ABSTRACT
The Book of the Duches is usually dated to various points between 1368 and 1374. This article suggests that these various datings imply different interpretations of the poem, because at some point in this period John of Gaunt began an affair with Katherine Swynford, Geoffrey Chaucer's sister-in-law and governess to the children of Gaunt and Blanche, duchess of Lancaster, whose death is the topic of the text. This article explores how dating Book of the Duchess to before or after their affair began alters our understanding of Chaucer's representations of himself and Gaunt and his intentions in writing this literary tribute.


    Sometimes a poem cannot be understood without an intimate knowledge of the personal lives it involves. Book of the Duchess is such a poem, Chaucer's motives in writing this work arid its effects on his contemporary audience were intricately linked to a network of intersecting private and public relationships between Chaucer and John of Gaunt that developed after the death of the nobleman's first wife, Blanche, duchess of Lancaster. For this reason, dating the poem is essential to understanding it. We have known that Blanche died on 12 September 1368 since J. J. N. Palmer published a letter written in late 1368 from Louis de Mâle, count of Flanders to Queen Philippa touching upon the duchess's death, and thus the poem could not have been written before that date.(FN1)
    To move from this terminus a quo to a terminus ad quern requires much interpretation of circumstantial evidence, and thus allows room for debate and disagreement John of Gaunt's order of an alabaster tomb for Blanche for the sixth anniversary memorial of her death in 1374(FN2) has invited some to argue that Chaucer wrote the poem for a performance coinciding with, or as a part of, that ceremony.(FN3) Others hold 1371 or 1372 as the last possible year of composition. Palmer believes the poem could not have been written after Gaunt's marriage to Costariza, princess of Castile, in 1371, when it would have been seen as a satirical portrait of Gaunt as an 'Inconstant Lover'(FN4). Howard Schless disagrees, and points out that although we might find the poem inappropriate after Gaunt's remarriage, such a view is 'a projection of our own morality' onto Chaucer's day, when political marriages were the norm amongst the aristocracy.(FN5) The fact that Louis de Male's letter was written a scant three and a half months after the death of Blanche to suggest a 'mariage de mon trescher cousin le duc de Lancastre, votre filz'(FN6) underlines the fact that marriage and love were not necessarily linked for John of Gaunt's social class.
    Schless suggests that we should date the poem to after 'the end of 1371, or perhaps 1372', after John of Gaunt had received Edward Ill's permission to assume the title of King of Castile and Leon, based upon line 1314: 'With that me thoghte that this kyng',(FN7) although Edward I. Condren notes that this 'king' is in fact Octavian, whose hart-hunt, narrated earlier in the poem, is the subject of the preceding three lines:

And with that word ryght anoon
They gan to strake forth; al was doon,
For that tyme, the hert-hunryng.(FN8)

