[In the essay below, Laurie investigates Deschamps's assertion in that poetry is an art form independent of music, finding the poet's views derived from his own poetry.]

In his exegesis of the passage in the Art de Dictier in which Deschamps describes poetry as ‘musique naturele’ and music as ‘musique artificiele’, Dragonetti points out that Deschamps' belief in poetry as an art form independent of music marks a break not only with the whole of French lyric tradition, in which the chanson is the synthesis of word and music, but also with that transcendental view of music which medieval theoreticians borrow most commonly from Boethius.¹ Dragonetti's argument is based entirely on Deschamps' theoretical work and on the treatises of his predecessors. It is worth examining Deschamps' theory in order to determine whether it had any practical application to his own work and to that of his contemporaries.

Deschamps explains that he describes music as ‘artificiele’ because it consists of nothing more than rules which can be learnt by the most insensitive of men and distinguishes it from poetry, a ‘musique naturele’ with which the poet is endowed by nature and which cannot be learnt. A musical accompaniment, so far from being necessary to poetry, is a disadvantage whenever it is desired to read poetry in private and without the services of several performers. There is an even more fundamental difficulty: poets, according to Deschamps, ‘ne saichent pas communement la musique artificiele ne donner chant par art de notes a ce qu'ilz font …’ (vii, 271). The importance of this admission may be appreciated when it is remembered that Deschamps was in this position himself and that he belonged to the first generation of French lyric poets, Froissart, Wenceslas de Bohême, and Oton de Grandson, who finally abandoned music as a constitutive element in their work. It is at this point that Deschamps' intention becomes most clear: he is attempting to justify not only his own practice but that of his contemporaries. Poets like himself have lost nothing at all by their incapacity to read or write music and their poetry has a better claim to be called music than what usually goes by this name.

Yet music has a claim to a special position amongst the arts. Deschamps describes it as ‘la medecine des. VII. ars’, an aid to relaxation and a distraction from the other arts (vii, 269). Dragonetti argues that this place should have been assigned to poetry, described in the Art de Dictier as the pre-eminent category of music, and believes that there is an illogicality in this part of Deschamps' argument. It is possible that there is no illogicality, for the place assigned to music by Deschamps is relatively inglorious: the last of the arts and the servant of them all; an escape from more serious matters. Dragonetti points out that there is precedent for Deschamps' non-transcendental view of the pleasure afforded by music in Roger Bacon and in Jean de Crocheo but does not note that there is a closer parallel for it in Machaut's Prologue:

Et Musique est une science
Qui vuet qu'on rie et chante et dence;
Cure n'a de merencolie
A chose qui ne puet valoir,
Eins met telz gens en nonchaloir.
Par tout, où elle est, joie y porte;
Les desconfortez reconforte,
Et nès seulement de l'oïr
Fait elle les gens resjoïr.(2)

Machaut includes music amongst the free gifts of Nature and considers that one of its chief glories is its place in the services of the Church on earth and in the praise offered to God by the angels and saints in heaven, but he is as far from making music a transcendental philosophical principle as Deschamps and reduces music to the level of an entertainment or even a pick-me-up.

Yet Machaut and Deschamps are more articulate in their praise of the aesthetic pleasure to be derived from the sound of music than from the sound of poetry. If there is an illogicality in this part of the Art de Dictier it might be explained by Deschamps' dependence on Machaut for his conception of the aesthetic qualities of vocal and instrumental music and by his separation from his master over the question of whether this music was, like poetry, a free gift of Nature. It is, however, improbable that this passage reveals nothing more than an editorial oversight on Deschamps' part. It is often argued that medieval poets had little appreciation of the possibilities of the mere sound of words as a means of conveying emotion.³ The fact that Deschamps, like Machaut, was well aware of the pleasure to be derived from the sound of music can give the critic no guarantee that he had anything more than a rudimentary appreciation of this aspect of the musicality of verse. Poetry seems to have this quality principally when it is provided with a musical accompaniment: ‘Et semblablement les chançons natureles sont delectables et embellies par la melodie et les teneurs, trebles et contreteneurs du chant de la musique artificiele’ (vii, 271-2). The fact that Deschamps describes poetry as music at all might be explained without any reference to the musicality of his own verse: Deschamps was arrogating to poetry alone the position which had belonged to poetry and music in earlier theoreticians, for example, in Roger Bacon. According to this view, again expounded in the fourteenth century by Philippe de Vitry,⁴ rhetoric and poetry are branches of music and it is not surprising that Deschamps should have continued to describe poetry in these terms in his treatise.

