111 65 2MB
English Pages 254 Year 2023
Gender and Position-Taking in Henrician Verse
Gender and Position-Taking in Henrician Verse Tradition, Translation, and Transcription
Rebecca M. Quoss-Moore
Amsterdam University Press
Cover illustration: Mary Magdalene Writing, Mistrz Półfigur Kobiecych (fl. 1500–1550); Muzeum Narodowe w Krakowie. Cover design: Coördesign, Leiden Lay-out: Crius Group, Hulshout isbn 978 94 6372 353 4 e-isbn 978 90 4855 157 6 doi 10.5117/9789463723534 nur 685 © R.M. Quoss-Moore / Amsterdam University Press B.V., Amsterdam 2023 All rights reserved. Without limiting the rights under copyright reserved above, no part of this book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the written permission of both the copyright owner and the author of the book.
Table of Contents
Acknowledgements 7 Introduction 9 1. Early Verse Position-Taking in the Henrician Court A. The King’s Courtly Verse B. Skelton and Early Henrician Courtiers’ Position-Taking
29 33 45
2. Traditions of Resistance and Verse Position-Taking A. Wyatt and Contextual Position-Taking B. Surrey and Self-Authorization
61 68 71
3. Translation and the Position-Taking Verse Tradition A. Titles, Tottel’s, and Verse in Context B. Language Choices and Communal Position-Taking Practice
83 86 96
4. Men’s and Women’s Approaches to Translation and Authority in the Late Henrician Court 113 A. Surrey’s Aeneid 115 B. Wyatt’s Psalms 126 C. Parr and Hybridized Position-Taking 138 D. Two Translations into Prose by Early Modern Englishwomen 144 Elizabeth’s Miroir146 Jane Lumley’s Iphigeneia153 5. Transcription as Translation: Writing the Language of Manuscript Poetry A. Context and ‘Correctness’ in Manuscript Transcription B. Responsive Reading and Composition C. Forms and Dialogues
163 171 183 189
6. Resistance and Unity in the Douglas-Howard Exchange A. The Epistolary Exchange B. Contextual Affect and Effect
201 206 223
Conclusion 235
Bibliography 241 Index 251
Acknowledgements My thanks go out to the faculty and staff of the University of Central Oklahoma who support my work in innumerable ways, particularly to those of the Department of English, Max Chambers Library, and the College of Liberal Arts. My thanks are also due to all of my students over the last decade, who always present opportunities to think about old texts in new ways. This work has benefitted enormously from my colleagues in the field, particularly from the meetings of the Renaissance Society of America and the Shakespeare Association of America. I want to offer special thanks to Raymond G. Siemens for his comments on an early form of the final chapters, here, and to Jennifer Lodine-Chaffey for her reading of a later version of the fifth chapter. Jonathan P. Lamb was remarkably generous with his time and energy, reading an entire draft of the manuscript, talking me through my own argument, and, invaluably, helping me rethink a particularly sticky term and, partly as a result, frame my goals much more clearly. I am also especially grateful to the editorial staff of Amsterdam University Press, most particularly to Erika Gaffney and to the peer reviewers who guided so much work on earlier versions of the manuscript. I am also grateful to the faculty and staff of the Department of English at the University of Arkansas, particularly to Dorothy Stephens, John DuVal, and Joseph Candido for their support and guidance. I also owe a great deal to Terrence Tucker, Amy Witherbee, and Danny Sexton, all of whom were important influences on the formation of my approaches to literary study and academic work. My thanks, also, go to the staff at David W. Mullins Library and to the staff of the British Library, whose labor enabled this and so many other projects. I am indebted to innumerable family, friends, and partners who have sustained and supported me since I began my work in academia. My particular thanks to my immediate family: Donald and Karen Moore, Jerry and Eden Moore, and my nieces Whitney, Maddie, and Ainsley. My friends and colleagues Grant Bain and Jenn and Lance Mallette are enduring sources of calm and joy. I really feel that I would never have gotten through graduate school, let alone this project or my first years in a tenure-track position, without my much-beloved husky, Gambit, who really understood the place of a good, long howl in any project. And, above all, thanks to my spouse and partner, Justin Quoss-Moore, who makes my work possible and whose unwavering support and understanding never stop feeling like a revelation.
8
Gender and Position-Taking in Henrician Verse
As I hope this work establishes, I believe all writing is, fundamentally, a collaborative and cooperative project. I owe something to every scholar cited here, to dozens if not hundreds of unnamed scholars whose work has shaped our field and my thinking, and, very materially, to everyone who has contributed to resources like Early English Books Online, the Social Edition of the Devonshire Manuscript, and the immeasurable gift of Interlibrary Loan. As in the texts we study, so in our scholarship: we are all writing together, and I am grateful to be part of this project.
Introduction Abstract: This introduction outlines the core goals of the project, which returns to the Henrician texts that underpinned much New Historicist work and re-reads these through the intervening scholarship on women’s writing, manuscript and print cultures, and revisions and challenges to that New Historicist framework. The project re-reads women into the Henrician canon to further illuminate the forces of canon formation that contributed to those women’s erasure and to restore more accurate depictions of a shared social system of verse position-taking. This restoration enables an exploration of how collections like Tottel’s—and their later influence—contributed to a misapprehension of single authorship, an initial diminution of Henrician verse’s political signif icance, and a damagingly inaccurate masculinizing of literary history. Keywords: Henrician translation; early modern women’s writing; Tudor verse transcription; Devonshire Manuscript; gendered canon formation; courtly love lyric
1536 was something of a landmark year for political scandal in England even considering the tumultuous history of the court of Henry VIII. One queen lost her head, another took her place, and Henry Fitzroy, the King’s only acknowledged son, died suddenly, within months of setting up house with his new wife. This context was nearly fatal for Margaret Douglas, niece to Henry VIII, who chose a particularly ill-omened time to contract an unapproved marriage to another of the King’s relatives, Thomas Howard.1 Given the unstable state of the dynasty, Henry reacted viciously to his niece’s transgression, imprisoning both Douglas and Howard. Howard fell 1 Howard was the brother of the Duke of Norfolk and uncle to the recently executed Anne Boleyn; to Henry Howard, Earl of Surrey; and to Mary Fitzroy, wife of the recently-departed Henry Fitzroy. See Heale, Devonshire; Irish, ‘Gender and Politics’; and the Oxford DNB.
