be understood. The circularity of many of the definitions of vice in narraciones results from this basic inadequacy. For instance, in the ‘Tale of Capaneus’ (I 1977), the character is a personification, and entirely exemplary. The tale does not define the vice, but works from a stated definition.

He was of such Surquiderie,
That he thurgh his chevalerie
Upon himself so mochel triste,
That to the goddes him ne liste
In no querelle to beseche.

(I 1981-5)

God's punishment of Capaneus does prove that he was vicious, but the significance of the narracio is essentially circular: Capaneus, a presumptuous man, is vicious in his presumptuous acts because presumption is a vice. In the same way, in ‘Diogenes and Aristippus’ flattery is vicious because it is a vice.

The other common restriction in Confessio Amantis comes about when the particulars of the narracio are arranged in sequences whose significance is unclear until a statement within the narracio explains it. In these cases, the particulars are not redundant, but they do not contribute directly to the significance of the ‘exposition’ because they do not reveal, let alone assert, their own significance. The particulars are used to express the ethical significance of the narracio, but in being explained the significance is prescribed. Once again, the inclusion within the narracio of a statement of its meaning results in a restriction of that meaning. These narraciones usually depict a situation in which an ethical truth is enacted. In ‘The Emperor and his Masons’ (VII 2412), the coronation ceremony described includes a little pageant that brings home the truth that all men, however mighty, must die. By narrating the incident, Genius does no more than reiterate the message. Similar, if less comprehensible, sequences of events occur in some longer narraciones, like ‘The Trump of Death’ (I 2021) and ‘Rosiphelee’ (IV 1245). The moralitas of ‘Rosiphelee’ links the narracio firmly to the ideas of the ‘exposition’, but ‘The Trump of Death’ is introduced in a way that deliberately emphasises the importance of the significance of its particulars. In both cases, however, the meaning of the events of the narracio must be explained by one of the characters. In ‘The Trump of Death’, the king tells his brother what the events have meant, and explains how he created a situation in which he could state his viewpoint through its events. In ‘Rosiphelee’, the events of the narracio form a mysterious pageant, whose significance is explained and applied by the lady on the wretched horse. In both cases, the events of the narrative are designed to teach a lesson to one of the characters in the narracio. When the lesson is stated and the events explained, the significance of the narracio is fixed, and restricted.

In all the exempla described so far, the significance of the narraciones has been restricted, and the ethical purpose and meaning of the narracio has been dominant. The narraciones have concentrated on vice, and not on the vicious, so that the particulars of the narraciones have been shown to be representative even as they were presented. At the same time, the narraciones have relied on stated definitions of both vices and significances, and there has been no opportunity for the particulars of the narraciones to express their own meanings. An independent assertion of significance is crucial if the narracio is not to be ancillary or subordinate to the ethical analysis of the ‘exposition’. If Gower was to achieve a balance between art and ethics in his exempla, he had to construct exempla in such a way that there were sufficient particulars in the narracio to evolve an independent depiction of human behaviour whose significance was established by the particulars themselves, without any reference outside the narracio. If the particulars could create this significance, as the theorists suggested they often did, the importance attached to particulars by Aristotelian ethics and the importance accorded to their significance by its central role in the processes of human understanding would ensure that close attention was fixed on the particulars and their meaning. Close reading of the particulars would be required, and the significance that emerged from this would make an independent contribution to the significance of the ‘exposition’, either to expand and enhance the significance of the statements of the ‘exposition’, or to contradict them. As I indicated earlier, the ethical nature of Gower's narraciones helps ensure their independence within the exemplum. At the same time, there are clear indications that Gower was aware that narraciones could have significances independent of those prescribed by the ethical machinery. These indications come from his creation of exempla in which the narracio asserts a meaning different from that required by the moralitas or ‘exposition’.

Some of these narraciones are simply poorly chosen. Whatever its importance in the context of the whole Confessio, ‘Apollonius of Tyre’ (VIII 271) is an uninformative and confusing narracio about incest. In other tales, it is possible to ignore the discrepancy. The vicious man who is supposed to be condemned by ‘The Travellers and the Angel’ (II 291) is in fact shown punishing an even more vicious one. This does not affect the ethical meaning of the narracio, however, and the discrepancy can be ignored. In the exemplum of ‘Orestes’ (III 1885) the main crime, the killing of Agamemnon, is somewhat overshadowed by Orestes' brutal murder of his mother. In this case, the successful outcome of his trial vindicates him to some extent, and the stated ethical meaning of the moralitas remains unchallenged. Exempla in which the discrepancies can be ignored are in a minority, however. As he clearly showed in a number of instances, Gower was well aware of the problems and possibilities of narraciones with independent meanings that did not exactly reflect the ideas stated in the ‘exposition’.

The first indications of Gower's awareness of the importance and independent significance of some configurations of particulars come when he can be observed altering the emphases and adjusting the significances of the particulars of a narracio in order to preserve the meaning expressed by the moralitas. In compensating for the inappropriate significances to which the narracio might give rise, Gower clearly acknowledges that particulars often do have their own meanings, and that these must be controlled. Such efforts at redress can be seen in the ‘Tale of Canace and Machaire’ (III 143), where the terrible and destructive cruelty engendered by the melancholy and wrath of Eolus is used as a warning against those vices. The problem for Gower is that the objects of the wrath are themselves guilty of incest, a major vice to which much of Book VIII of the Confessio is devoted. His response is to emphasise the exonerating inevitability of love, and the power of love and nature, by describing the children's sin with much gentleness, and without condemnation.