Yet Deschamps does attempt to explain the theory of poetry as music. Poets have the right to call their work music ‘… pour ce que les diz et chançons par eulx faiz ou les livres metrifiez se lisent de bouche, et proferent par voix non pas chantable, tant que les douces paroles ainsis faictes et recordées par voix plaisent aux escoutans qui les oyent …’ (vii, 270-1). Poetry is: ‘une musique de bouche en proferant paroules metrifiées …’ (vii, 270). It ought to be read aloud: ‘… pour ce que neant plus que l'en pourroit proferer le chant de musique sanz la bouche ouvrir, neant plus pourroit l'en proferer ceste musique naturele sanz voix et sanz donner son et pause aux dictez qui faiz en sont’ (vii, 271). In these statements Deschamps returns to the same idea: poetry is music because it is read aloud and the human voice establishes its metre, rhythm, and stress. By ‘son et pause’ Deschamps may be referring to the stress on or pause after the syllables forming the caesura and rhyme. Music itself often played an auxiliary role in establishing the ‘pause’ on caesura and rhyme, sometimes by stressing both these points in the line with long notes. Deschamps may be referring to these advantages of music as a means of emphasizing metrical accents when he praises the accompanied lyric: ‘Et aussi ces deux musiques sont si consonans l'une avecques l'autre, que chascune puet bien estre appellée musique, pour la douceur tant du chant comme des paroles qui toutes sont prononcées et pointoyées par douçour de voix et ouverture de bouche …’ (vii, 271).

It is clear that the abandonment of music in the French lyric involved the loss of one of its most important auxiliary metrical elements. Deschamps' concept of the musicality of verse as residing in metre and competent elocution, and his insistence on the necessity of reading aloud, may indicate that he was aware that the lyric was in danger of becoming impoverished by the loss of a musical accompaniment on the one hand and the growth of silent reading on the other. If so, and if he had anything more than a superficial conception of the musicality of verse, it might have been expected that he would have interested himself in compensatory devices in order to avoid the dangers of an indeterminate metrical structure and ensure that his verse was not read like prose.

To judge by the Art de Dictier the chief device of this kind which appears to have interested Deschamps, apart from the regularization of the fixed lyric forms themselves, was the development of rhyme. He attempts to define grammatical rhyme, illustrates the meaning of the words equivoque, retrograde, leonine, and sonant, and recommends a mixture of masculine and feminine rhymes. He also explains the construction of the balade equivoque, retrograde et leonine and describes these poems as ‘les plus fors balades qui se puissent faire’ (vii, 277). This passage is the subject of an extraordinary piece of exegesis by Lote:

[Deschamps] donne la mesure de son esprit frivole. … Il mesure donc l'intérêt que présentent les poèmes à la difficulté de leur facture, ce qui apparaît de la manière la plus évidente au cours du développement qu'il consacre à la Ballade equivoque, retrograde et leonine. Il ne nous cache pas l'admiration qu'il ressent pour de pareils tours de force, où se manifeste l'habileté de l'ouvrier. … Ces raffinements le transportent et le ravissent. Son bonheur, quand il songe à d'aussi prodigieuses merveilles, ressemble à celui d'un poète parnassien qui aurait écrit un sonnet sur des rimes rarissimes, ou à celui d'un enfant qui aurait mis la main sur quelque jouet compliqué. …⁵