Quoss-Moore, R.M., Gender and Position-Taking in Henrician Verse: Tradition, Translation, and Transcription. Amsterdam: Amsterdam University Press, 2023 doi 10.5117/9789463723534_intro
10
Gender and Position-Taking in Henrician Verse
under an act of attainder that simultaneously established as treason the act of which he was accused: marriage to a member of the King’s family without the King’s consent. Though contemporary observers suspected that the King intended clemency, his intentions were mooted by Howard’s death in the Tower from ague. Certainly, the outline of events has all the elements of a storybook tragedy: young lovers imprisoned by an embittered king who twists the law to suit his temper. What sets this tragedy apart, though, is the role these particular young lovers had in the production of poetry at court. Douglas and Howard were major contributors to, if not the creators of, the Devonshire Manuscript.2 Helen Baron and Elizabeth Heale are among the scholars who suggest that Howard was, in fact, the original owner and circulator of the manuscript, which perhaps passed to Douglas upon his death.3 In that manuscript, a distinctive hand transcribes, along with verse epistles between the lovers, reconstructed versions of works both by Chaucer and misattributed to him by the 1532 Thynne edition of his works, as well as translations from Alain Chartier’s La Belle Dame sans Merci. The placement of the verses and their content, as Baron, Heale, and Paul G. Remley all suggest, strongly reflect a connection to the verse epistles widely agreed upon as a product of the Douglas-Howard affair. 4 Within this context, the changes made by the writer or writers, both in translation and transcription, become particularly striking. Through these verses, we can examine fruitfully clear examples of the uses Henry’s courtiers made of transcription, translation, and tradition as systems for coding and preserving sociopolitical position-taking. At times, the changes made in the transcriptions are fairly isolated. In a selection from The Remedy of Love, a text attributed to Chaucer in the 1532 Thynne, the Devonshire transcriber changes a single line that significantly alters the theme of the piece. In the Thynne edition, the speaker postulates that if the entire world were turned to writing materials, still ‘The cursydnesse yet and disceyte of women / Coude not be shewed by the meane of penne’.5 The transcription in the Devonshire, though, reads as follows: yff all the erthe were parchment scrybable spedy for the hande /6 and all maner wode 2 MS BL Add. 17492. 3 Heale, ed., Devonshire, and Baron, ‘Fitzroy’s Hand’. 4 Remley, ‘Mary Shelton’. 5 Thynne, ed., Workes of Geffray Chaucer, fol. ccc.lxvi (v). 6 I adopt the slashes used midline as a common typographical interpretation for a similar mark in the manuscript. While my transcriptions are conf irmed by my own work with the
Introduc tion
11
were hewed and proporcyoned to pennes able al water ynke / in damme or in flode euery man beyng a parfyte scribe & goode The faythfulnes yet and prayse of women cowde not be shewyd by the meane off penne7
As most critics who have worked with this section of the Devonshire, including Heale and Bradley J. Irish, have remarked, the effect of the change of the penultimate line is striking—altering the complaint from a misogynist tradition to align instead with the medieval tradition of the defense of women. Further, as Irish goes on to discuss, ‘the line shows [the writer] clearly read with enough active interest to imagine the radical altering of textual meaning’.8 The strategies this reader/writer uses reflect how Henry’s courtiers, as a group, read and write: with an eye towards the adaptability of texts and voices to accommodate new ideas and to memorialize new situations. Indeed, the exact situation of this poem, as for others in the manuscript, renders the question of the precise writer less important. Whether the writer is, as Heale would suggest, Howard himself, or, as Remley argues, Mary Shelton—another active contributor to the manuscript, and one whose work with the text lasted several years longer than Howard’s—the defense of women gains poignancy from its situation near the epistle verses exchanged between Howard and Douglas. The writer chooses to place this transcription alongside those prison exchanges, orienting the reader to understand the context of the defense of women as also a defense of the woman involved in those verses, whether as writer, transcriber, or subject. The transcription of translated lines from La Bella Dame sans Merci which directly follows the above transcription may seem to depart suddenly in an entirely other direction, but again the context of the lines becomes key to a useful interpretation. The lines open with an apparent condemnation of a disdainful lover: manuscript in 2015, they are always indebted to the excellent work of the contributors to the Social Edition, a collaborative digital project that provides facsimiles and transcriptions of the entire manuscript and thorough references to the distinguishing features of each hand. My transcript conventions preserve original spelling, expand contractions as indicated by italics, and indicate lines struck through in the original as accurately as possible. ^…^ is used to indicate writing included above the rest of a line, though not necessarily superscript, while […] is used to indicate uncertain transcription. 7 MS BL Add. 17492, 90r, ‘yff all the erthe’, italics mine. 8 Irish, ‘Gender and Politics’, 103.
12
Gender and Position-Taking in Henrician Verse
O marble herte / and yet more ^harde^ perde wych mercy may not perce for no labor more stronge to bowe than ys a myghty tree what avanay avayleth yow to shewe so great rygor pleasyth ^yt^ yow more to se me dye thys hour before yowr [yowr] eyen for yowr dysporte and play than for to shewe some comforte and socour to respyte death / wych chaseth me alway9
In their original context, these lines seem a more appropriate match for the unaltered version of The Remedy. However, given their placement in the manuscript and the apparent date of transcription, Irish suggests the applicability of these verses to Henry VIII, rather than as a sudden departure in tone from defense of women to marked misogyny.10 This analysis is entirely logical in the context of a court system that often understood such love lyrics as political critiques, and the piece’s divorce from its original title emphasizes the applicability of the appeal to the cruel mistress as instead an appeal to a despotic tyrant. In this approach, the ‘death’ which chases the speaker, and the ‘comforte and socour’ for which the speaker pleads, lose their metaphorical sense. Instead, the speaker addresses a figure who holds their courtiers’ lives at pleasure as well as, in this case, their hearts. The King controls the outcome of the love match, encompassing the metaphor of the original love lyric, and controls his subject’s life, making the hyperbole of the translated poem instead immediate and literal. This sense of the address is emphasized by the context within which the translation is placed—both historically and within the manuscript. These selections highlight the importance of the basic structures through which I aim to interpret the poetry of Henry’s court. First, though nearly acting as a translation, the lines from Chartier nonetheless update La Belle Dame within a markedly different framework. Second, the placement of each transcription suggests the contextual importance that contemporary writers ascribed to verse, as well as the interpretive freedoms with which they felt comfortable. Finally, the works together function within the traditions of courtly love, of medieval defenses of women, and of verse as an outlet for political position-taking. I appropriate the term ‘position-taking’ from
9 MS BL Add. 17492, 90r, ‘O marble herte’. 10 Irish, ‘Gender and Politics’, 105.
Introduc tion
13
Bourdieu for this analysis,11 as a way of adapting Greenblatt’s concept of ‘self-fashioning’ to offer a term that more thoroughly integrates the concepts of communal verse production and its inherently social and negotiated nature. The poets and their poems reference and take part in the systems they critique, and they depend on their audience’s understanding of those systems to create meaning. I intend this initial engagement with verse position-taking in the Devonshire to demonstrate the core structure and goals of this work. First, and most simply, I want to return to the Henrician texts that underpinned so much New Historicist work and re-read these through the intervening scholarship on women’s writing, on the relationship between manuscript and print cultures, and on revisions and challenges to that New Historicist framework. As part of that new critical framework—and particularly crucially—I want to re-read women into the Henrician canon. That re-reading functions in two directions: to further illuminate the forces of canon formation that contributed to those women’s erasure and to restore more accurate depictions of courtly writing and reading networks. To offer one application as a sort of definition by example: when we restore Margaret Douglas, Mary Shelton, and Mary Fitzroy to their place in manuscript circulation and creation, we can see that ‘women’s writing’ was integral to the production that gave us those works we now call Thomas Wyatt’s or Henry Howard’s. Simultaneously, then, this means that such production was not women’s writing—or, more precisely, that women’s writing and men’s writing are not two separate things. We can better reconstruct the system, and so we can interrogate new interpretations of its products: particularly, we can better understand how those products participated in a shared social system of position-taking. When we turn to the transfer to public, print culture, we better see how collections like Tottel’s—and their influence on the modern canon—contributed to a misapprehension of single authorship, an initial diminution of Henrician verse’s political significance, and a damagingly 11 Bourdieu’s term also integrates the extent to which position-takings are relative and defined by the social spaces within which they occur (and the options understood as available within those spaces, by both producer and consumer): The space of literary or artistic position-takings, i.e. the structured set of the manifestations of the social agents involved in the field […] is inseparable from the space of literary or artistic positions defined by possession of a determinate quantity of specific capital (recognition) and, at the same time, by occupation of a determinate position in the structure of the distribution of this specific capital. The literary or artistic field is a field of forces, but it is also a field of struggles tending to transform or conserve this field of forces. (Field of Cultural Production, 3)
14
Gender and Position-Taking in Henrician Verse
inaccurate masculinizing of literary history that persisted even as that verse’s political significance was restored. David Norbrook emphasizes the importance of starting from the precept that early modern artists were aware of, and resistant to, the problems of too much or too strict a structure; in response, ‘they developed elaborate strategies to try to preserve a degree of independence for their writing’.12 Repression is often the ground for creativity; drawing on Foucault, Butler articulates clearly the essential claim that ‘the culturally contradictory enterprise of the mechanism of repression is prohibitive and generative at once’.13 The response of the aristocratic court poets to repressive measures was shaped by their understanding of their social roles. In keeping with humanist tradition, educated nobles ‘believed themselves to be educated for public service, believed that they could persuade princes, in Church and state, to reform’.14 As a result, learned aristocrats ‘used their writings, and the various forms of license that their culture allowed them […] to influence government policy through the medium of eloquence’.15 In fact, Henry’s early approach to government and privileging of humanist education had only reinforced these social tropes, and ‘the habit of speaking boldly on issues of principle and practice and the capacity to […] apply biblical and classical examples to illuminate contemporary politics were ingrained in elite English culture’.16 The legislation of the 1530s suddenly made different claims. The abstract conflict of humanist thought crystallized in a real conflict between self-interest and social interest. Suddenly, law restricted the ability of the court poets to shape social change and offer critique, at the time that social codes and long-engrained practice seemed most to demand that they exercise precisely those abilities. While men’s roles in public life may offer allowances for the previous framing of this tension as an especial concern of courtly men, the Devonshire Manuscript shows that court women considered critique a key element of their identities as courtiers, as well. In restoring women’s manuscript production to its central role in court verse work, we can realign our own understanding of position-taking verse systems at the Henrician court and better understand the successes and failures of the court poets who worked within these systems. 12 Norbrook, Poetry and Politics, 5. 13 Butler, Gender Trouble, 126. 14 Brigden, New Worlds, 4. 15 Walker, Writing Under Tyranny, 15. 16 Walker, 23.