Machaire mihte it noght asterte
That he with al his hole herte
His love upon his Soster caste …
Cupid bad hem ferst to kesse.

(III 163-5, 169)

They are ‘as who seith, enchaunted’ (III 178), and innocently blind. By contrast, Gower later in the narracio emphasises the senseless and destructive viciousness of the wrathful father, and the pathos of the daughter's death. There is nothing but sympathy for the children, and condemnation of the father, and in this way Gower succeeds in retaining the ethical meaning he wished to exemplify. Again, in ‘Albinus and Rosemund’ (I 2459), Gower has to deal with the problem of having too many sinners in the narracio. An exemplum against boasting, the stated ethical significance of the narracio is that Albinus is quite rightly brought to a nasty end for his boasting. Unfortunately for the moralitas, the narracio also shows the trial and execution of the conspirators who have murdered him. To redress the ethical balance, Gower has to broaden the moralitas to indicate that the boasting is the direct cause of all the deaths. ‘And al this made avant of Pride’ (I 2647): boasting not only destroys the boaster, but like many vices it has a widespread destructive effect.

Since Gower is aware of the need to control his narraciones, and since he is prepared to revise his moralitates to take account of the more complex significances expressed by the particulars of the narraciones, it is possible to argue that in the exempla where he does not compensate for the discrepancies, he is deliberately using the independent significance of the narracio. Aware of the potential of the particulars of his narraciones, Gower exploited the narraciones to alter the significance of the ‘exposition’, and to point out the need for more complex responses to ethical questions. Some of the clearest examples have already been noticed by John Burrow.³⁷ At the beginning of Book IV, the exempla of ‘Eneas and Dido’ (IV 77) and ‘Ulysses and Penelope’ (IV 147) are used as illustrations of sloth and ‘lachesse’. The significance of the narraciones is substantially different from that stated by the moralitates. Eneas is not shown to be in love. There is only a veiled reference to Dido's faith in ‘the wordes whiche he seide’ (IV 89). He is seen to make no commitment to return, and moreover his travel to Italy is of the greatest importance, as Genius makes clear.

Bot after that, as it be scholde,
Fro thenne he goth toward Ytaile.

(IV 92-3)

Eneas does have ‘his thoghtes feinte / Towardes love and full of Slowthe’ (IV 118-19), but this is explicitly a result of his high destiny. Only Dido is shown to be concerned with love. As John Burrow observed, ‘Perhaps—though Venus' Confessor does not say so—“lachesse in loves cas” is not a vice at all.’³⁸ The discrepancy is even clearer in the next exemplum. Penelope has been abandoned a long time, some of her suitors ‘maken gret manace’ (IV 173) and her husband should honour his ‘trowthe’ (IV 199). Yet he still delays his return. His fault seems incontrovertible. The narracio, however, introduces a number of exonerating emphases and statements, and some that are directly contradictory of the condemnatory moralitas. Thus the narracio indicates that he will return, but only ‘what time that he mai’ (IV 219). He thinks ‘of a day / A thousand yer’ (IV 220-1) until he can see her again, but does not stir until

… the time is so befalle
That Troie was destruid and brent.

(IV 224-5)

Only then does he make ‘non delaiement’ (IV 226). The narracio makes it quite clear that the siege is of far greater importance than ‘trowthe’ in love. In both these narraciones, the particulars supply sufficient information for a reader to form an understanding of the significance of the narracio as a whole. The emphasis laid on the significance of the narracio by the exemplum form ensures that the particulars are carefully read, and their significance understood. Once this has been done, it becomes clear that narraciones and moralitates are in opposition. In this case there is no adjustment of the statement made by the moralitates, but the fact that Gower knew that particulars could assert their own significances justifies us in arguing that those significances are here being used to challenge the statements of the moralitas and to suggest the inadequacy of such statements for the assessment of the complexity of the particulars of human experience.

When the independent significance of the particulars of the narracio does force an alteration in Genius' moralitas, as we have already seen in ‘Albinus and Rosemund’, the independence of the narracio within the exemplum is made absolutely explicit. If the particulars of the narracio force an acknowledgement of their own meaning onto the moralitas, they force it into the ‘exposition’. Genius' ethical statements in the ‘exposition’ lose their dominance, and the importance of a reader understanding the particulars if he is to understand the ethical analysis is emphasised. This is the situation that arises in the exemplum of ‘Vulcan and Venus’ (V 635). The exemplum is part of an ‘exposition’ in which jealousy is defined in terms of the husband's unrelenting and entirely unmotivated suspicion of his wife. In the narracio, however, much emphasis is laid on the ‘lothly and malgracious’ appearance of Vulcan (V 647), the strangeness of his association with the beautiful Venus, and the readiness with which Venus accepts the advances of Mars. Vulcan is entirely right in his mistrust of his wife, whom he catches in bed and in flagrante delicto. The result of all these particulars is an extremely complex situation, in which the love of Mars and Venus, the sympathy she receives in the narration, and the rebuking of Vulcan by the other gods must be reconciled with the justice of Vulcan's claims to sympathy and compensation, and his completely justified suspicion and sense of grievance. The solution is provided by a very detailed moralitas, that alters the general definition in the ‘exposition’ and that very carefully interprets the particulars of the narracio. Vulcan, it is shown, has made the situation worse by choosing, through his jealousy, the worst of various evils. ‘His schame was the more’ (V 703) in betraying his wife and exposing himself as a cuckold and a jealous husband.