Hoepffner's comment upon the same passage is that Deschamps reveals a typical medieval outlook and that it led him to compose worthless poems which are frequently contorted and incomprehensible.⁶ None of Deschamps' critics have done him the elementary justice of pointing out that in the total number of his one thousand and fourteen⁷ ballades only “Ballades 461” and “477” are equivoques, retrogrades and leonines, and only “Ballades 9” and “18” are retrogrades. In “Ballade 999,” which is used by Deschamps as an example of grammatical rhyme (vii, 277-8) the technique is used consistently only in the first strophe. Even when “Rondeau 618,” which is equivoque and retrograde and “Rondeaux 930-1,” which are equivoques, are included, the total is astonishingly small for a form which Deschamps is supposed to have considered as the summit of his art. Elsewhere in the Art de Dictier, Deschamps does not hesitate to make express recommendations about the length of line and strophe, the position of the rhymes, the mixture of masculine and feminine rhymes, the structure of the envoi, the number of strophes in the virelai, etc. (vii, 274-81). Yet when describing the above word-games, he does not explicitly recommend their use but contents himself with describing them as extremely skilful. In this respect his own practice accords with his theory: acrostics and other word-games make no more than an occasional appearance (“Ballades 73,” “460,” “540,” “947,” “1312”; “Rondeaux 655,” “1326”; “Virelai 743”), the rime retrograde and grammaticale are rare curiosities and even the rime equivoque is seldom used in more than two lines of any rondeau, virelai or ballade. The sparing use made of grammatical rhyme is typical; Deschamps tends to use it, not as a frivolous ornament but to emphasize the subject of a poem in the first two lines:

Qui faire veult aucun fort edifice,
Neuf choses fault a son ediffier

“Chanson Royale 391,” ll. 1-2 (cf. “Ballades 1458,” “1476,” “1485,” “1486,” “1491,” “1492”)

In these respects he is as restrained as the trouvères and as Guillaume de Machaut.⁸

Deschamps' account of rime leonine and rime sonnante and his recommendation of a mixture of masculine and feminine rhymes (vii, 274-6) have more relationship to his own practice. The rime leonine bears on two vowels and the rime sonnante on one (described by Deschamps as ‘entiere sillabe’ and ‘demie sillabe’ respectively). He does not follow the Leys d'Amors in distinguishing between perfect and imperfect rime leonine and rime sonnante, according to whether they are preceded or not by a supporting consonant. As he reckons the feminine ending in mute ‘e’ as an additional vowel in the line, but not as a sufficient rhyme without a supporting vowel, every feminine rhyme is also leonine: homme, Romme. Only masculine rhymes are sonnantes: clamer, oster; in order to become leonine the masculine rhyme, like the feminine rhyme, must bear on two vowels: defenir, maintenir.

Two of the three ballades given by Deschamps to illustrate these distinctions contain only leonine rhymes, and present an irregular mixture of masculine and feminine rhymes. The remaining example, which consists of a regular alternation of masculine and feminine rhymes, is described as ‘moitié leonime [sic] et moitié sonant’. A mixture of masculine and feminine rhymes is explicitly recommended in the paragraph immediately following: ‘Et se doit on tousjours garder en faisant balade, qui puet, que les vers ne soient pas de mesmes piez, mais doivent estre de .ix. ou de .x., de .vii. ou de .viii. ou de .ix., selon ce qu'il plaist au faiseur, sanz les faire touz egaulx, car la balade n'en est pas si plaisant ne de si bonne façon.’ Raynaud notes that Deschamps gives no indication as to the proportions in which masculine and feminine rhymes ought to be mixed,⁹ and Lote argues that these lines are merely a recommendation of ‘un mélange à peu près équilibré’ and must not be mistaken for the late fifteenth-century rule of regular alternation.¹⁰ According to Lote, alternation is never more than an exceptional and rare phenomenon in medieval French verse, but Reaney argues that ‘the alternation of masculine with feminine rhyme seems to have been almost an accepted thing in the fourteenth century, though it was not yet a rule’ and illustrates this from Machaut.¹¹ In view of the subsequent importance of this rule it is worth while analysing Deschamps' ballades and chansons royales in order to discover how close he had come to it himself and what practical importance may be attached to his own recommendations in the Art de Dictier. Only one hundred and fifty-four of his one thousand and fourteen ballades and fifteen of the one hundred and thirty-eight chansons royales have only masculine rhymes; eleven ballades and one chanson royale have only feminine rhymes. Two hundred and twenty ballades and twenty-four chansons royales have regular alternation of masculine and feminine rhymes; the remaining six hundred and twenty-nine ballades and ninety-eight chansons royales have an irregular mixture of masculine and feminine rhymes: the ouvert and the clos, or the tierce, or the envoi may be the only part of the poem to present alternation. The progress towards regular alternation in these poems may be appreciated when they are compared with the twelfth- and thirteenth-century chanson. Dragonetti analyses one thousand and eight chansons of the trouvères and counts six hundred and ninety-eight with a mixture of masculine and feminine rhymes, two hundred and eighty-nine with only masculine rhymes and twenty-one with only feminine rhymes.¹² Sixty-nine per cent of the poems analysed by Dragonetti have alternation or mixture of rhymes as against over eighty-three per cent of Deschamps' ballades, and eighty-eight per cent of his chansons royales.