    That this king is not in fact Gaunt opens up a pre-1372 dating, and the possibility that the poem was written shortly after Blanche's death, perhaps at the request of Queen Philippa. Another clue which confirms such a date is the reference to Richmond in line 1319 ('ryche hil'), of which Gaunt was an earl until 20 July 1372, when Edward III transferred the earldom to John de Montfort to secure his loyalty to England in the struggle against France.(FN9) A pre-1372 dating has held much sway in recent years.(FN10)
    The poem's date is rarely related to its content and Chaucer's possible motivations in writing it. Here I would like to examine how each potential dating of the poem brings with it different assumptions about Chaucer's representations of himself and Gaunt, and I would like to argue that the relationship between Chaucer and Gaunt by way of Katherine Swynford, the poet's sister-in-law and the nobleman's mistress (and later wife), is central to our understanding of Book of the Duchess and should not be as marginalised as it has often been.
    At the time of Blanche's death, Chaucer was probably not in the country. On 17 July 1368 he was granted a licence to pass at Dover with 'vint soldz por ses despenses',(FN11) and appears next in the records on 20 October 1368, when a mandate ordered by Edward III was given to pay Chaucer his annuity in arrears.(FN12) At the same time, John of Gaunt was in England, quite probably, as Donald R. Howard suggests, at Blanche's bedside when she died.(FN13) Chaucer would most likely have heard of the duchess's death over a month after the fact, when the initial wave of Gaunt's and England's grief had begun to ebb. Surviving records give the impression that Chaucer's connections to Gaunt at this time were minimal. Possibly, Chaucer first met Gaunt when he was a page in the household of Elizabeth, countess of Ulster, where he served from 1357 to 1359,(FN14) but his minor position as one of the many pages in the Ulster household would not have allowed for anything more than a formal meeting. The records do not provide a tangible link between the two by the time of the duchess's death, and although it is possible that they gained some familiarity in the years 1361-1365, for which we have no records, it is more likely that the two did not interact too greatly, given Chaucer's low social status in comparison to Gaunt, then one of the most powerful aristocrats in England.
    A personal link between the two would not emerge until 30 August 1372, when the duke began to give annuities to the poet's wife, Philippa Chaucer.(FN15) Gaunt would not award Chaucer an annuity until nearly two years later,(FN16) which, as Paul Strohm has noted, did not compromise or contradict his loyalty to the king's court.(FN17) Although Chaucer's sister-in-law, Katherine Swynford, was attached to Blanche's court until the duchess's death, this does not suggest that John of Gaunt had more than a passing familiarity with Katherine, let alone any connection to her sister or brother-in-law, until she had begun her duties as the governess of Gaunt's and Blanche's children upon the death of their mother.(FN18) It was probably around this time that the two began having a romantic relationship. Sydney Armitage-Smith maintains that the affair between Gaunt and Katherine probably began in 1371 or 1372 since John Beaufort, their first child, was born in 1373.(FN19) Thus the first opportunity for meaningful contact between Gaunt and Chaucer would have been around this time and would have been the result of a private, intimate and non-political relationship. Whatever connection we attempt to reconstruct between these two men must begin about 3 years after Blanche's death in the private personal realm as far away from the public as Gaunt could be. The significance of this fact has not yet been brought to bear on Book of the Duchess.
    Whereas earlier scholarship accepted that the poem was most likely written for John of Gaunt, either under his patronage or as a plea for it, recent attention has focussed upon the heterogeneity of Chaucer's mixed audience, which included bourgeois readers. Paul Strohm notes that, 'Although the composition of this audience cannot securely be known, the confidence with which Chaucer approaches it would argue for relatively greater social equality than that existing between Chaucer and John of Gaunt'.(FN20) Previously held notions of Chaucer as a court poet have invited an about-face approach to the poem's audience, which ever increasingly excludes John of Gaunt. This false dilemma -- was the poem written for Gaunt or for Chaucer's social equals -- has assumed that the poem could not have been written for both, and it does not provide a satisfactory account of how Gaunt relates to this very intimate poem about his bereavement for his, first wife. Even if it were written for just about everyone else but Gaunt, as Strohm implies and more recent scholars have suggested,(FN21) it strains credulity to imagine that Chaucer would carry on writing his first lengthy narrative, on a highly contemporary topic involving the most powerful man in England without expecting the poem to catch its subject's attention at some point. If we maintain that Book of the Duchess was written at least partially for John of Gaunt, Chaucer's anticipation of the nobleman's reaction is a part of the fabric of the poem. Gaunt publicly mourned Blanche's death for the rest of his life, even after marrying Katherine Swynford over 20 years after their relationship had begun in January 1396,(FN22) so we can expect Book of the Duchess to have elicited thoughts of Gaunt's feelings toward Blanche while expressing, or creating, an emotional bond between the nobleman and poet regardless of when it was written. What changes with the various dates proposed are Chaucer's motivations in writing this work and creating or confirming this bond. The identification of the poem's audience with Gaunt invites questions of the poem's function in relation to his grief. One possibility is that it is a consolation. A recent proponent of this view is David Carlson, who argues that the poem is an instance of Chaucerian self-promotion.(FN23) Such an interpretation explains the flattering representation of Gaunt as the Black Knight and explains the narrator's apparent bumbling personality as a self-deprecating rhetorical persona employed to assert Chaucer's inferiority to the nobleman. These aspects of the work would best be understood as part of the poem's rhetorical fibre, constructed to sidestep the conundrum succinctly described by Arthur W. Bahr: 'Faced with the task of comforting her [Blanche's] grieving husband, yet unable to assume the appearance of superior emotional wisdom that would allow him to do so directly, Chaucer must work obliquely'.(FN24)
    A difficulty noted with this view is that the poem does not provide any sort of consolation whatsoever. Joerg O. Fichte argues that 'Chaucer gives an alternative to the obviously unsatisfactory forms of conventional consolation, and thus he sets out to eternalise the portrait of Blanche in the form of a literary encomium'.(FN25) The inquisitive narrator, according to this theory, serves as a strawman, 'who is built up in the beginning to be knocked down at the end, in order to contrast a contrived experience with real loss'.(FN26) However, if the poem becomes a eulogy, it is presented as something more:

Me thynketh in gret sorowe I yow see;
But certes, sire, yif that yee
Wolde ought discure me youre woo,
I wolde, as wys God helpe me soo,
Amende hyt, yif I lean or may.
Ye mowe preve hyt be assay;
For, by my trouthe, to make yow hool
I wol do al my power hool.(FN27)

    Here the dreamer timidly offers, without pretension, any sort of consolation he can. Yet we are made to feel this promise cannot be fulfilled as soon as it is made. The dreamer's repeated rhyme 'hool/hool' foreshadows the narrator's incompetence to offer the Black Knight any sort of comfort in words in the dialogue that follows. The narrator's promise of consolation unfilled, Chaucer can conclude the dialogue between the dreamer and knight only with a bathetic and meek acknowledgment of the Black Knight's suffering:

'She ys ded!' [says the Knight] 'Nay!'
'Yis, be my trou the!'
'Is that youre los? Be God, hyt ys
routhe!'(FN28)

    Commenting upon these lines, Minnis asks: 'For a climax to a long apotheosis of love and the lady, is not this rather disappointing?... The chime of the rhyme diminishes the emotional force of the exchange, making it sound inappropriately pat and curt'.(FN29) One possible reason why the narrator fails to console is that he lacks the linguistic tools with which to help. Phillip C. Boardman suggests that, in Book of the Duchess, Chaucer expresses the inadequacies of courtly language to console John of Gaunt.(FN30) Davis qualifies this observation by pointing out that Chaucer heavily borrows from Machaut for the bulk of the courtly language of the poem, resulting in a representation of the literary world originating with Machaut as 'fatally capricious in the historical context that Chaucer evokes'.(FN31)
    This observation, however convincing in light of the poem itself, does not provide much explanation for Chaucer's intentions in producing such a dynamic beyond a criticism of courtly language. The fact that this criticism is couched in the context of Gaunt's real loss suggests that Chaucer was communicating something more to Gaunt and his contemporaries. Chaucer attempts to offer language as a response to Blanche's death, but his language will ultimately fall flat in light of the reality of the death and Gaunt's human response, try as he may. And in trying and failing, he is able to maintain his apparent lack of 'superior emotional wisdom' as Bahr puts it, in relation to John of Gaunt At the same time, Chaucer's attempt and failure to use courtly discourse to relieve Gaunt's suffering gives clerkish affirmation to the overwhelming severity of his grief -- an approach which is comforting without presumption. This is the paradox of the poem: it consoles by admitting an inability to console.
    Such an interpretation assumes that an intimate emotional connection is developed within the poem -- a perception which is complicated by dating the poem to a point before any connection between Chaucer and Gaunt existed. If the poem were written before Gaunt's affair with Katherine Swynford, Chaucer would have been writing a poem about the bereavement of a figure far above himself in the social hierarchy. His motivation in writing this poem to John of Gaunt in this context would most obviously be to gain the favour of an incredibly powerful man -- in other words, the poem was a vie for patronage.
    Many other possible scenarios alter our understanding of Chaucer's motivation. If the relationship between Gaunt and Chaucer's1 sister-in-law had already begun and recently been made public,(FN32) an entirely different interpretation arises, whereby the poem becomes public testimony to Gaunt's love for Blanche during her lifetime and an attempt to denounce more cynical contemporaries who might have wondered (or gossiped) that Gaunt's relationship with Blanche's lady-in-waiting began not after her death and during Katherine's duties as governess to Gaunt's children, but during Blanche's lifetime -- perhaps even while Blanche's health waned. Chaucer's poem, when read by a person who has held this opinion or heard this rumour, appears to assert Gaunt's noble qualities and love for Blanche, which, Chaucer suggests, was so devout as to put Gaunt's soul in jeopardy.(FN33) If nothing else, the poem would be a strong argument against the gossipmongers. At the same time, Gaunt is allowed an entrance into literate courtly discourse via the poem and space to make his complaint known in the dignity of verse. In this context; Chaucer's motivation for writing such a work appears to have been to confirm a bond of loyalty to his potential protectorate which the newfound connection offered by Chaucer's sister-in-law would have made tenable, but was1 as of yet publicly unconfirmed by the duke.(FN34)