… where a man schal nedes lese,
The leste harm is forto chese.
Bot Jelousie of his untrist
Makth that full many an harm arist
Which elles scholde noght arise.

(V 715-19)

The particulars create their own significances and emphases, and the exemplum form forces the moralitas to take account of them, and to alter the definition of the vice in the ‘exposition’. Gower is here very close to achieving the conditions of balance. All that prevents it here is the fact that the moralitas initially tries to impose a restricted meaning. When the particulars are understood, the result is tension, and a balance is not fully possible. Already the ethical analysis is being expanded in response to the particulars of human behaviour and experience in the narracio, but the tension leaves some residual doubt about the truth and value of the ethical analysis, a doubt that would not be present in a balanced exemplum.

One final example will serve to indicate that Gower's occasional use of the independent significance of the narracio is not an accidental result of inappropriate narraciones. His acknowledgement, in altered moralitates, of the importance of particulars, is one proof. His use of a narracio independent from the moralitas at a crucial moment in the poem is another, striking, confirmation. The incident occurs at one of the most important junctures of the Confessio, when Amans argues with the Confessor, and comes very close to besting him. The argument is crucial to the whole question of Genius' authority in the poem, and to the authority of his ethical analysis. The disagreement arises when Genius commends crusading as a means of attracting favourable attention from one's lady. Amans emphatically denies this, pointing out that any killing is contrary to Christ's law and that distance from one's lady can never be a good thing. He even uses an exemplum of his own, to show how Achilles himself acknowledged a time for love and a time for war. Genius finally adduces an exemplum, ‘a tale thou schalt understonde’ (IV 1812), which will prove to Amans that love is to be forsaken for war, through which greater glory and success in love may be won. Unfortunately, his exemplum of ‘Nauplus and Ulysses’ (IV 1815) does nothing of the kind, for the narracio has a quite different significance. When Nauplus and other Greeks come to persuade Ulysses to join them at Troy, his reaction against war and in favour of love is immediate:

Ulysses anon upon Penolope
His wif, whom that he loveth hote,
Thenkende, wolde hem not behote.

(IV 1822-4)

This reaction is emphasised by the fact that he straightaway

… schop thanne a wonder wyle,
How that he scholde hem best beguile,
So that he mihte duelle stille
At home and welde his love at wille.

(IV 1825-8)

Ulysses does eventually go, which saves some of Genius' argument, but not because he is persuaded of the glory or necessity of war. He must be tricked into acquiescing, and remains to the end with ‘his herte fyred / Upon his wif’ (IV 1882-3). He is only ‘halvynge aschamed’ (IV 1885) of his attempts at evasion, and he does not set off with any enthusiasm.

… be him lief or loth,
To Troie forth with hem he goth.

(IV 1889-90)

The events of the tale finally bear out Genius' argument, but the significance of the particulars does not. At a vital moment in Genius' ethical analysis, the independent meaning of the narracio, emphasised by the exemplum, suggests a different version of the truth. Genius is not an infallible guide, clearly, and Gower uses the properties of his exempla to bring home this fact.

Gower was aware, then, of the independent significance that could reside in his narraciones by virtue of their particulars. He was prepared to alter moralitates to take account of the significances of the particulars, and to exploit the fact that the exemplum form concentrates attention on the particulars. He even shows Amans responding to the significance of the particulars. Sometimes this response is comic, as in his famous reply to the exemplum of ‘Phebus and Daphne’ (III 1685):

Bot while I se mi ladi is
No tre, but halt hire oghne forme,
There mai me noman so enforme, …
That I unto mi lyves ende
Ne wol hire serven everemo.

(III 1730-5)

Sometimes, however, Amans gives the clearest proof of the importance of particulars to any understanding of Genius' ethical analysis. In enjoining him to chastity, Genius uses the exemplum of ‘Valentinian’ (V 6395) to support his argument. The narracio does not elaborate its meaning through particulars, but relies on a stated definition which depends for its meaning on the ‘exposition’:

He was a virgine, as he seide;
On that bataille his pris he leide.

(V 6415-16)

Without particulars, however, Amans is loath to accept the argument, and argues that if all were chaste the race would die out. In any case, God bade mankind to ‘multeplie’ (V 6423). The inadequacy of arguing without particulars is reflected in the only answer Genius can find: ‘Mi Sone, take it as I seie’ (V 6428).

In every respect, Gower's exempla and narraciones were capable of being used to effect a balance between art and ethics. His narraciones are ethical and detailed, and their particulars often evolve significances of their own, independently of the ‘expositions’. As Derek Pearsall has taught us, the narraciones are also told with the greatest artistry and human insight; Gower's narrative art is considerable.³⁹ Gower was clearly aware of the independent importance and significance of the particulars of many of his narraciones, and acknowledged these aspects in a number of his exempla. At the same time, his approach to ethical analysis made an understanding of particulars a preliminary to ethical understanding. The exemplum form, through which he chose to present many of his particulars, made particulars more than preliminary: particulars made understanding possible, and exempla thus direct much attention to, and lay great emphasis on, the independent significance of the particulars of the narraciones. By using the form, Gower was able to bring at once the particulars of his stories and their independent significances to the notice of his readers, to require of those readers the closest possible reading and the fullest understanding of the particulars and their meanings, and to make that full understanding crucial to an understanding of the ethical analysis in the poem. It was because of, not despite, the exemplum form that Gower was able to strike a balance between art and ethics in his exempla. It is to these balanced exempla that I now turn.