Deschamps appears to consider the avoidance of an uninterrupted succession of masculine or feminine rhymes as an ornament to verse and perhaps as a structural merit: ‘… la balade n'en est pas si plaisant ne de si bonne façon’ (vii, 276). One reason for the high proportion of feminine, leonine or polysyllabic rhymes in his ballades and chansons royales may be that music no longer served to accentuate and differentiate rhyme, and the poet, continuing to borrow the musician's technique, lengthened the rhyme by extending it to more than one syllable. This does nothing to explain Deschamps' concern with the mixture of masculine and feminine rhymes. The use of stereotyped form and more particularly the disappearance of the heterometric ballade and chanson royale in Deschamps and his contemporaries had resulted in monotony and in a lack of variation in the strophe; a development of the practice of mixing lines of seven and eight, eight and nine, nine and ten syllables, as Deschamps calls them, together with the frequent use of the ballade layée where the line is decasyllabic, may have been the most accessible compensatory devices. It is likely that the predominance of the stereotyped isometric lyric in the fourteenth and fifteenth centuries was one of the most important contributory elements in the process which led to the formulation of the rule of alternation at the end of the fifteenth century.

Deschamps' concern with the accentuation and diversification of rhyme in the ballade and chanson royale did not lead him to formulate any rules for the reinforcement of the other metrical ‘pause’ in the line, namely the caesura. The caesura is regularly observed in Deschamps' decasyllables¹³ but he does not appear to favour the most common method of ornamenting and reinforcing the caesura: that is, by making it rhyme with the following caesura or with the rhyme itself (“Ballades 31” and “1315” are exceptions). The ordinary method of emphasizing ‘son et pause’ in Deschamps' lines is not the construction of these jingles but the use of alliteration as a structural device.

Lote, who considers examples of alliteration on the caesura and rhyme or on successive caesuras in medieval French verse, is inclined to dismiss them as mere chance and denies that they ever act as a structural device.¹⁴ In his review of the whole question of alliteration in medieval French verse, including the work of Köhler, Mone, Stengel, Hoepffner, Du Méril, Meyer and Riese, Lote appears to base his sceptical attitude principally upon twelfth- and thirteenth-century examples. The same impression may be derived from Dragonetti, who overlooks this question in La technique poétique des trouvères dans la chanson courtoise. By the fourteenth century it is more difficult to dismiss the practice as a mere accident of language; alliteration is frequently used in successive lines on the first syllable of the rhyming word and of the word forming the caesura and also on the first syllable of the line. Examples of the type:

Loer m'en doy en tant que desirer
Tres loiaument, sans penser deshonnour,
Me fait dame qui est de flours la flour.
Mais je m'en doy pleindre aussi durement

are common in Guillaume de Machaut and Oton de Grandson.¹⁵ In Deschamps the technique is most frequently used in the first two lines, in the refrain or in the envoi:

Prince, que veult avoir dominion
Doit doubter Dieu, es vertuz se deporte

(“Ballade 1019,” ll. 25-6)

(Cf. “Ballades 34,” “43,” “87,” “132,” “161,” “162,” “166,” “167,” “174,” “276,” “391,” “417,” “456,” “494,” “503,” “857,” “858,” “871,” “1015,” “1045,” “1050,” “1087.”) In these instances alliteration (as was the case with grammatical rhyme) brings an important line or word into relief: a word at the caesura or rhyme bears one of the principal metrical accents and is emphasized further by alliteration. The device is also used frequently on the first syllable of the line and occasionally on the first syllable of any other word in a line which the poet wishes to emphasize. There are also some instances in which Deschamps uses it to mark a pause within the line in a series of run-on lines:

Couvoiteus suis, blans et chanus,
Eschars, courroceux; j'adevine
Ce qui n'est pas, et loe plus
Le temps passé que la dotrine
De temps present; mon corps se mine

(“Ballade 1266,” ll. 11-15)

Alliteration used for any purpose other than emphasis in Deschamps is less common. “Rondeau 1326,” which is a series of almost unintelligible and disconnected nouns beginning with ‘p’, may be described as humorous or relentless according to the reader's point of view. In “Virelai 548,” the unpleasant features of campaigning in Flanders are set in relief by laborious, arthritic alliterative effects in f, p and s:

Puis que j'ay passé le Lis,
Je seray gais et jolis
En ce doulz pais de France,
Et vivray a ma plaisance,
Maugré Flandre et la pais
Ou j'ay toudis fait penance,
Porté bassinet et lance,
De cote de fer vestis,
Geu aux champs en grant doubtance,
En faim, froit, pluie et soufrance
Sanz couvert, sanz avoir lis,
Et encor me faisoit pis
Wacarme, alarme et les cris
Des Flamens, que ma finance
Ne que toute ma despence

(ll. 1-15)

It is precisely this kind of ‘musique naturele’, which includes not only alliteration but the arrangement of words, the rhythm of the line, the quality of vowels and consonants, onomatopoeia, which it is most interesting to find in Deschamps. He appears to have been well aware that the sound and rhythm of a line of verse might be suited to the words and add to their literal meaning. Reflections on the vanity of life are presented in solemn, slow-moving decasyllabic lines:

Charongne a vers, povre fragilité,
Qui puez estre comparée a la rose
Qui est boutons et naist ou temps d'esté.

(“Ballade 969,” ll. 1-3)

His satire on affected courtesy takes the form of a conversation in disjointed octosyllables:

Je m'en voys—Demourez icy,
Beaus cousins.—Certes, non feray:
Je ne vous lairay pas ainsi.

(“Ballade 1031,” ll. 1-3)

When Deschamps attacks Montagu as an effeminate dandy, the lines themselves take on a derisively mincing parallel rhythm:

Le baut, le doulx, le poupinet,
Le long, le droit, le gay, le savoureux,
Le gentil corps et le chief crespelet,
Megre ne gras, au viaire piteux.

(“Ballade 784,” ll. 1-4)¹⁶

The alliterated refrain of the same poem, uninterrupted, solid and emphatic, discloses the concrete reason for Deschamps's resentment and offers a contrast with the first lines:

Milleur marchié a fait de ma maison.

“Ballade 1202,” in which Deschamps describes birds screeching around the tower of Fismes, contains some of his most successful onomatopoeic effects:

A souleil couchant sur le soir
Deslogent de leur carrefour
Cahuans, suettes, pour voir,
Qui chantent chans plains de tristour;
Toute la nuit font grant freour
Aux veillans; de mort sont appeaulx.

(ll. 25-30)

Other examples of these techniques are not difficult to find, but even when every concession is made to subjective judgement in evaluating them, it is impossible to pretend that they prove that Deschamps was more sensitive than his predecessors to the musicality of the unaccompanied lyric. Similar techniques are studied by Dragonetti in the poetry of the trouvères¹⁷ and there is no reason to suppose that they are any less effective. If the musicality of Deschamps' verse can be distinguished from that of the trouvères in any respect it is by techniques which were common in Machaut and in Deschamps' contemporaries: the more frequent use of alliteration for emphasis, the preference for polysyllabic rhymes and a development of the practice of mixing masculine and feminine rhymes. It would be difficult to argue on this evidence that Deschamps' concept of poetry as music has any great practical importance in his own work. He may have been the first medieval theoretician to claim a pre-eminent place for the purely literary lyric but his subordination of music to poetry might be explained as nothing more than an attempt to justify the incapacity of himself and contemporary poets to write music by maintaining that they were dedicated to a higher art. The idea of poetry as natural music might be described as an intuition or merely a counter in argument, the true importance of which was lost upon its author.