    Regardless of when the poem was written, it appears to be a public avowal to a wide public of loyalty to Gaunt at a time when he was incredibly unpopular amongst London society, as he would remain for most of the rest of his life.(FN35) Whether the Book of the Duchess was an attempt to secure further patronage or, as Strohm describes Chaucer's 1374 annuity from Gaunt, 'the protection of a second lord', or whether it is an impassioned attempt to console a grieving man, it appears to have been a public assertion of a personal relationship regardless of when the poem was written and performed, and regardless of whether that relationship actually existed or not. I would argue from this that Chaucer was acutely aware of how his poetry could be employed to mediate his own positionality amongst the aristocracy of his time and of the potential impact his poetry could have on his own person.
    This changes our understanding of the structure of the poem and the production of the 'narrator', who appears less of a literary figure -- or, as Stephen Davis calls him, a 'Machauldian figure'(FN36) -- and more as a rhetorical representation of Chaucer himself vis-à-vis Gaunt and a public audience at the centre of the socio-political structures of the early 1370s. A. C. Spearing's recent criticism of modern conceptions of narrators and narrative demands a reconsideration of what we mean by-the 'narrator' of the poem.(FN37) According to Spearing, we need not see a mimetic representation of a human being in the narrative voice of Middle English poetry, which he holds to be merely an aspect of narrative necessitated by the genre. So, in Spearing's view, Book of the Duchess 'almost completely separates the functions of the experiencing "I" from those of the narrating "I" The former is richly characterised, the latter not characterised, at all (and only briefly defined even as the writer of the poem)'.(FN38) This renders the narrator of the Book of the Duchess nothing more than a part of the narration, as Spearing suggests in no unclear terms: 'The subject of narration is not a self'.(FN39) Such a perspective allows us to look for a less literary and more rhetorical motivation for creating a first-person narrator in the poem, and it allows us to make sense of inconsistencies in the text without resorting to an anachronistic disapproval of a lack of unity in the poem's narrative voice.(FN40)
    A good example of this inconsistency in the narrator occurs fairly early on; when his usual obtuseness to everything human disappears to express a tender sensitivity to the Black Knight's emotional pain: When the narrator first notices the Black Knight, the Black Knight does not notice him. Chaucer is quick to justify the knight's behaviour by emphasising the depth of his sorrows:

Hym thoughte hys sorwes were so
Smerte
And lay so colde upon hys herte.
So, throgh hys sorwe and hevy
thoght,
Made hym that he herde me noght;
For he had wel nygh lost hys mynde,
Thogh Pan, that men clepeth god of
kynde,
Were for hys sorwes never so
wroth.(FN41)

    From this observation we know that the narrator is immediately aware of the Black Knight's agony, although the poem begins with a focus on the narrator's insensitivity:

But men myght axe me why spo
I may not slepe and what me is.
But natheles, who aske this
Leseth his asking trewely.
Myselven can not telle why
The sothe; but trewly as I gesse,
I holde hit be a sicknesse
That I have suffred this eight yeer;
And yet my boote is never the ner,
For there is phisicien but oon
That may me hele, but that is don.(FN42)