The one aspect of Gower's exempla that would seem from my discussion so far to stand in the way of a balance of art and ethics is Genius' use of moralitates. In almost every case described so far, these have attempted to restrict the significance of the narracio by establishing close links between the significances of narracio and ‘exposition’. The links either made the narracio subordinate to the meaning of the ‘exposition’ or dependent on it, and the moralitates were either successful in their restriction or they were contradicted. There has been no suggestion so far that the moralitates could simply mediate between two equally important kinds of ethical significance, or that they could be used to indicate the importance in the ethical analysis of an understanding of the particulars of the narraciones. Such moralitates do exist, however, and their presence in the poem is proof that Gower was determined to use the independent meanings of his narraciones not only to contradict but also to expand and enhance his ethical analysis.

Many of Genius' moralitates draw Amans' attention to the particulars of the narracio and to their intrinsic meanings. Some simply commend the narracio to his notice:

… wherof I rede
A tale, and tak therof good hiede,
Of that befell be olde tyde.

(V 137-9)

For who these olde bokes rede
Of suche ensamples as were ar,
Him oghte be the more war
Of alle tho that feigne chiere,
Wherof thou schalt a tale hiere.

(II 2140-4)

More importantly, others introduce the narracio as an independent source of instruction, separate from and supplementary to the ‘exposition’:

And forto knowe how it so is,
A tale lich to this matiere
I thenke telle, if thou wolt hiere,
To schewe proprely the vice
Of this Envie and the malice.

(II 286-90)

Mi Sone, that thou miht enforme
Thi pacience upon the forme
Of olde essamples, as thei felle,
Nou understond what I schal telle.

(III 1753-6)

Wherof, if I schal more seie
Upon the nature of the vice, …
A tale, which was whilom soth,
Of fooles that so drunken were,
I schal reherce unto thin Ere.

(VI 530-1, 534-6)

These moralitates all acknowledge the independent significance of the narraciones, and the importance of this to an understanding of the ‘exposition’. They are a clear sign that Gower was not only concerned with establishing the independence of his narraciones, but also sought to create an ethical framework within which this independence could be put to use. Not all his balanced exempla have, or require, this kind of moralitas, but the examples just given do indicate Gower's intention of finding every possible means of manipulating his exempla so that the ethical analysis could benefit to the greatest extent from the significances of his narraciones. In introducing some of his narraciones in this way, Gower showed that he did not just hope for a balance, but that he actively worked to make one possible.

A balance between art and ethics is achieved in an exemplum when narracio and ‘exposition’ are used collaboratively to discuss the same ethical question or category, without the independent significance of the particulars of the narracio being restricted in any way by the ‘exposition’ or the moralitas, and without the ‘exposition’ being challenged by the narracio. In a balanced exemplum in Confessio Amantis, the particulars of the narracio independently establish the significance of a sequence of events concerned with human behaviour. This significance is used either to expand the discussion in the ‘exposition’, or to supply discussion and definition if it is lacking. What is crucial is that the ethical purpose of the exemplum should not be allowed to impair the independence or particularity of the significance of the narracio, and that the narracio should contribute the fullest meaning of all its particulars, including their ‘artistic’ ‘response of imaginative sympathy to the human situation’, to the ‘exposition’. When the art and independent significance of the narracio are given recognition, and the full significance of the narracio is used to make an independent contribution to the ethical analysis, Gower's exempla do indeed ‘stant betwene ernest and game’. It is Gower's use of the exemplum form that makes this possible. The narraciones are part of the ethical analysis, but they are independent. The importance of their contribution to the meaning of the ‘exposition’ requires the reader to arrive at the fullest possible understanding of the particulars of the narraciones. The pleasure and artistry of the narraciones are crucial to the investigation of ethics, and the reader's awareness of, and attention to, these qualities is enhanced and increased by the demands of the exemplum. At the same time, the interest, informativeness and accuracy of the ‘exposition’ are greatly increased by the addition of the insights into human behaviour and experience supplied by the particulars of the narraciones. By achieving a balance of art and ethics in his exempla, Gower was able to pursue his investigation of mankind's behaviour, to display his narrative skills, and by linking the two together to enhance both.

‘Acis and Galatea’ (II 97) is an exemplum in which this balance is struck. The narracio itself contains a detailed portrayal of a situation, in which the thoughts, actions and emotions of the three participants are constantly specified. At each stage in the development of the sequence of events, it is not only clear what is happening, but why, and how each character is reacting. The reader is thus given all the information necessary to understand the events, and to pass judgement on them. The particulars create a meaning, and it may be understood without reference outside the narracio. The particulars also achieve a very clear depiction of a complex series of emotional states and of the effects these have on others, and thus provide a considerable insight into one aspect of human behaviour. Much of the effectiveness of the narration comes from its concentration on Poliphemus, and its use of his point of view to focus the significance of the events. Having failed to win the love of Galatea, Poliphemus spies on her, eagerly and obsessively watching ‘be alle weies’ (II 111) to see what happens. The extent of his mania is suggested in the long duration of his watch: only ‘ate laste’ (II 112) does he discover what he fears, that another is not only favoured, but successful. The painfulness of Poliphemus' exclusion and failure, and the contrast of his sorrow to Acis' joy, is emphasised by rhyme:

… ate laste he knew and herde
How that an other hadde leve
To love there as he mot leve.