Yet this idea does have a practical importance in Deschamps' work in so far as it is an attempt to define the inspiration of the poet: ‘L'autre musique est appellée naturele pour ce qu'elle ne puet estre aprinse a nul, se son propre couraige naturelment ne s'i applique …’ (vii, 270). It is easy to exaggerate Deschamps' originality in this conception of the inspiration of the poet. Siciliano has no reservations: ‘C'est seulement à la fin du XIVe siècle que ce brave homme d'Eustache Deschamps s'aperçut d'une chose simple et profonde, à savoir que la poésie est une “faculté naturelle” qui “ne puet estre aprinse a nul”.’¹⁸ It is not necessary to look any further than Machaut's Prologue to demonstrate the superficiality of Siciliano's view. Machaut makes it perfectly clear that he regards the poet as a privileged being, who has not chosen his own vocation but has been called to it by Nature. His conception of poetry is similar to that of the troubadours and trouvères: poetry and love are two aspects of the same thing; true love is the proper inspiration of the lyric. Deschamps' view may also be a development of the earlier theory that the true poet, as distinct from the mere technician, cannot write without the inspiration of love. The trouvères frequently argue that the poet's true feelings about his lady must be the same as those he describes in his verse; if he relies solely on rhetoric and writes without genuine love his verse is a counterfeit.¹⁹

The difference between this view of the sincerity of the poet-lover and Deschamps' account of the inspiration of the poet is probably best understood in terms of Deschamps' own contribution to the subject-matter and style of the French lyric. As long as it was maintained that love was the only theme worthy of lyric poetry it was possible, at least in theory, to distinguish between the craftsman and the true poet solely on the grounds that the one merely adopted the postures of the conventional lover while the other was genuinely in love with his lady. Deschamps himself asserts that love is his only source of inspiration: ‘Car sanz amours ne puis faire chanson.’ “Ballade 493,” in which this line occurs, is a companion piece to “Ballade 447,” in which Deschamps, with apparent lack of respect for Machaut, asks the Péronne of the Voir Dit to receive him as her lover now that Machaut is dead. In “Ballade 493” he informs another lady, Gauteronne, that as Péronne has turned down his request, she may, perhaps, consider acting as a stand-in herself. In both poems he is ironically adopting the postures of the conventional lyric poet, here represented by Machaut, and at no point is his misrepresentation of his own work more grotesque than in his mocking claim that he cannot write lyric poetry without the inspiration of love. He wrote no less than one hundred and seventy-three rondeaux,²⁰ one hundred and thirty-eight chansons royales, eighty-four virelais²¹ and one thousand and fourteen ballades. Only seventy-five of the rondeaux, fifty-seven of the virelais, seven of the chansons royales and one hundred and eighty-seven of the ballades can be described as love lyrics; the remaining poems deal with a wide variety of other subjects, ranging from scenes of criminal life to descriptions of royal châteaux, from the horrors of campaigning in Flanders to toothache, from moralizing on ecclesiastical and political disorders to the price of cabbages in Paris. Deschamps' lyric is the vehicle for the begging letter, the moralist's lament, the humorous dialogue, the medical prescription, the political diatribe and the traveller's tale. These poems are not in the tradition of the trouvère chanson or of the fixed lyric forms of the thirteenth and fourteenth centuries, which were ordinarily confined to conventional love. Exceptions amongst the rondeaux and virelais are negligible: Adam de la Halle has a humorous rondeau on a broken-down horse, one of the rondeaux interpolated in the Roman de Fauvel condemns the vices of society in the person of Fauvel, ten of the thirty-six rondeaux in the Tresor amoureux praise Nature, Free Will, Reason and Virtue, two by Froissart are on the themes of contentment and Fortune, and there is one anonymous political virelai, the Chanson sur le nouveau fort of 1375. Amongst the ballades exceptions are more numerous: Adam de la Halle uses it for a noël; in the Roman de Fauvel and the Tresor amoureux it becomes didactic; Guillaume de Machaut, Philippe de Vitri, Jehan Campion and Jehan de le Mote occasionally use it to discuss the principles of poetry and Jehan de le Mote also uses it for a funeral lament. Froissart writes a few humorous and satirical ballades and provides Deschamps with the model for the political pastourelle. Oton de Grandson writes a ballade on the injustices to which he was subjected after the death of the Comte Rouge in 1391. With these exceptions the fixed lyric forms of the thirteenth and fourteenth centuries are indistinguishable in their inspiration from the trouvère chanson. What is exceptional in Deschamps' predecessors and contemporaries becomes the rule in his own work. Most of his lyrics are in the tradition of non-strophic or unaccompanied verse, for example, the dit, or even prose, rather than of the Northern French lyric. There is nothing extraordinary about the fact that one of the first French lyric poets who abandoned music as an element in his work should have taken his inspiration from the unaccompanied dit rather than from the accompanied lyric. It is more surprising that his own contemporaries, Wenceslas de Bohême, Oton de Grandson, and Jean de Garancières should have been so conservative in this respect. Account must be taken of the exceptions in Froissart but even in his work the lyric is normally a love poem: for him as for Machaut other themes are best treated in the dit. Even in the dit the same tendency is evident: Froissart explains to Philosophy in Le joli buisson de Jonece that one of the reasons he has given up writing poetry altogether is on account of the corruption of the world. Machaut, in Le jugement dou roy de Navarre, finds the horrors of the Black Death so inexplicable that he decides it is not his business to set the world to rights and consoles himself instead with the problems of love. Deschamps occasionally presents the view that it is useless to teach or preach in a decadent age (“Ballades 43,” “63,” “97,” “100,” “1063,” “1389”), but never argues, as Machaut does, that the poet ought to take refuge in conventional love. In “Ballade 1246” he explains that it is precisely because of the decadence of the age that he refuses to write love poetry and turns instead to condemning disorders and vices in Church and State. In the Art de Dictier he associates the idea that poets do not concern themselves with writing music with the argument that the lyric is not necessarily concerned with love and that the subject-matter of verse does not compromise the claims of any of it to be described as natural music: ‘Et ja soit ce que ceste musique naturele se face de volunté amoureuse a la louenge des dames, et en autres manieres, selon les materes et le sentement de ceuls qui en ceste musique s'appliquent, et que les faiseurs d'icelle ne saichent pas communement la musique artificiele ne donner chant par art de notes a ce qu'ilz font, toutesvoies est appellée musique ceste science naturele …’ (vii, 270-1).