    The nature of the narrator's sicknesse has been debated for the better part of a century. Suggestions have, ranged from lovesickness to head melancholy.(FN43) Whatever it may be, the narrator professes his inability to provide a definitive answer, leaving his audience guessing and the narrator curiously undefinable. This opens potential responses to the narrator without giving too much information away -- a characterisation easily described as politic. And the narrator's insensitivity to and disinterest in his own suffering -- whatever its cause may be -- invites Chaucer's audience to see his character however they want, with the only caveat that he has a derth of human emotion.
    This lack is suspended when Chaucer focuses on explaining the Black Knight's behaviour when the narrator first stumbles upon him. The narrator's insensitivity soon returns again when he asks the grieving knight specific, direct, awkward questions, such as:

'Now, goode syre' quod I thoo,
'Ye han wel told me here before;
Hyt ys no nede to reherse it more,
How ye sawe hir first, and where.
But wolde ye tel me the manere
To hire which was your firste speche-
Therof I wolde yow beseche-
And how she knewe first your thoght,
Whether ye loved hir or noght?
And telleth me eke what ye have lore,
I herde yow telle herebefore'.(FN44)

    This graceless inquisition allows the Black Knight space to complain without appearing self-indulgent -- in fact, he appears to be indulging the narrator with an almost inhuman patience. The narrator's keen sensitivity to the Black Knight's misery upon first seeing the nobleman has a similar effect. As the apology for the Black Knight's inattention continues, his behaviour appears not only excusable, but quite sensible:

But at the last, to sayn ryght soth,
He was war of me, how y stood
Before hym and did of myn hood,
And hadygret hym as I best koude,
Debonayrly, and nothyng lowde.
He sayde, 'I prey the, be not wroth.
I herde the not, to seyn the soth,
Ne I sawgh the not, syr, trewely'.(FN45)

    The Riverside Chaucer glosses 'Debonayrly' as 'courteously, modestly', and Chaucer seems to emphasise the former meaning by adding the phrase 'and nothyng lowed' immediately after, inviting an understanding of the narrator at this particular moment as an otherwise lewd man who is trying his hardest to meet the courtly nature of the Black Knight, and failing miserably.
    What is made to seem most extraordinary is the Black Knight's following apology. Not only is the gap between Gaunt's and Chaucer's social positions enormous, but the audience is invited to assume that the humble, hat-holding narrator is easily missed, and understandably ignored:

I stalked even unto hys bak,
And there I stood as stille as ought,
That, soth to save, he saw me nought.(FN46)

    While the 'wel-farynge knight' sits 'even upright'(FN47) and makes his complaint, the narrator stands awkwardly behind him. The gap in civility between the two, implicit in the rigidity of their social code, is made palpable by the narrator's hat-holding and standing behind the Black Knight, waiting nervously for recognition.
    A comparison with one of Chaucer's primary sources for Book of the Duchess, Guillaume de Machaut, will help clarify how a narrator can be used as a figure of self-deprecation while the knight is described in unending acclaim. In Le Dit de la Fonteinne Amoureuse, the narrator lays in bed when he overhears the love complaint of a knight, whose identity is made clear in an anagram of Jean, duc de Berry at the beginning of the poem.(FN48) Machaut heaps superlative praise upon this figure:

Mais onques en jour de ma vie
Maniere qui fust plus jolie
En homme, n'en femme ne vi,
Et s'ot corps trop bien assevi,
Car il estoit grans, Ions et drois,
Bien façonnés en tous endrois,
Gens, joins, jolis, juenes et comtes.(FN49)

    This is a marked contrast from the 'narrator', who is little more than a caricature of the unmartial, bookish clerks often seen in French university towns in the fourteenth century:

Adont durement me doubtay
Et dedens mon lit me boutay
Il sambloit que j'eüsse fievres,
Car je sui plus couars qu'uns lievres,
Et si trambloie, et goute a goute
Suoie d'effroy et de doubte.
La gisoie en si petit point
Que, s'aucuns preist mon pourpoint
Ou ma sainture ou ma chemise,
Par moy n'i fust deffense mise;
Non, par Dieu! qui mon corps pre"ist
Ja remuer ne m'en va"ist.(FN50)