(II 112-14)

Despair follows, implicit in his inability to think of any course of action except further spying. This time, however, his intention is to seek revenge. Interestingly, what he hates is not Galatea herself, but her capacity to love: he hopes

That he hire love myhte grieve,
Which he himself mai noght achieve.

(II 119-20)

His reaction is shown to be far more complex than simple despisal of the woman or hatred of his rival. The portrayal of the lovers that follows emphasises the contrast between their virtue and Poliphemus' vice. Acis is ‘trewe’ (II 124), Galatea ‘above alle othre was unmete / Of beaute’ (II 122-3), and each is lovingly devoted only to the other. ‘Hierof was Poliphemus wo’ (II 134). He now waits and watches with undiminished determination (‘upon every side’ (II 136)), but no longer just for Galatea. By a subtle development, Poliphemus' envious hatred is seen progressing, and now encompasses both lovers. When he does find them, the scene is carefully ‘pointed’ to prepare for the climax of the story. Poliphemus invades the lovers' ‘prive’ (II 141) meeting place, where they stand innocently to ‘have here wordes goode’ (II 142); again, his actions are condemned by the contrast. The high bank and the sea necessary to future action are specified, as is Poliphemus' position above them. There follows a remarkable scene in which Poliphemus' emotions are expressed, realised and judged through a description of his actions. His heart burns, and he flees like some huge flaming arrow, burning like Etna he runs around. The judgements are made possible by reference to elements of a series of key ideas in the Confessio: he is like a beast without reason, and mad. His thoughts are described as ‘pure Envie’ (II 150). Again rhyme stresses the basic situation; he is

Fulfild of sorghe and gret desese,
That he syh Acis wel at ese.

(II 165-6)

Finally, with a last subtle emotional shift, Poliphemus turns against Acis, thinking to gain some relief by thwarting his rival. There is a slight suggestion that he pursues Galatea, but his killing of Acis is essentially irrational, a confused effort to deny another the happiness he may not have and to end his torment by removing its source. The metamorphosis of Acis and the displeasure of the gods add final confirmation to the judgement. Poliphemus' viciousness is complete.

The particulars of the narracio achieve a number of important objectives. They provide a subtle and complex portrait of a series of emotions. Starting from a hatred of rejection, they show the development of the desire to prevent the happiness of others. The particulars thus provide a full explanation of the nature of the emotions, what gives rise to them, and what effect they have. When the complex of feelings is identified as Envie, the reader's understanding of the vice is considerable. The particulars also enable the reader to assess the vice. Its destructiveness, its irrationality, its selfish obsession, all condemn it, even before the gods do. The particulars of the narracio thus enable the reader to understand a complex human situation, to be aware of its subtleties, and to judge it. As a statement about one aspect of human behaviour, the narracio lacks nothing. It is entirely ready to sustain its half of the balance.

What makes the balance work is the fact that the moralitas acknowledges the independent significance of the narracio. The narracio is not in this case the sole source of information on the vice. The ‘exposition’ is in fact quite explicit, and gives, through the mini-exemplum of the dog in the manger, a clear definition of the nature of the vice, and some description of the action it entails. The narracio thus makes its points within a framework whose outline is already obvious. This fact is not allowed to restrict the narracio, however, for the independent meaning of its particulars is fully recognised. The moralitas is explicit: Genius knows, and Amans will hear, a story ‘to this pourpos’ (II 96). The narracio is not simply an instance of the vice, it is about it. The narracio is linked to the ‘exposition’, by the reference back of ‘wherof’ (II 95) and, to some extent, of ‘this’ (II 96). ‘This pourpos’, however, refers both to the specific preceding argument, and to general discussion of the vice. The narracio is such a discussion, and the moralitas emphasises its independent value. The second part of the moralitas makes a similar acknowledgement of the independent significance of the narracio. ‘And thus … thou myht understonde’ (II 200-1) directs attention back to the narracio as a source of comprehension, and helps to ensure that the narracio is granted an independent function in the ‘exposition’. Coupled to the natural tendency of exempla to emphasise the importance of the narracio, these moralitates grant the narracio its independence, and allow it to make its own contribution to the discussion of the vice. To understand Gower's discussion of Envy, it is necessary to have as full an understanding of the narracio as possible.

The effects of the independence and the balance are considerable, and clearly signalled. Genius began the ‘exposition’ by defining Envy as an internal sorrow for the success or virtue of others. Amans confesses in particular terms to burning in his ‘hertes thought’ (II 23) when another is successful in love. This is only the case when the success is with his lady; anybody else may be loved with Amans' blessing. Genius then reverts to the more general definition, of the envious as those who wish to prevent others from loving at all (II 88-91). The first importance of the significance of the narracio is thus that it discusses a case analogous to Amans', a discussion which would otherwise be lacking. The narracio also makes explicit the destructive viciousness of Envy. The ‘exposition’ had concentrated on the turmoil in the thoughts of the envious, and only briefly touched on the effect this turmoil might have on others. The narracio thus increases the reader's sense of the horror and evil consequences of these emotions. Finally, the narracio supplies a detailed consideration of the emotions associated with the vice, and in showing them makes possible a much fuller and more sensitive understanding of the nature of the vice. Amongst other things, its irrationality, and the fact that its selfish destructiveness leads to no benefit to the envious man, are made clear. The ‘exposition’ is concerned with the existence of Envy, and its effect on those who suffer from it. By using the full independent significance of the narracio, Gower is able to enlarge the description and make it more informative, and also to provide a warning against the vice. It is the independence and particularity of the narracio that make these effects possible. Amans avers that

… how so evere that it stod
With Poliphemes love as tho,
It schal noght stonde with me so.