Introduc tion
15
As we will see in Chapter One, Henry VIII himself was one of the most sophisticated practitioners of this kind of manipulation. However, as he gained power, he moved out of the more coded strategies of verse production; simultaneously, courtiers deployed his own methods against him. Royal energies are frequently at odds with the efforts of disgruntled courtiers like Skelton and Douglas, but they also provide fuel for those energies. This negotiated intersection, of royal strategy and courtier strategy, frames these poets’ efforts in a new, more comprehensive context. The conditions of repression imposed by the patron system and by royal authority create the conditions necessary for the artist to imagine a privileging of their work. This creation of a privileged creative space is designed to create a protectiveness of that space in the otherwise oppositional subject: the court poets’ positions are also necessarily entwined with the social structure they critique. Such a perspective does not lessen the potential resistance or even subversion of a work; the poet claims that the divisions or circumstances that exist around the court are not ‘natural’, and the push for a return to the natural order necessarily calls to destabilize the artificial order which has taken its place. Through appeals to tradition, authority, and community, the courtly poets sought to establish their own counternarratives, not least through the creation of interpretive possibility in their verse. Each strategy considered here also establishes each author’s work within a larger continuity; this continuity serves to give verses greater context, greater interpretive potential, and greater authority for their contemporary readers. The details of courtly context not only restore the important roles of courtly women, but also further allow us to analyze the energies of these poems as moments of political position-taking—whether resistant, critical, radical, or conservative. The impact of these varied and often conflicting modes and goals must be recovered in a complex negotiation of simultaneously separating and relating political, personal, and poetic strategies as understood and used by the Henrician court poet. Poetic position-taking in Henrician England is cultural and political position-taking, partially because of the inherent link between poetry and the established social structure; this is political work both of a highly particular and of several broad types. Specifically, this cultural positioning exists within and because of a structure that necessarily mixes the personal, cultural, and political: the court poet’s home and primary household is often also the seat of cultural and political power. Generally, though, the complications of this social-structural position for the poet create very diverse responses, determined by each poet’s goals, but then filtered through their strategies, social position, and audience.
16
Gender and Position-Taking in Henrician Verse
Translation, transcription, and poetic tradition then emerge as methods that not only allow the poets to distance themselves from their content, but also even explicitly alert the audience to particularly controversial potential interpretations. After all, the sophisticated audience for court poetry already understood the separation between poem and poet; if poets sometimes felt the need to emphasize or widen that separation, they alerted the audience to some difference in the content of such poems. Such authors aimed to cloak their dissent with some reasonable deniability, using genre, the distancing claim of ‘translation’, and doubled language to protect poetry of protest. The atmosphere of court necessitates this cloaking. The charged atmosphere in which these authors lived has been thoroughly explored by literary scholars and historians alike, but, integrating the considerations of position-taking as highly contextual, my work aims to examine that atmosphere as a tool used by writers, rather than solely as a limiting or provoking factor. Courtly milieu serves, in its way, as a context for the courtiers in the same way that translation, transcription, and tradition, variously, serve as context for their works: lending legitimacy, offering a vocabulary evocative of the genre, and cloaking, protecting. Behavioral codes and verse coding intersect. Working in the highly prescriptive social ‘language’ of court courtesy and custom, Henrician courtiers learned to use courtly behavior to protect themselves; they were also able to use these codes of behavior at times to justify, defend, or hide intentions that ran counter to Henry’s own. In this re-examination of courtly verse coding, and particularly in my concern with the re-inscription of women’s verse work, I hope to offer some responses to particular calls for reconsiderations of manuscript culture and practice. Victoria E. Burke has articulated the need for ‘close attention to material characteristics, including handwriting and layout, as well as to content, [which] can sometimes reveal patterns in these apparently random collections’.17 Speaking to a different element of manuscript production, Jason Powell calls for ‘a balanced sensitivity to the possible uses for anonymity […] alongside a discussion of likely authorship [to represent] the richness of this manuscript, its social environments, and the community of authors, compilers, and scribes who participated in its production’.18 And Deborah Solomon speaks particularly directly to my concerns, here, in writing If we agree that context affects content, that every textual version in the messy history of textual transmission, every ‘misreading’, as McKenzie 17 Burke, ‘Materiality and Form’, 219. 18 Powell, ‘Marginalia’, 12.
Introduc tion
17
would say, represents a cultural artifact rich in significance, why not read the Devonshire MS for its potential literary value rather than simply mine it for information? […] While we have long delighted in the effect linguistic forms have on our interpretation of the lyric, we have yet to do justice to the effect material forms have on our experience of poems.19
In engaging the Devonshire, I have focused on answering these calls for greater attention to the work of anonymity, of placement and patterns, and of the significance of linguistic, textual, and structural choices. While I work both backward and forward from that 1536 moment to consider the sources and the later manifestations and alterations of the themes I identify here, the Devonshire was where I first understood these strategies as a coded system of sociopolitical position-taking—and first became concerned with re-inscribing the ways women wrote in this system to more fully understand Henrician poetic production. When I reference this verse coding system, I do not mean an exact and precise code—manuscript reality is necessarily messier than that. We are, though, familiar and comfortable with the idea that love poetry of the early modern period frequently offered coded commentary on political and social events. The influence of Stephen Greenblatt’s Renaissance Self-Fashioning is inescapable, and I join some of those fundamental claims to Jeff Dolven’s work on re-fashioning to consider the alteration of poetry as one form of poetic representation or creation of self and self-interest.20 I want to continue our work to update those frameworks for understanding political verse with our enriched approaches to manuscript studies, with a greater consideration of materiality and form as active poetic work, and with a more equitable consideration of gendered forces in verse production. Arthur F. Marotti’s foundational ‘Love Is Not Love’ further established our understanding of many of these works as poetry which coded men’s ambition and grievances; the updated considerations outlined here allow us to demasculinize that understanding. The foundational work from Greenblatt and Marotti also establishes our sense of ‘coding’ as a useful term for verse analysis: the verse is both part of a social ‘code’, in the sense of a system of behaviors that reflects one’s cultural position and embeddedness, and part of a system of ‘coding’ positions, in the sense of obfuscating through a system that is translatable by those with the correct knowledge. Marotti had himself, as echoed in Burke and Solomon, pushed for further engagement with the material 19 Solomon, ‘Representations’, 682. Solomon also cites D. F. McKenzie in this argument. 20 Greenblatt, Renaissance Self-Fashioning; Dolven, ‘Reading Wyatt’.