    This passage occurs less than 50 lines after the narrator has identified himself, also in an anagram, as Guillaume de Machaut, so, little exegesis is required to conflate the voice of this docile coward with Machaut himself.
    Chaucer's praise of the Black Knight is not so lavish, and his self-deprecation is not so bald, but subtlety does not mean absence, and there, seems to be little reason to understand Chaucer's representations of himself and Gaunt as anything other than an act of social deference to a man of considerable power and a political gesture to proclaim (or, depending on how we date the poem, to confirm) an alliance. The narrator in the dream maintains an inferior position, characterised by gaucheness, to allow Chaucer to affirm his inferior position to John of Gaunt in the real world of the poem's initial reception. If Chaucer's audience is invited to conflate the narrator with the poet himself, then he is inviting his contemporaries to see him as an awkward, hat-holding clerk outside the world of courtly love and, as is made abundantly clear, unable to understand it or its language no matter how explicit the case is made. When we consider the poem in its historical context, the reason for these characterisations becomes obvious: Chaucer wrote a poem with a keen sensitivity to Gaunt's suffering for all to hear in order to affirm his allegiance to the nobleman, while appearing beneath the courtly world he is at pains to represent.
    I hope that this study demonstrates that Gaunt's relationship with Katherine Swynford is essential to our dating and interpretation of Book of the Duchess because of the implications the affair could have had oh Chaucer's personal motivations in writing it. If we date the poem to before Katherine Swynford and John of Gaunt began their relationship, it can be seen as Chaucer's attempt to create a connection with Gaunt that was tenuous if extant. If it was written after the affair began and was well known, it can be seen as an attempt to confirm Gaunt's love for Blanche of Lancaster and to render questionable any rumour that he had been unfaithful with Swynford while publicly confirming an allegiance and loyalty that Chaucer's and Gaunt's private connections offered. Although we do not need to know the details of the human relationships behind the poem to appreciate it as a work sensitive to the sorrows of loss, our appreciation of it is heightened by a consideration of how Chaucer was reacting to the complexities of Gaunt's quickly changing personal life and emotional state while remaining tactful. Book of the Duchess displays Chaucer's awareness of the realities of life as it reveals his sensitivity to the sadness of death not in the abstract, but as they affected real human beings, including himself.
ADDED MATERIAL
    MICHAEL FOSTER
    The University of Nottingham