(II 212-14)

The emotions, and some of the actions, of the narracio are representative: ‘Ther ben of suche mo than twelve’ (II 97). The particulars as a whole are not, however. By acknowledging this fact, and making use of it, Gower was able to create his skilful narrative portrayal of human behaviour and to incorporate its full significance into his ethical analysis. The reader is encouraged to give the particulars his fullest attention, and to appreciate all that the art of the narrative can convey. At the same time, the ethical analysis and discussion is greatly enhanced by the addition of the independent significance of the narracio. In the exemplum of ‘Acis and Galatea’, Gower succeeded superbly in balancing art and ethics. In doing so, he demonstrated to the full the great gains, in pleasure, comprehension, and coherence, that result from such success.

In other exempla, Gower's ability to use the independent significance of his narraciones achieves still more. ‘Pyramus and Thisbe’ (III 1331) is an exemplum that follows a somewhat confused ‘exposition’, in which no very clear statement about the vice has been made. In this case, the particulars of the narracio supply almost all the information about the vice which the reader is given. The vice in this case is a complex of interlinked impulses. Contek (impetuosity) is characterised by Foolhaste, and leads directly to Homicide. They all arise when the reason is overthrown. Amans turns the discussion to love, and associates the vices with despair. He also suggests that the vices arise from an internal conflict of will, reason and love. What the ‘exposition’ lacks is any sense of the particular effects associated with the vices, of what the vices mean in terms of behaviour and emotional response. All this is left to the narracio. The moralitas indicates that the narracio does supply the information. The narracio is ‘behovely forto wite’ (III 133): to understand the vice, it is helpful to understand what the narracio has to say about it, independently from the ‘exposition’. At the end of the narracio, in the second part of the moralitas, the point is made again. Having heard and understood the independent meaning of the narracio, Amans should beware of indulging in the behaviour shown by Pyramus to be so disastrously destructive.

The narracio itself, like that of ‘Acis and Galatea’, provides a detailed portrayal of a human situation, evolved through a wealth of particulars. The circumstances of the central events are carefully described, with the reason for Pyramus and Thisbe falling in love specified, the urgency of their love suggested, and the reason for, and details of, their plan to meet outside the city given. Thisbe's absence from the meeting place, and the reason for the presence of her bloody wimple, are explained in great emotional and circumstantial detail. Pyramus' behaviour is then used to define the emotions and behaviour that constitute the vice. His horror is more than mankind can bear, and strikes straight to his emotions. His hand-wringing captures the inarticulate anguish that defeats his thought. Irrationally (a fact emphasised by the audience's knowledge that Thisbe is safe), he acts hastily and impulsively. No reason is involved in his action, as he ‘sodeinly’ (III 1428) takes out his sword. The false logic of love leads him to what he sees as a justifiable act, and he dies. The pathos of Thisbe's distress, and the destructive wastefulness of both their deaths, is finally used to confirm the significance of the narracio. Through its particulars, the narracio makes the audience aware of the irrationality of the vice, of the circumstances in which it arises and the emotions and failure of reason that characterise, and give rise to, it. As in ‘Acis and Galatea’, we are shown that vice is destructive of both self and others. By establishing the importance of this independent depiction of human behaviour to our understanding of the general ethical analysis of the vice, Gower was able to incorporate all these particular lessons into his ethical analysis. He also made a close and sensitive reading of the narracio essential, for in this case it is the only source of information. The richness, detail and art of his narrative are brought, in this and other balanced exempla, into the forefront of attention, to enrich not only the ethical analysis, but to enhance the whole poem.

V. CONCLUSION

Gower's was a sophisticated audience, capable of responding to his narrative art, and to the subtlety of his investigation of human behaviour. They did not need to rely on simple statements of ethical truth, but could be trusted to learn from ‘the truth about human nature and human behaviour … which depends on the initial response of imaginative sympathy to the human situation’, that is, from the particulars of the stories. An Aristotelian approach to ethics stressed the importance of such particulars. More important, the medieval theory of imagination, as applied to exempla, meant that the fullest understanding of the significance of the particulars was essential to the fullest understanding of the ethics. Exempla also allowed and encouraged Gower to establish the significance of the particulars of his stories as crucial and independent contributors to his investigation of ethical knowledge. By making his stories ethical themselves, and by acknowledging the independent importance within the ethical framework of the significance of their particulars, he succeeded in ensuring that his stories would be fully and sensitively appreciated. At the same time, he ensured that this full appreciation would become part of the full appreciation of his ethical analysis. The art of the stories is given full recognition in the poem, and the ethical analysis is made more precise, comprehensible, discerning and informative. There is no hierarchy of types of significance. Gower achieved a balance in his exempla that exalted his art and enhanced his ethics, and gave both a central role in his attempt to understand and define the human situation. An appreciation of this need not be restricted to his original audience. By recovering the values which he held, and the understanding he had of the techniques he used, Gower's achievement is available to all of us. ‘Ernest’ and ‘game’ are not distinct or in opposition in Confessio Amantis, but exist in a fruitful balance between pleasurable instruction and instructive pleasure. It is a balance whose creation required considerable skill and judgement, and Gower maintained it, with the greatest artistry.