18
Gender and Position-Taking in Henrician Verse
specificity of manuscripts and with materiality’s impacts on both social worlds of production and social coding;21 women’s work in the Devonshire and in courtly writing more generally is one key piece of that material specificity. In reconsidering the composition of those social worlds—and so of the groups that used such social coding—we gain an access point for restoring women’s places in these scribal communities. Manuscript studies has long understood the importance of communal production in these manuscript miscellanies. Alongside Harold Love,22 Mary Hobbs makes some of the most influential arguments for understanding manuscripts as whole documents—that is, for considering how the pieces all come together to influence readers’ experiences, a consideration I unite to Burke’s emphasis on patterns and materiality and to Powell’s arguments about anonymity. H.R. Woudhuysen, similarly concerned with the concept of scribal community, builds from Love and Hobbs and especially considers questions of how manuscripts reach the intended audience and of the broader complications of manuscript production;23 these considerations undergird my arguments, throughout, about the availability of wide ranges of interpretations and about the work writers and editors offer to control or, in some cases, promote the possibility of misunderstanding, misinterpretation, or obfuscation. Peter Beal anchors similar concerns around audience in a particular consideration of the unique nature of each manuscript, encouraging us to consider the implications of the intersection of replication and singularity,24 a point I particularly apply when considering the changes a verse undergoes between manuscript and print versions. And Steven W. May’s consideration of manuscript production traces some points especially salient to the works under consideration, here, as he links work with scribal communities to these kinds of transfer from manuscript forms to print—and to Tottel’s, specifically.25 These works on manuscript studies influence my primary approach to the works under consideration, here, where I unite material concerns in the Devonshire—including precise spelling choices, space on the page, and placement—with the communal concerns of this group of courtiers. Particularly, I follow Solomon’s call to reconsider the work of textual 21 Marotti, ‘“Love Is Not Love”’. 22 Love, Scribal Production; Hobbs, Verse Miscellany Manuscripts. 23 Woudhuysen, Circulation of Manuscripts. 24 Beal, In Praise of Scribes. 25 May, ‘Popularizing Courtly Poetry’. Our approaches differ in that May, broadly, sees most alterations as just mistakes and misreadings; I think at least in the Devonshire and its closest ilk and, to move the case to a different situation, Tottel’s, this is not necessarily the case.
Introduc tion
19
transmissions and the interpretive potentials of apparent misreadings.26 The structural categories, here, of tradition, translation, and transcription, are each essentially about the transmissions of verse, images, ideas, or approaches. The choices writers make in those transmissions are their primary method of textual production. That is, if poetry is a form of selffashioning and position-taking in the Henrician court, and if most poetry in manuscripts has a stronger relationship to transcription or translation than to modern ideas of ‘original’ composition, it then makes sense that the scribal community adopts norms and methods for systems of production that make the translated or transcribed verse work for their larger products. More precisely, given these constitutive relationships, we need to read every aspect of the manuscript as a form of self-fashioning and position-taking, and specifically as part of a system of position-taking to which women and men made equal contributions. Outlining some of the functions of that adaptive compositional work in the Devonshire offers new perspectives on the position-taking and -making functions of other courtly verse. Raymond G. Siemens’s and Peter C. Herman’s arguments on Henry’s early poetic work helped me to understand later courtiers’ strategies as specifically a response to the King, pitched very intentionally to his own early work.27 The long history of criticism understanding Skelton’s and Wyatt’s work as subversive or resistant— including, beyond Greenblatt and Marotti, critics like David R. Carlson, Jane Griffiths, W. Scott Blanchard, and Powell—takes on new dimensions when we join those understandings to this specific manuscript genealogy, highlighting this work as communally informed rather than reflective of more individualized political or aesthetic positioning.28 The transfer of those communal products through different publics can be accessed through comparative work by Jonathan Gibson and by Solomon,29 as well as through the work on Tottel’s by May, Megan Heffernan, and Christopher J. Warner, all of which suggest the efficacy of considering the manuscript work of the Devonshire and the print decisions in Tottel’s alongside one another.30 This 26 Solomon, ‘Representations’. 27 Herman and Siemens, ‘Poetry of Politics’; Herman, Royal Poetrie; and Siemens, ‘Henry VIII as Writer’. 28 Greenblatt, Renaissance Self-Fashioning; Marotti, ‘Love Is Not Love’ and English Renaissance Lyric; Carlson, ‘Revels and Erudition’ and ‘Henrician Courtier’; Griff iths, Poetic Authority; Blanchard, ‘Voice of the Mob’; and Powell, ‘Plainness and Dissimulation’. 29 Gibson, ‘Miscellanies’, and Solomon, ‘Representations’. 30 May, ‘Popularizing Courtly Poetry’; Heffernan, Making the Miscellany; and Warner, Making and Marketing.
20
Gender and Position-Taking in Henrician Verse
consideration allows us to access the ways writers work to reach different audiences and how audiences do or do not read particular types of coding and verse work. These critics establish that verse by, for example, Wyatt or Surrey, does different work in different places; we can use that distinction to more clearly map the work being done on the manuscript page and the transfer or loss of different position-taking strategies between forms. In outlining those strategies, I have drawn extensively on the vibrant conversation about the Devonshire itself. The many contributors to the Social Edition have created a rich repository of critical lenses, historical and biographical information, and details of material and mechanical considerations. Something of Heale’s, Remley’s, and Irish’s influence has already been indicated by the opening analyses, but Christopher Shirley’s work, particularly, also informs some of my directions here in consideration of gender.31 Beginning the work with the Devonshire emphasizes one of the key opportunities for further contextualization. Much of the extant work on Henrician poetry emphasizes methods of interpreting poetry as masculine courtly performance. In foundational work like Greenblatt’s, there is often extensive engagement with the idea of masculinity grating against submission, duplicity, and so on. However, the Devonshire shows us these themes are actually quite present in women’s work, as well. Studies of women’s manuscript work, particularly those by Margaret J. M. Ezell, Burke and Gibson, and Jane Stevenson and Peter Davidson, have made invaluable contributions to the field. These influential works on women’s manuscripts, though, largely pick up in the mid-sixteenth century or intentionally (and importantly) focus on work outside of courtly contexts.32 I focus, here, on courtier poets. These are poets who, regardless of gender, consider themselves as possessing a certain amount of political power, and their relationships to power are quite different from those of the authors featured in much of the excellent work on women’s manuscript production in other, more localized contexts.33 Ezell evinces a concern with reconsidering 31 A Social Edition; Heale, Devonshire, ‘Love Lyric’, and ‘Female Voices’; Remley, ‘Mary Shelton’; Irish, ‘Gender and Politics’; and Shirley ‘Reading Gender’. 32 Ezell’s Social Authorship focuses on the extension of manuscript cultures into the seventeenth and eighteenth centuries. Burke and Gibson’s Manuscript Writing includes Heale’s ‘Female Voices’, as well as work on Katherine Parr and Elizabeth (Gibson, ‘Katherine Parr’), but the other chapters of the collection primarily consider later women’s manuscripts. While Stevenson and Davidson’s Women Poets does include entries that date between 1520 and 1550, these entries make up, at most, less than 10 percent of the anthology, where a priority is instead greater representation of women poets throughout the British Isles. 33 White also engages considerations of social networks of women’s writing, but her focus around women’s manuscript transmissions, specifically, is primarily concerned with local networks
21
Introduc tion
whether manuscript culture was feminine—her argument outlines the ways in which that alignment is used in a patronizing sense.34 The corollary that applies to my work, here, is that the Henrician manuscript has not been yet fully considered as a space of women’s work.35 When the manuscript is the highest form, it is treated on masculine terms; once the manuscript site is de-privileged, it becomes feminized. This process of recursive redefinition occurs both historically—print privileges itself and privileges male visibility over women’s access36 —and critically, as Ezell points out when she summarizes previous characterizations of Elizabethan and Jacobean manuscript writing as ‘“aristocratic”, dilettante literature, insignificant in terms of literary history because the texts formed no school’.37 This rather depressingly predictable arc has a corrective at one end of the period in Ezell’s reconsideration; I hope to offer an additional entry into the balance for these earlier decades. *** This work is divided into three sections, each addressing one of the three strategies of versified position-taking that I have used for our system of categorization. The work begins with a section on what may initially seem to be a more, or even overly, general theme in Henrician works: tradition. The first chapter here deals with the tradition of the courtly love lyric, beginning with Henry’s own verse. Henry worked almost exclusively in original compositions, many of which were likely meant to be set to music. This analysis follows Siemens’s and Herman’s established critique of Henry’s self-assertion through verse in the first decade of his reign,38 but also draws outside of the court (‘Women Writers’). Her work on Katherine Parr, though, foregrounds a similar sense of the distinct strategies used by women with marked political power and underpins the approach to Parr in my fourth chapter (‘Literary Collaboration’ and ‘Royal Iconography’). 34 Ezell, Social Authorship, especially 21–44. 35 Heale, for example, characterizes women’s involvement in the Devonshire as ‘both more frustratingly uncertain than many modern critics […] would like, and at the same time more extensive and central than is often recognized’ (‘Female Voices’, 9). 36 Heale considers one consequence of this shift when she outlines that ‘[a]s the early Tudor balet moved from manuscript to print, so it became an almost exclusively male-voiced genre with the female-voiced poems of passion and retaliation largely silenced. In moving into print, the role of women as crucial to the culture and the production of courtly verse disappeared from sight’ (26). 37 Ezell, Social Authorship, 36. 38 See Herman, Royal Poetrie; Herman, ed., Reading Monarchs; Herman, ed., Rethinking the Henrician; Siemens, ‘Henry VIII as Writer’; and Siemens, ed., Lyrics.