It appears that the traditional theory of poetical inspiration was too circumscribed to be made to fit Deschamps' lyrics. His own formulation marks an appreciable progress towards generalization and abstraction. He exchanges the naïve view of the sincerity of the poet-lover for the theory that the art of poetry, unlike that of music, cannot be learnt, except by those who are born poets. It is possible that this idea is a version of the divine frenzy topos, described by Curtius as a curious concomitant to what he believes to be the more typical medieval view: that is, that poetry was ‘sweat-producing labour’.²² Yet Deschamps' description of the unaccompanied lyric as the noblest of art forms and his rejection of the view that the essential element in the poet's inspiration was the reality and depth of his feelings for a particular lady are so convenient a reflection of his own practice in the lyric that it is difficult to believe they are nothing more than a variant of a commonplace and that his theory was any less distinctive than his practice. He is the first French poet to break decisively with trouvère conventions of subject-matter in the lyric and also the first to describe the French lyric as poetry rather than song. It is in this sense that Deschamps' lyric is a substantial and autonomous art form, a ‘musique naturele’.

Notes

1. R. Dragonetti, ‘La poésie … ceste musique naturele; Essai d'exégèse d'un passage de l'Art de Dictier d'Eustache Deschamps’, in Fin du Moyen Age et Renaissance, Mélanges de Philologie française offerts à Robert Guiette (Anvers, 1961), pp. 49-64, on Deschamps, Art de Dictier, vii, 270-2. The notes in the text refer to the volume and page numbers of the SATF edition of Deschamps' work.

2. V. Chichmaref, Guillaume de Machaut: Poésies lyriques (Paris, 1909), Prologue, ll. 198-206.

3. Cf. G. Lote, Histoire du vers français, 3 vols. iii (Paris, 1949-55), 25-8, 49, 59-60, 67-70.

4. G. Reaney. ‘The Ars Nova of Philippe de Vitry’, Musica Disciplina, x, 5-33.

5. Lote, ‘Quelques remarques sur l'Art de Dictier d'Eustache Deschamps’, Mélanges Hoepffner, Publications de la faculté des lettres de l'Université de Strasbourg, cxiii (Paris, 1949), 365.