Notes

1. For instance, G. C. Macaulay averred that ‘After endeavouring to “give an account of his stewardship” in various ways as a moralist, Gower at length found his true vocation, and this time happily in his native tongue, as a teller of stories. The rest is all machinery, sometimes poetical and interesting, sometimes tiresome and clumsy, but the stories are the main thing’. Gower: English Works I p. x. J. Fisher supports a very different view of the emphasis of the poem, but he too often refers to the parts of the poem as though they were quite distinct and irreconcilable: ‘The moralistic comment which Professor Lewis took as peripheral is central, the allegory of love which he took as central is peripheral’. John Gower, Moral Philosopher and Friend of Chaucer (London 1965) p. 191. Neither of these approaches, which are broadly representative of the critical division over Gower, suggests any possibility of Gower having achieved a unity in his poem.

2. St Gregory was cited as an authority for the use of exempla throughout the thirteenth and fourteenth centuries. His authority derived in large measure from remarks like that in Book I of the Dialogues, ‘And there are many who are drawn to love of the Heavenly Kingdom more by exempla than through direct preaching’. Dialogues I Prol. 9, ed. Adalbert de Vogue (Paris 1979) II 16.

3. For further discussion, see R. H. Rouse and M. A. Rouse Preachers, Sermons and Florilegia (Toronto 1979) p. 72. The fact that, at this time, exempla were recognised as being of value, but were not widely used, is reflected in Alan of Lille's failure to use any exempla in his model sermons, despite his strong recommendation that they should be employed. Migne PL CCX 114.

4. The new preaching ministry arose in response to new demands from congregations, which were being met by unauthorised or heretical preachers. The Church's official response was codified in the Fourth Lateran Council in 1215. This was translated, in the Paris schools, into new emphases in teaching, which affected the techniques of lecturing and the commentaries produced. Above all, the teaching of preaching was stressed, and the principle that good students should be encouraged to preach. One of the first results was a greatly increased use of exempla by academics in sermons and commentaries. Another was the development, during the thirteenth century, of a form of sermon based on a fixed structure of theme, protheme, and so on, as opposed to the older type based on an exegetical technique, the distinctio. The new form was an aid to both preachers and audiences, and helped to codify the place of exempla in sermons. Of importance in all these developments were the new Orders: the Dominicans, founded specifically as a preaching Order, and the Franciscans, who were quickly drawn into the field. For analysis see Rouse and Rouse op. cit. pp. 17, 43-90; P. B. Roberts Stephanus de Lingua-Tonante. Studies in the Sermons of Stephen Langton (Toronto 1968) pp. 39-74; B. Smalley The Study of the Bible in the Middle Ages 2nd ed. (Oxford 1952) pp. 196-263, and ‘Exempla in the Commentaries of Stephen Langton’ Bulletin of the John Rylands Library XVII (1933) 121-9; Artes Praedicandi ed. Th.-M. Charland (Paris 1936) passim.

5. Quoted in J. Th. Welter L'Exemplum dans la littérature religieuse et didactique du moyen âge (Paris 1927) p. 72 n.

6. See Rouse and Rouse op. cit. pp. 3-42; T. F. Crane The Exempla of Jacques de Vitry (London 1890) pp. xlvi-vii, xlix-xi, lxx-cxvi; Welter op. cit. pp. 109-376. Alphabeticisation was a popular organising method: see H. Pfander ‘The Mediaeval Friars and some Alphabetical Reference Books for Sermons’ MÆ III (1933) pp. 19-29.

7. Welter op. cit. p. 1. Welter's is still the best general survey of exempla. There are useful sketches of the history of the form in Roberts op. cit. pp. 39-104 and Crane op. cit. pp. xvii-cxvi. Rouse and Rouse op. cit. pp. 1-90 provide an invaluable introduction to the ecclesiastical and academic background. J. A. Herbert A Catalogue of Romances in the Department of Manuscripts in the British Museum III (London 1910) provides a wealth of details of manuscripts, while R. Schenda ‘Stand und Aufgaben der Exemplaforschung’ Fabula X (1969) 69-85 contains a useful survey of scholarship and a good bibliography. By far the most important studies of the ways in which exempla actually work are K. Stierle ‘L'Histoire comme exemple, L'Exemple comme histoire’ Poetique X (1972) 176-198, and J. A. Burrow Medieval Writers and their Work (Oxford 1982) pp. 86-118, esp. pp. 107-118. I have not been able to see J. Le Goff and J. C. Schmidt Les ‘Exempla’, to be published by the Catholic University of Louvain in the series Typologies de sources du moyen âge occidental.

8. See, for instance, the attempt to distinguish one category of material by L. Thorndyke ‘The Properties of Things of Nature Adapted to Sermons’ Medievalia et Humanistica XII (1958) 78-83; also Roberts op. cit. pp. 79-94.

9. Stephen of Bourbon speaks without distinction of ‘similitudes, parables, miracles and exempla’: Anecdotes historiques ed. Lecoy de la Marche (Paris 1877) p. 4. James of Vitry equates exempla and ‘similitudes’, Welter op. cit. p. 120 n., while John of Gobi differentiates between them, Crane op. cit. p. xxi n. Arnold of Liège distinguishes ‘narratives’ (narraciones) from exempla, Welter op. cit. p. 76; Ralph Higden speaks of ‘figures’, ‘examples of canons and saints’ and ‘similitudes of visible things’: Ralph Higden: Ars Componendi Sermones ed. M. M. Jennings (Ph.D. thesis, Bryn Mawr College 1970) p. 171.

10. I borrow the terms narracio and moralitas from manuscript collections of exempla, although the words are never used in precisely the sense I give them. They are used in preference to terms like ‘tenor’ and ‘vehicle’ which are less neutral in significance.