22
Gender and Position-Taking in Henrician Verse
on analysis of Henry’s centralization of power as a radical act. Henry’s attempts at centralization created reactive, conservative responses from his courtiers, who then themselves created spaces for new acquisitions of power. Henry and his courtiers wanted the ability to assign power while also wanting to preserve their own power on traditional terms, and drawing on traditional forms and themes to justify their assignments of power helped bolster such claims. The second chapter engages with the manipulation of particular historic traditions when addressing Henry’s reign. Surrey, of course, famously uses the figures of both David and Sardanapalus to critique Henry’s policies, but Surrey is far from isolated in using kings of classical antiquity as allegorical stand-ins for a reigning king. Wyatt’s work provides an earlier example of such interpretation, as do the King’s own material choices in self-presentation. Henry’s choices underline the political verse-making process that takes place when courtiers manipulate his choices of royal precedent so that those precedents become critiques rather than validation or valorization. Essentially, the engagement with tradition seems to work best as a method for the negotiation of power through verse. Courtiers felt that their assumption of power was normal and right, just as Henry felt about his own monarchal powers. Both groups, though, were simultaneously aware of and resistant to the need to put their most radical claims to power in codes that both protected such claims and robbed them of at least some force. Tradition, in this section, is considered as functioning in verse production on similar terms to translation or transcription—as a composing choice that actively modifies extant material to reflect participation in a shared system of position-taking. The second section deals with the related uses of translation. In the third chapter, the focus is on translation of Petrarch and other poets in the courtly love lyric tradition, bridging the intersection of translation and engagement with tradition as position-taking strategies. Though Wyatt and Surrey are the most well-known actors in this genre, the chapter also looks at works preserved in the Devonshire Manuscript. As the chapter emphasizes, the value of translation had been understood for centuries, by its many separate acolytes, as adaptable to engage different messages, rather than as a strict transference of the original meaning from one tongue to another. Further, Henrician poets borrowed from their Continental predecessors an appreciation and understanding of the ability of the courtly love lyric to convey political frustration. By joining translation to similar coding techniques, we can highlight some consistent systems of political response and resistance at play across poetic genres, while the joint analysis
Introduc tion
23
also highlights the ways that coding works differently in translations than through other outlets, with more clarity in its critiques and so more risk. The fourth chapter then shifts attention to translation from classical and ancient sources, engaging translations like Surrey’s Aeneid and Wyatt’s Psalms and considering Katherine Parr’s uses of biblical and political authority. Parr’s work offers a bridge into an examination of the use of translated work by young women in the last years of Henry VIII and in the brief reign of his son Edward, specifically translations from verse into prose by Elizabeth I and Jane Lumley. These works are considered, together, as translations of texts that had already been assigned cultural importance and authority, thus increasing the interpretive cultural framework. Surrey and Wyatt work in verse translations and appear to align their speaking voices with the relatively isolated and individualized leader-subjects of their works. While both poets draw on the communal position-taking practices of translation, they also both engage the potential of more individualized authorities. The question of authority underpins the subsequent engagement with Parr’s work as a hybridized intersection of the coded strategies of verse positiontaking, the alternative interpretive frameworks suggested by prose and print, and the negotiation of meaning between monarchical authorization and courtly, communal position-taking. Considering Parr’s work alongside that of Wyatt and Surrey contextualizes Elizabeth’s and Jane Lumley’s decisions to work away from the verse versions they translate into prose. All three women offer visions of rhetorical authority for their narrators or primary characters, but these visions also appeal both to communal experiences and to established, higher, and masculinized figures of authority. The section on translation emphasizes the interpretive frameworks audiences were expected to bring to their reading and offers us practical demonstrations of how power—inflected by gender, status, and age—is played out on the page. The final section moves into a consideration of the strategies of transcription, where play on the page is the core concern. This section opens with the fifth chapter of the work, devoted to the practice of transcription as practiced in circulated manuscripts of the time. Transcription, as shown in the opening analysis here, often involved studied revision on a scale ranging from particular pronouns to entire stanzas, offering an entirely different effect from that of an original piece. Like the practice of translation, transcription was understood as a method through which courtiers could reimagine a text, making works more immediately resonant with their world. Compared to translation, transcription allowed for both greater flexibility and greater deniability. The imaginative engagement with the text is of a different sort, because the text can be more explicitly rearranged or broken
24
Gender and Position-Taking in Henrician Verse
apart to join with other texts. This chapter examines the products of the Devonshire manuscript as a collaborative project where transcription unites with multiple authorship. The final chapter continues the consideration of transcription, but shifts into a focus on dialogue within manuscript creation. The focus of the chapter is the exchange between Margaret Douglas and Thomas Howard during their imprisonment in the Tower and the process by which that exchange came to be preserved in the Devonshire. The chapter also, though, looks at the ways that other sections of the manuscript preserve exchanges and conversations, wherein writers respond to and riff on one another’s works. The widespread use of such strategies emphasizes the particular literacies involved in the use of poetry at court; those particular literacies must be established for position-taking verse to circulate successfully via context-dependent codes. The established context of and comparison to translation and tradition allows for greater insight into exactly what transcription offered its authors that other outlets did not, while also structuring a framework for considering how and why these three strategies intersected in courtiers’ versified position-taking. In the communally created space of manuscript verse—and in the print versions of that verse that then arose—reader interpretation is understood as integral to the project. Because part of the argument throughout this manuscript engages the reasons for multiple versions of texts—or, more precisely, what is indicated by how a text is written for each particular audience—I have focused on Henrician texts for which some work is already available in terms of textual history. This is particularly true of my work with Tottel’s and with the Devonshire, but applies throughout, as in the engagement with the Thynne Chaucer, with a young Elizabeth’s translation work, and with Surrey’s Aeneid. In order to craft an argument about the position-taking work these verses did at court, I engage, for the most part, texts where a great deal of analytical, interpretive, and historicizing work has already been crafted, combining considerations of the politicized nature of Henrician verse with work on manuscripts that frame the book as a material cultural product. I attempt to avoid any claims of a totalizing argument by considering difference as much as alignment. I hope that the project traces some part of the shifts texts undergo as they move between manuscript and print and indicates something of what those shifts reveal about the specific functions of the Henrician courtly manuscript. The outlets through which the King and his courtiers attempted to preserve and privilege their positions often opened up new opportunities, and the humanist atmosphere of the court created subjects (in both senses) eager to take advantage of those new chances and changes. Taken together,
Introduc tion
25
these analyses should establish not only that verse was used by men and women at court for coded position-taking, but that courtier-poets at all levels were trained to recognize and use this poetic coding. Moreover, that recognition should allow for new readings of Henrician verse that emphasize the interpretive range available to these courtly reading and writing communities and, crucially, restore a more extensive network of production and reception in which women took on many roles. This demasculinizes our approach to Henrician verse not only through a more equitable consideration of gender’s functions in that social world, but also in de-emphasizing individualized self-fashioning or authorial intent in favor of an engagement with communal production and shared sociopolitical engagement. The availability of a wide range of interpretations is essential to the coded energy and potential of the verses considered here; the risk of misunderstanding is inherent in all communal position-taking projects, as creators and audiences work together to move between available meanings. While I hope to have produced an exploration of how these systems fit together, it is important that this system is not holistic. I have tried to offer some key moments of departure from this unifying model—that these systems were available to courtly verse producers does not mean they were universally used by or useful to them. The creation in this system is not of a code, but of systems for coding and recognizing position-taking; the poet does not create a self, a singular poetic voice. Rather, the communal systems offer a site for the intersection of reader and writer, of transcriber and composer, and of King and courtier in a space that questions, creates, and troubles power in the Henrician court.