6. E. Hoepffner, Eustache Deschamps Leben und Werke (Strasbourg, 1904), p. 132.

7. Hoepffner, op. cit. p. 126, counts 1013 ballades and 137 chansons royales by Deschamps; G. Raynaud, Eustache Deschamps, Sa vie, ses œuvres, son temps (vol. xi of the SATF edition of Deschamps' works), pp. 115 and 122 counts 1017 ballades and 139 chansons royales; the present writer counts 1014 ballades and 138 chansons royales. Raynaud, op. cit. p. 122, note 2, describes Ballade 351 as an incomplete chanson royale in four strophes. As the last ‘strophe’ begins like an envoi: ‘A Nynive, roys, princes, contes, ducs, / Aiez regart …’ (ll. 29-30), the poem is best described as a ballade; in Chansons Royales 346 and 1164 the envoi is also identical in form with the strophe. Ballades 829, 1379, and part of Ballades 145, 146, 147, 149 are attributed by Deschamps to other writers, but have not been deducted from the total. Poems attributed to Deschamps in vol. x have not been included: see Raynaud, op. cit. 107, 334.

8. On these techniques in the trouvères see Dragonetti, La technique poétique des trouvères dans la chanson courtoise. Contribution à l'étude de la rhétorique médiévale (Bruges, 1960), p. 421. On Machaut see Reaney, ‘The poetic form of Machaut's musical works’, Musica Disciplina, xiii, 37.

9. Raynaud, op. cit. p. 116.

10. Lote, op. cit. ii, 120-1. He points out that similar recommendations are made in the Leys d'Amors (op. cit. ii, 111-12).

11. Reaney, Musica Disciplina, xiii, 34; Lote, op. cit. ii, 119.

12. Dragonetti, op. cit. p. 147.

13. Lote, op. cit. i, 181-6, 221-46, maintains the contrary view. There are only twelve instances of the caesura splitting a word in the 262 decasyllabic ballades and chansons royales contained in the first two volumes of the SATF edition.

14. Lote, op. cit. iii, 49-70.

15. Chichmaref, op. cit. i, 19, ll. 3-6 and cf. 4, ll. 16-17, ll. 21-2; 17, ll. 3-4; 18, l. 2, ll. 14-15; 20, l. 1; 73, ll. 8-10, etc. G. Reaney, Musica Disciplina, xiii, 34-5, considers that not only rhyme but other forms of word-play (including alliteration) help to give unity to Machaut's rondeau. For examples in Grandson see A. Piaget, Oton de Grandson, sa vie et ses poésies (Paris, 1941), 197-8, l. 2, ll. 5-6, ll. 13-15, ll. 24-8, l. 32.

16. There are two syllables missing in l. 1.

17. Dragonetti, op. cit. pp. 515-38.

18. I. Siciliano, François Villon et les thèmes poétiques du moyen âge (Paris, 1934), p. 425.

19. Cf. Dragonetti, op. cit. pp. 21-30.

20. Hoepffner, op. cit. p. 134, counts 174 rondeaux by Deschamps; Raynaud, op. cit. p. 124, counts only 171. 175 rondeaux appear in the SATF edition, not 170 or 172 as is stated by Raynaud, op. cit. p. 124, note 2. In the total 175, Rondel 574 which appears again in the Art de Dictier in a different form has been reckoned again as vii, 285, but other rondeaux in the Art de Dictier which appear elsewhere have not been included; neither has Rondel 653, which is the same poem as Rondel 480. Rondels 1396 bis and 1396 ter, which appear as examples in the Art de Dictier, are ascribed by Raynaud to Deschamps and included in his total, but they were in fact written by Machaut: cf. Raynaud, p. 114, note 1; for the texts see Chichmaref, op. cit. i, 210; ii, 572. If these two poems are excluded there are in all 173 rondeaux by Deschamps.

21. Hoepffner, op. cit. p. 136, followed by Lote, op. cit. ii, p. 265 counts 76 virelais by Deschamps; Raynaud, op. cit. p. 126, reckons correctly 84 virelais.

22. E. R. Curtius, European Literature and the Latin Middle Ages, translated by W. R. Trask (1953), p. 475.