FOOTNOTES
1 J. J. N. Palmer, 'The Historical Context of the Book of the Duchess: A Revision', The Chaucer Review, 8 (1974), 253-61.
2 N. B. Lewis, 'The Anniversary Service for Blanche, Duchess of Lancaster, 12th September 1374', Bulletin of the John Rylands Library, 21(1937), 3-419.
3 For one of the most influential of these datings, see D. W Robertson, 'The Historical Setting of Chaucer's Book of the Duchess', in Medieval Studies in Honor of Urban Tigner Holmes, eds John Mahoney and Jon Esten Keller (Chapel Hill, NC, 1966), 169-95.
4 Palmer, 'The Historical Context', 259-60.
5 Howard Schless, 'A Dating for the Book of the Duchess: Line 1314', The Chaucer Review, 19 (1985), 272-6.
6 'A marriage between my beloved cousin and the-Duke of Lancaster, your son, in Palmer,' The Historical Context, 253.
7 Schless, 'Dating' 274. All Chaucer quotations are taken from The Riverside Chaucer, ed. Larry D. Benson (Boston, MA, 1987).
8 Book of the Duchess, 1311-13. Possibly significant is the fact that reference is made to Lancaster and Richmond, but not to Castile.
9 Sydney Armitage-Smith, John of Gaunt (London, 1904, reprinted 1964), 202-3.
10 Beryl Rowland, 'Chaucer's Duchess and Chess', Florilegium, 16 (1999), 42 and Kathryn L. Lynch, Chaucer's Philosophical Visions (Cambridge, 2000), 31. Robert A. Watson confidently dates the poem to 1369: Watson, 'Dialogue and Invention in the Book of the Duchess', Modem Philology, 98 (2001), 545. Jenny Adams argues that the poem must have been written before 1372, although without explanation: Adams, 'Pawn Takes Knight's Queen: Playing with Chess in the Book of the Duchess', The Chaucer Review, 34 (1999), 135. A. J. Minnis, Oxford Guides to Chaucer: The Shorter Poems (Oxford, 1995), 79-80 notes in parentheses that the reference to Richmond confirms that the poem was written before Gaunt lost that earldom. Three exceptions are Phillipa Hardman, who dates the poem to 1374, Nicole Lassahn, who suggests that the poem could have been written as late as after 1374, and William A. Quinn, who bases a date 'closer to 1374 than 1368' upon a reading of the poem as an attempt to 'put the Duke's grief to rest, not just to eulogise the memory of the dead Duchess': Hardman, 'The Book of the Duchess as a Memorial Monument', The Chaucer Review, 28 (1994), 205-15, Lassahn, 'Literary Representations of History in Fourteenth Century England: Shared Technique and Divergent Practice in Chaucer and Langland', Essays in Medieval Studies, 17 (2000), 51-2, and Quinn, 'Medieval Dream Visions: Chaucer's Book of the Duches', in Readings in Medieval Texts: Interpreting Old and Middle English Literature, eds David Johnson and Elaine Treharne (Oxford, 2005), 323-36. Larry D. Benson and Colin Wilcockson date the poem to before 1372-see Riverside Chaucer, xxix and 329, respectively.
11 'Twenty shillings for personal expenses', reproduced in Martin M. Crow and Clair C. Olson, Chaucer Lifer-Records (Oxford, 1966), 29.
12 Reproduced in ibid., 128.
13 Donald R. Howard, Chaucer: His Life, His Works, His World (New York, 1987), 122.
14 Crow and Olson, Life-Records, 13-18.
15 The first record of an annuity paid to Philippa Chaucer by John of Gaunt is from 30 August 1372, reproduced in Crow and Olson, Life-Records, 85-6.
16 The grant for the annuity is reproduced in Crow and Olson, Life-Records;, 271.1
17 Paul Strohm, Social Chaucer (Cambridge, MA, 1989), 34-6.
18 Armitage-Smith, John of Gaunt., 390-1.
19 Ibid., John of Gaunt, 389 and 462-3. For support, Armitage-Smith quotes the Monk of Evesham, who affirms the relationship began during Gaunt's marriage to Constance, and Froissart, who tells us that the relationship began around the time of the death of Katherine's first husband, Hugh Swynford, who died in 1372. The year of John Beaufort's birth is not given in surviving records. Anthony Goodman accepts Armitage-Smith's dating in his biography, John of Gaunt: The Exercise of Princely Power in Fourteenth-Century Europe (London, 1992), 365. G. L. Harriss suggests that he was born in 1372 'before she [Katherine] was widowed' see Harriss, Cardinal Beaufort: A, Study of Lancastrian Ascendancy and Decline (Oxford, 1988); in his 2004 entry for the Oxford Dictionary of National Biography; Harriss suggests c.1371 for Beaufort's birth. All of these dates suggest that the liaison began at some point in 1371. Due to length restrictions and a paucity of documents, I shall not explore whether the liaison began before or after Hugh Swynford's death nor on how this might have further complicated public perceptions of Gaunt's relationships with Katherine and Blanche.
20 Paul Strohm, Social Chaucer (Cambridge, MA, 1989), 55.