Works Cited Baron, Helen. ‘Mary (Howard) Fitzroy’s Hand in the Devonshire Manuscript’. The Review of English Studies 45, no. 179 (1994): 318–335. Beal, Peter. In Praise of Scribes: Manuscripts and Their Makers in Seventeenth-Century England. Oxford: Clarendon, 1998. Blanchard, W. Scott. ‘Skelton: The Voice of the Mob in Sanctuary’. In Herman, Rethinking, 123–144. Bourdieu, Pierre. The Field of Cultural Production. New York: Columbia UP, 1993. Brigden, Susan. New Worlds, Lost Worlds: The Rule of the Tudors, 1485–1603. New York: Penguin, 2002. Burke, Victoria E. ‘Materiality and Form in the Seventeenth-Century Miscellanies of Anne Southwell, Elizabeth Hastings, and Jane Truesdale’. In English Manuscript
26
Gender and Position-Taking in Henrician Verse
Studies 1100–1700, Vol. 16: Manuscript Miscellanies c. 1450–1700, edited by Peter Beal and A.S.G. Edwards, 219–241. London: British Library, 2011. Burke, Victoria E. and Jonathan Gibson, eds. Early Modern Women’s Manuscript Writing. Burlington, VT: Ashgate, 2004. Butler, Judith. Gender Trouble. New York: Routledge Classics, 2006. Carlson, David R. ‘The Henrician Courtier Writing in Manuscript and Print: Wyatt, Surrey, Bryan, and Others’. In A Companion to Tudor Literature, edited by Kent Cartwright, 151–177. Malden, MA: Wiley-Blackwell, 2010. Carlson, David R. ‘Skelton, Garnesche, and Henry VIII: Revels and Erudition at Court’. The Review of English Studies 66, no. 274 (2015): 240–258. The Devonshire Manuscript. 1525?–1567? MS BL Add. 17492. British Lib., London. Dolven, Jeff. ‘Reading Wyatt for the Style’. Modern Philology 105, no. 1 (2007): 65–86. Ezell, Margaret J.M. Social Authorship and the Advent of Print. Baltimore: Johns Hopkins UP, 1999. Gibson, Jonathan. ‘Katherine Parr, Princess Elizabeth, and the Crucified Christ’. In Burke and Gibson, Manuscript Writing, 33–49. Gibson, Jonathan. ‘Miscellanies in Manuscript and Print’. In A Companion to Renaissance Poetry, edited by Catherine Bates, 103–114. Hoboken, N.J.: Wiley, 2018. Greenblatt, Stephen. Renaissance Self-Fashioning. Chicago: U of Chicago P, 1980. Griffiths, Jane. John Skelton and Poetic Authority: Defining the Liberty to Speak. New York: Oxford UP, 2006. Heale, Elizabeth. ‘“Desiring Women Writing”: Female Voices and Courtly “Balets” in some Early Tudor Manuscript Albums’. In Burke and Gibson, Manuscript Writing, 9–31. Heale, Elizabeth, ed. The Devonshire Manuscript: A Women’s Book of Courtly Poetry. Toronto: Center for Reformation and Renaissance Studies, 2012. Heale, Elizabeth. ‘Women and the Courtly Love Lyric: The Devonshire MS (BL Additional 17492)’. The Modern Language Review 90, no. 2 (1995): 296–313. Heffernan, Megan. Making the Miscellany: Poetry, Print, and the History of the Book in Early Modern England. Philadelphia: U of Pennsylvania P, 2021. Herman, Peter C., ed. Reading Monarchs Writing: The Poetry of Henry VIII, Mary Stuart, Elizabeth I, and James VI/I. Tempe, AZ: MRTS, 2002. Herman, Peter C., ed. Rethinking the Henrician Era: Essays on Early Tudor Texts and Contexts. Urbana: U of Illinois P, 1994. Herman, Peter C. Royal Poetrie: Monarchic Verse and the Political Imaginary of Early Modern England. Ithaca: Cornell UP, 2010. Herman, Peter C. and Ray G. Siemens. ‘Henry VIII and the Poetry of Politics’. In Herman, Reading Monarchs, 11–34. Hobbs, Mary. Early Seventeenth-Century Verse Miscellany Manuscripts. Aldershot: Scolar Press, 1992.
Introduc tion
27
Irish, Bradley J. ‘Gender and Politics in the Henrician Court: The Douglas-Howard Lyrics in the Devonshire Manuscript (BL Add 17492)’. Renaissance Quarterly 64 (2011): 79–114. Love, Harold. Scribal Production in Seventeenth-Century England. Oxford: Clarendon, 1993. Marotti, Arthur F. ‘“Love Is Not Love”: Elizabethan Sonnet Sequences and the Social Order’. English Literary History 49, no. 2 (1982): 396–428. Marotti, Arthur F. Manuscript, Print, and the English Renaissance Lyric. Ithaca: Cornell UP, 1995. May, Steven W. ‘Popularizing Courtly Poetry: Tottel’s Miscellany and Its Progeny’. In Pincombe and Shrank, Tudor Literature, 418–433. Norbrook, David. Poetry and Politics in the English Renaissance. New York: Oxford UP, 2002. The Oxford Dictionary of National Biography Online. Oxford UP, 2015. Web. https:// www.oxforddnb.com/. Accessed 30 June 2022. Pincombe, Mike and Cathy Shrank, eds. The Oxford Handbook of Tudor Literature, 1485–1603. New York: Oxford UP, 2009. Powell, Jason. ‘Marginalia, Authorship, and Editing in the Manuscripts of Thomas Wyatt’s Verse’. In English Manuscript Studies 1100–1700, Vol. 15: Tudor Manuscripts 1485–1603, edited by Peter Beal and A.S.G. Edwards, 1–40. London: British Library, 2009. Powell, Jason. ‘Thomas Wyatt and Francis Bryan: Plainness and Dissimulation’. In Pincombe and Shrank, Tudor Literature, 187–202. Remley, Paul G. ‘Mary Shelton and Her Tudor Literary Milieu’. In Herman, Rethinking, 40–77. Shirley, Christopher. ‘The Devonshire Manuscript: Reading Gender in the Henrician Court’. English Literary Renaissance 45, no. 1 (2015): 32–59. Siemens, Raymond G. ‘Henry VIII as Writer and Lyricist’. The Musical Quarterly 92, no. 1 (2009): 136–166. Siemens, Raymond G., ed. The Lyrics of the Henry VIII Manuscript. Grand Rapids, MI: English Renaissance Text Society, 2013. A Social Edition of the Devonshire Manuscript MS (BL Add. MS 17492). Wikibooks, The Free Textbook Project. https://en.wikibooks.org/w/index.php?title=The_Devonshire_ Manuscript& oldid= 3469218. Accessed 1 July 2022. Solomon, Deborah. ‘Representations of Lyric Intimacy in Manuscript and Print Versions of Wyatt’s “They flee from me”’. Modern Philology 111, no. 4 (2014): 668–682. Songes and Sonettes, Written by the Right Honorable Lorde Henry Howard Late Earle of Surrey, and Other. Ed. Richard Tottel. London: Tottel, 1557. Stevenson, Jane and Peter Davidson. Early Modern Women Poets (1520–1700): An Anthology. Oxford: Oxford UP, 2001.