11. P. Boitani English Medieval Narrative in the Thirteenth and Fourteenth Centuries trans. J. K. Hall (Cambridge 1982) p. 8.

12. For discussion of Gower's narrative sources, see D. Speed Gower's Narrative Technique as Revealed by his Adaptation of Source Material in the Confessio Amantis (Ph.D. thesis, University of London 1970).

13. R. W. Frank Chaucer and the Legend of Good Women (Cambridge, Mass. 1972) esp. pp. 6-10. See also W. Francis ‘Chaucer Shortens a Tale’ PMLA LXVIII (1953) 1126-1141.

14. See the examples given by Speed op. cit. p. 94.

15. The Canterbury Tales, General Prologue I 844, ed. Robinson Chaucer: Works p. 25.

16. D. A. Pearsall ‘Gower's Narrative Art’ PMLA LXXXI (1966) 477-8.

17. An early comment illustrating this shift in taste is William Blake's note in the margin of a copy of Reynolds' Discourses, ‘To Generalise is to be an Idiot, To Particularise is the Alone Distinction of Merit’. William Blake: Complete Writings ed. G. Keynes (Oxford 1966) p. 451.

18. J. A. Burrow Ricardian Poetry (London 1971) p. 83. Modern difficulties with exemplification manifest themselves, for instance, in the fact that we are loth to accept Gower's simple moralisation of Troilus and Criseyde (Confessio Amantis V 7597-7608), and regard as inadequate the meaning ascribed to Ben Jonson's plays Sejanus and Volpone by their author's simple exhortatory and monitory moralisations of them: Sejanus IV v 898-903, Volpone V v xii 149-51, in Ben Jonson ed. C. Herford and P. Simpson (Oxford 1925-52) IV 470, V 136.

19. Ricardian Poetry p. 109.

20. Li Livres dou Tresor de Brunetto Latini ed. F. J. Carmody (Berkeley and Los Angeles 1948) pp. 175-225. Brunetto's discussion of particulars is limited by his omission of the crucial chapter of the Ethics. Giles draws together Aristotle's whole definition of ethics: see H. E. Childs A Study of the Unique Middle English Translation of the De Regimine Principum of Aegidius Romanus (MS Digby 233) (Ph.D. thesis, University of Washington 1932) pp. 191-7.

21. Childs op. cit. p. 191.

22. Ibid.

23. Ibid. p. 194.

24. Ibid. pp. 195, 196.

25. Ibid. p. 196.

26. Ibid.

27. Anecdotes historiques ed. de la Marche p. 4.

28. Artes Praedicandi ed. Charland pp. 396-7.

29. A. J. Minnis ‘Langland's Ymaginatif and Late-Medieval Theories of Imagination’ Comparative Criticism III (1981) 72.

30. James of Vitry, quoted in Crane op. cit. p. xli n.

31. James of Vitry, quoted in Crane, op. cit. p. xxxix n.

32. Humbert of Romans, quoted in Welter op. cit. p. 73 n.

33. James of Vitry, quoted in Crane op. cit. p. xlii n.

34. Quoted in M. D. Howie Studies in the Use of Exempla (London 1923) p. 11. The suspicion of certain kinds of material is still present in the fifteenth century, as this heavily sarcastic priest had found to his cost: ‘I rede in haly wryte, I sey noght at I rede in ouidie, noyther in oras. Vor þe last tyme þat I was her ich was blamyd of som men word, because þat I began my sermon wyt a poysy. And ter-vorn, I say þat I red in haly wryt, in þe secund book of haly wryt, þat I suppoise be sufficiant inowgh of autoritee …’. Three Middle English Sermons from Worcester Chapter MS F. 10 ed. D. M. Grisdale, Leeds School of English Language, Texts and Monographs V (Leeds 1932) 22.

35. For Boccaccio, see Stierle op. cit. pp. 186-90. Guillaume de Machaut has the lady in Le Jougement du Roi de Navarre accuse the poet-narrator of using exempla that do not fit the significance he is trying to impose upon them, so creating a complex situation in which uncertainties about truth and authority create the complex tone of the work: Oeuvres de Guillaume de Machaut ed. E. Hoepffner (Paris 1908-21) I 136-282. In The Canterbury Tales, Chaucer displays a similar interest in the process by which tales signify, and repeatedly used moralisations that he showed to be unsatisfactory, inappropriate or misleading to comment on tales, tellers, and the difficulty of defining a necessarily subjective experience. For a close analysis of some of Chaucer's techniques, see Anne Middleton ‘The Physician's Tale and Love's Martyrs: “Ensamples mo than Ten” as a Method in the Canterbury Tales' The Chaucer Review VIII (1973) 9-32.

36. D. A. Pearsall Gower and Lydgate British Council: Writers and their Work CCXI (London 1969) p. 6.

37. Ricardian Poetry pp. 94-5.

38. Ibid. p. 95.

39. The best criticism of Gower as a narrative artist is Pearsall ‘Gower's Narrative Art’ pp. 475-84; C. S. Lewis The Allegory of Love (Oxford 1936) pp. 201-13; and Speed op. cit. passim. There is an important discussion of Gower's interests and techniques in R. Woolf ‘Moral Chaucer and Kindly Gower’ in J. R. R. Tolkien, Scholar and Storyteller: Essays in Memoriam ed. M. Salu and R. Farrell (London 1979) pp. 221-45.