21 For example, Nicole Lassahn suggests that Gaunt was not the poem's audience, and that it 'constitutes a literary discussion: a reworking of the ways in which this representation [of poet-patron] has been used before', 'Literary Representations', 52. In her assessment, the poem appears to be 'engaged and yet still... wholly theoretical', 'Literary Representations', 58.
22 In his testament, dated 3 February 1398, Gaunt asked to be buried next to Blanche, and he was buried there in 1399. The testament is reproduced in Armitage-Smith, John of Gaunt, 420. However, one must consider the protestations made by Derek Pearsall, The Life of Geoffrey Chaucer (Oxford, 1992), 89-91, who notes that it was conventional to be buried beside one's first wife, and thus we cannot take Gaunt's will to signify any emotional connection between Gaunt and Blanche.
23 David Carlson, Chaucer1 Jobs (New York, 2004), 34-40.
24 Arthur W Bahr, 'The Rhetorical Construction of Narrator and Narrative in Chaucer's the Book of the Duchess', The Chaucer Review, 35 (2000), 43.
25 Joerg O. Fichte, 'The Book of the Duchess -- A Consolation?', Studia Neophilologica, 45 (1973), 53-67.
26 Ibid., 59-60.
27 Book of the Duchess, 547-54.
28 Ibid., 1309-10.
29 Minnis, The Shorter Poems, 84.
30 Phillip C. Boardman, 'Courtly Language and the Strategy of Consolation in the Book of the Duchess' ELH, 44 (1977), 572.
31 Steven Davis, 'Guillaume de Machaut, Chaucer's Book of the Duchess, and the Chaucer Tradition', The Chaucer Review, 36 (2002), 396.
32 Although we do not have written record of when the affair was made public, 'the Duke's gifts to Katherine... begin to become significant' around 1371-2 (Armitage-Smith, John of Gaunt, 463).
33 After, the Black Knight describes how his lady syt so in myn herte That, by my trouthe, nolde noght For al thys world out of my thoght Leve my lady; noo, trewely! the narrator warns him that: Me thynketh ye have such a chaunce As shryfte wythoute repentaunce. (Book of the Duchess, pp. 1108-14).
34 The first mention of Philippa Chaucer as a member of Gaunt's household on'30 August 1372 furthers the possibility of an intimate social, but not political, relationship between Gaunt, his mistress, her sister Philippa, and, by extension, her husband Geoffrey Chaucer. At the very least, by this date Katherine, Swynford would be pregnant with Gaunt's child, John Beaufort -- thus the relationship between Philippa Chaucer and Gaunt was confirmed in blood, and quite, possibly this was, Gaunt's motivation for including her in his household.
35 As Armitage-Smith notes, 'The Duke of Lancaster was never peculiarly sensitive to public opinion' ('John of Gaunt', 121). His unpopularity amongst the bourgeois and aristocrats peaked for the first time in 1372 as a result of his military failures in Aquitaine-see ibid., 121-2.
36 Davis, 'Guillaume de Machaut', 392.
37 A. C. Spearing, Textual Subjectivity: The Encoding of Subjectivity in Medieval Narratives and Lyrics (Oxford, 2005), 17-31.
38 Ibid., 156.
39 Ibid., 157.
40 Another scholar who has attempted to untangle Chaucer's complex narrative voice suggests that we 'cannot look at a medieval author with the eighteenth-century microscopic eye, and modern strictures on logic or taboos against anachronism would mean little to him [Chaucer]. It seems harsh to argue, for instance, that a naive Chaucerian pose does not exist in the Prologue to the Canterbury Tales because the voice of such a narrator is used only intermittently in the tales', Thomas J. Garbáty, 'The Degradation of Chaucer's "Geffrey"', PMLA, 89 (1974), 98.
41 Book of the Duchess, 507-13.
42 Ibid., 30-40.
43 See R. M. Lumiansky, 'The Bereaved Narrator in Chaucer's The Book of the Duches', Tennessee Studies in Literature, 9 (1959), 5-17 for a classic description of the narrator as lovesick; for the head melancholy theory, which draws heavily upon Robert Burtons Anatomy of Melancholy, see John M. Hill, 'The Book of the Duchess, Melancholy and That Eight-Year Sickness', Chaucer Review, 9 (1974), 35-50.
44 Book of the Duchess, 1126-36.
45 Ibid., 514-21.
46 Ibid., 458-60.
47 Ibid., 451-2.
48 The anagram is found in Fonteinne Amoureuse, 44-51. All quotes from Machaut are from Oeuvres de Guillaume de Machaut, ed. Ernest Hoepffner, 3 vols (Paris, 1908-1914). I would like to thank Grover Furr for taking on the Herculean task of scanning and making available all three volumes at my request.
49 'But never in a day of my life have I seen a manner so fine in either a man or woman, and his body was flwaless, for, it was grand, tall and up-right, well fashioned in all ways, noble, handsome, fine, young, and graceful' Fonteinne Amoureuse, 1101-7.
50 'Then as if I had a fever, for I am more cowardly than a hare, and I trembled and cowered for fear. I lay there in such a sorry state that, if anyone would try to take my doublet, belt, or shirt, I would put up no defence at all; No, by God! Anyone who would see my body would not see it move ibid., 89-100.