28
Gender and Position-Taking in Henrician Verse
Warner, Christopher J. The Making and Marketing of Tottel’s Miscellany, 1557: Songs and Sonnets in the Summer of the Martyr’s Fires. Burlington, VT: Ashgate, 2013. Walker, Greg. Writing under Tyranny: English Literature and the Henrician Reformation. Oxford: Oxford UP, 2005. White, Micheline. ‘Katherine Parr, Henry VIII, and Royal Literary Collaboration’. In Gender, Authorship, and Early Modern Women’s Collaboration, edited by Patricia Pender, 23–46. Palgrave: 2017. White, Micheline. ‘The Psalms, War, and Royal Iconography: Katherine Parr’s Psalms or Prayers (1544) and Henry VIII as David’. Renaissance Studies 29, no. 4 (2015): 554–575. White, Micheline. ‘Women Writers and Literary-Religious Circles in the Elizabethan West Country: Anne Dowriche, Anne Lock Prowse, Anne Lock Moyle, Ursula Fulford, and Elizabeth Rous’. Modern Philology 103, no. 2 (2005): 187–214. The Workes of Geffray Chaucer Newly Printed, with Dyuers Workes Whiche Were Neuer in Print Before. Ed. William Thynne. London: Thomas Godfray, 1532. Woudhuysen, H.R. Sir Philip Sidney and the Circulation of Manuscripts, 1558–1640. Oxford: Clarendon, 1996.
1.
Early Verse Position-Taking in the Henrician Court Abstract: In this chapter, considerations of Henry’s use of verse as an outlet for political self-assertion join with analysis of his increasing centralization of power as a radical act, considering the two together as context for reactive verse work from Henrician courtiers. The increasing emphasis on the monarch as the source of power and privilege created restrictions under which courtiers chafed, but Henry also assigned particular privilege to courtly poetry, establishing its power as a fact of court life and so creating one of the outlets through which courtiers resisted those restrictions. Skelton’s work offers one entry in the early establishment of the position-taking power of verse, infused with the choices established by the King for privileging and coding such position-taking. Keywords: Henry VIII; John Skelton; Henrician verse; early Tudor poetry; courtly love lyric
In January 1510, Henry VIII surprised his new queen and her ladies, ‘appearing unannounced in her Chamber at Westminster with ten of his companions dressed as Robin Hood’s men’.1 Detailing the innovations of the performance, W.R. Streitberger notes the ‘elements of breaching ordinary decorum, concealed identity, and surprise’.2 Henry’s personal participation in the masque reflected the larger influence of his personality on all elements of court life and, ultimately, on England. Henry emphasized, through his early masques, his spontaneity, youth, and ostensible accessibility. At the same time, as Streitberger adds, ‘as the head of a household that governed the realm, his ceremonies, spectacles, and revels were public
1 Streitberger, Court Revels, 67. 2 Streitberger, 69.
Quoss-Moore, R.M., Gender and Position-Taking in Henrician Verse: Tradition, Translation, and Transcription. Amsterdam: Amsterdam University Press, 2023 doi 10.5117/9789463723534_ch01
30
Gender and Position-Taking in Henrician Verse
and so expressed the character of the state itself’.3 Henry was changing the particular investment of interest reflected in the masque; as Streitberger argues, speaking specif ically about the revels of 1510, ‘the interests of prestige diplomacy were being served, but the manner of their presentation, emphasizing participation by the King and his friends, personalized diplomacy in a manner calculated to overwhelm’. 4 Thus, in these early masques, Henry set the tone of his new administration through the performance that would come to be seen as an essential characteristic of Tudor monarchy. Seen through this lens, the King’s control of the masque emphasizes his aims to expand his control more generally. The changes which Henry VIII wrought to the masque entertainments reflect the tension of law and social code that characterized the Henrician era: Henry aimed to expand his control through every possible avenue. As a result, his courtiers would find themselves confronted by constraining and conflicting interpretations of the legality of Henry’s aspirations and of their own obligations to the court and country.5 That Henry was already planning a character of state that centered more thoroughly on him, personally, and through which he planned to overwhelm his opposition is apparent even as early as his coronation oath, which he tried to alter ‘to emphasize the primacy of regal authority’.6 Alice Hunt has usefully discussed the complex dual ideation of coronation in the period, summarizing that ‘the doctrine of divine right […] pledged the already-sacred nature of the king and the legitimacy of his rule but […] did not alter the ceremony’s insistence on transformation through anointing or its political and cultural prominence’.7 Because of the rules of hereditary monarchy, that is, ‘the king was king before the coronation […] the Church now had to signify, rather than render; its function was to make visible and tangible the divine power that had already been granted’.8 This complicated interplay of divine right and symbolic act were at work in every country whose monarchy rested on heredity; in England, though, a third factor was also in play: ‘at the coronation the king’s oath was sworn and thus the 3 Streitberger, 5. 4 Streitberger, 69, italics mine. 5 My understandings of the court have been particularly shaped by Scarisbrick, Henry VIII; Weir, The King and His Court; Brigden, New Worlds; Loades, Henry VIII; and Anglo, Spectacle, in addition to the sources more specifically cited throughout this chapter—especially those by Walker, Carlson, Herman, and Siemens. 6 Herman, Royal Poetrie, 23. 7 Hunt, Drama of Coronation, 12–13. 8 Hunt, 13–14.
Early Verse Position-Taking in the Henrician Court
31
constitutional purpose of the ceremony remained paramount’.9 In his first thoroughly public performance, Henry attempted to change that oath, making even the constitutional element instead a marker of his rights and his divine power. The oath which Henry attempted to revise was a 300-year-old contract between the King and those he governed, written during the reign of Edward I and used for his son’s coronation. Hunt outlines the purpose and the importance of this oath for England’s constitutional, hereditary monarchy, as it ‘defines, and limits, the king’s powers in relation to past and present laws, and his power over laws that shall be made subsequently during his reign. It is the contract between king and clergy, king and people, king and law, and ties him to promises for which he can subsequently be held accountable’.10 Through the traditional ceremony, the king acknowledged his place as below the law and his purpose as serving the people. The careful wording of the oaths he swore in the coronation located the king ‘below Parliament and its law-making capacities. He promised to strengthen and defend laws that his people had already made and will make’.11 From our retrospective position, Henry’s resistance to such an oath is hardly surprising. His desired alterations were, in fact, much in line with essentially every major change he would bring to England during his reign: they focused on his conscience and his power. He added to the promise to ‘mayntene the lawfull right and the libertees of old tyme’ that he would do this insofar as such were ‘nott preiudyciall to hys Jurysdiccion and dignite ryall’; he would ‘kepe all the londes honours and dignytes’ which were ‘nott preiudiciall to hys Jurysdiction and dygnite ryall and fredommes of the crowne of Englond’.12 However, at the beginning of his reign, Henry often found his more extreme efforts, including this attempted change, thwarted; 13 during the coronation, Henry swore the traditional oath, without these additions. Perhaps because of these limitations, or perhaps because his strategies of control were more successful when aimed at altering social codes,14 Henry still ‘ascended the throne as the darling of the humanists’.15 Admittedly, the grandiose claims of perfection made by diplomats, courtiers, and poets ‘were […] stylized and conventional [;…] ascribing to a new prince 9 Hunt, 14, italics mine. 10 Hunt, 26. 11 Hunt, 27. 12 Legg, English Coronation Records, 240. 13 Herman, Royal Poetrie, 23–26. 14 Herman, 26. 15 Walker, Writing Under Tyranny, 5.
32
Gender and Position-Taking in Henrician Verse
the sum of all intellectual and physical excellence was a commonplace of political eulogy’.16 The particular energy with which the people accepted Henry, though, indicates that at least some of those initial hopes were seen as fulf illed in the new King, particularly in light of the common wisdom that ‘good kings did not need to be perfect, only to be willing to listen to the criticism that would reveal and correct their shortcomings’.17 Nearly a decade into his reign, in 1517, Erasmus’s praise of the King is particularly effusive, and so particularly poignant with the hindsight of history: ‘what chiefly commands my approbation is this, that whereas, being gifted with an extraordinary clearness of mind, you have no lack of wisdom yourself, you still delight in familiar converse with men of prudence and learning, and most of all with those who do not know how to flatter’.18 Greg Walker attributes to exactly such positive perceptions the fact that, in the early sixteenth century, ‘a generation of politically active individuals grew to maturity secure in the knowledge that England was indeed a stable polity in safe royal hands’.19 The widespread admiration of the King was, in fact, essential to the discontent that would follow as that perception changed. The long tradition of interpreting courtly love lyrics in terms of political resentment emphasizes that courtier-poets chafed at social as well as legal restrictions.20 Returning to the masque as a site of practiced power, a relevant addition is Skiles Howard’s analysis of dancing and festivity in the Henrician court as ‘a privileged site for the production of hierarchy and gender difference’,21 placed ‘at the center of the constellation of nonverbal practices that consolidated political power in the sovereign by performing the work of social stratification’.22 Henry was shifting the social code to privilege his own power, and ‘behaviour became a new heraldry by which the courtier created his nobility and signaled his allegiance’.23 Brian Lockey emphasizes this point, acknowledging that ‘traditional blood links were 16 Walker, 9–10. 17 Walker, 11. 18 Erasmus, ‘To Henry VIII’, in Renaissance Letters, 255–257. 19 Walker, 13. 20 Carlson’s ‘Henrician Courtier’ offers several key points in this interpretive tradition; Crane specifically connects the intensity around ‘a fear of lacking voice or presence and an awareness of the difficult of controlling language’, drawing on Derrida, to ‘Henry VII and Henry VIII [‘s erosion] of the feudal nobility’s power and [establishment of] a powerful, centralized monarchy’ (Framing Authority, 13). 21 Howard, ‘Performing Hierarchy’, 17. 22 Howard, 17. 23 Howard, 19.
Early Verse Position-Taking in the Henrician Court
33
replaced by a new chivalric code of virtuous service to the sovereign and kingdom’. 24 Despite Howard’s own use of the masculine pronoun, the consideration of the social space of courtly entertainments emphasizes the key point that these changes applied to all courtiers—not in ungendered terms, but certainly in ways that impacted people of all genders at court. Educated court poets, in the latter third of Henry’s reign, found themselves in the impossible position of believing that service to the kingdom under Henry’s increasing despotism necessitated displeasing the King even as the shifting social code emphasized service to the King as the primary site of privilege. Henry created the restrictions under which they chafed, but, as this chapter explores, he also played a key role in assigning privilege to courtly poetry and in establishing its power as a fact of court life—creating one of the outlets through which his courtiers would resist those restrictions.
A. The King’s Courtly Verse Henry’s understanding of law as dependent on his conscience and subject to royal interpretation seems evident from even his earliest actions, even if initially manageable; Henry VIII’s own early lyrics reveal similar aims to his early revisions to the masque, and to that first, failed attempt to alter the English legal tradition in his own interest.25 Like the bridges of preserved practice Henry used in his masque, his early poems ‘directly respond to the anxieties caused by the crowning of a new king whose policies and personality differ radically from the previous monarch’s’;26 they served as an outlet, ‘establishing his independence from Henry VII’s policies [… and] articulating as forcefully as possible that he, and only he, rules the land’.27 Through his verse and performance—and verse as performance—Henry uses enough tradition to establish his reign as a continuity within a larger historical arc, while introducing innovations and manipulations always designed to privilege not only all royal power, but specifically Henry’s power. Peter C. Herman and Raymond Siemens explicitly make this connection, as well: ‘Like Henry’s masques and disguisings, his lyrics also constitute vehicles for depicting the hierarchy of the court and for both defending and 24 Lockey, Law and Empire, 34. 25 Herman first outlined this arc in Royal Poetrie, 15–51. 26 Herman and Siemens, ‘Introduction’, Reading Monarchs Writing, 6–7. 27 Herman and Siemens, ‘Introduction’, 7.
34
Gender and Position-Taking in Henrician Verse
reinforcing the power of his monarchy’.28 From early in his reign, Henry uses every element of royal performance at his disposal to respond to the limitations of the law upon his power and to manipulate social and legal codes in his own favor. Henry wishes to manipulate the contemporary system, working with the tools of poetic expression, to overthrow that system, reestablishing power according to his design. Herman and Siemens argue that, in most of his love lyrics of the 1510s, ‘Henry is drawing upon and manipulating the generic conventions of the courtly love lyric to provide an acceptable face for marital relations, which themselves are suggestive of national political strength, stability, and Henry’s ability to deal with political discord’.29 Such a strategy combines the conventional equation of marital stability with national stability alongside a reworking of the privilege generally afforded the speaker of Petrarchan love lyrics to accommodate royal authority. Henry’s intentional centering of his own marriage, moreover, invites a consideration of women’s multiple involvements in courtly verse. While I am not aware of any evidence that Katherine of Aragon herself wrote poetry, her place in Henry’s verse requires that we reconsider a simple label of ‘object’ or ‘subject’; if Henry’s verse equates marital and national relations, then his work necessarily also imbues Katherine’s position with political power. Henry does, unsurprisingly, exploit this positioning to emphasize his own virtues; correspondingly, claims of national stability are deployed to advance new positions, as Henry aims to supplant the extant, constitutional system with a more absolute monarchy by emphasizing his own worth. The bulk of Henry’s surviving poetry is recorded in BL Additional 31922, which ‘dates from c. 1515, when Henry VIII was twenty-four years old’.30 Aside from the physical evidence that the manuscript was bound by 1522, from earlier works,31 the dating is also reinforced by ‘the topicality of some songs which are related to specific events of this period, a time of frenetic festive activity’.32 In this manuscript, for those verses which Henry composed, ‘attribution to “The Kynge H. VIII.” […] is centered at the top of the leaf on which each piece begins’.33 Siemens observes that this attention-grabbing style ‘sets Henry’s works apart from that of others collected in the manuscript’;34 the 28 29 30 31 32 33 34
Herman and Siemens, ‘Henry VIII’, 15. Herman and Siemens, 21. Stemmler, ‘Songs and Love-Letters’, 98. Siemens, ‘Henry VIII as Writer’, 147. Rupp, ‘Performing Court’, 106. Siemens, ‘Henry VIII as Writer’, 139. Siemens, 139.
Early Verse Position-Taking in the Henrician Court
35
style also distinguishes Henry’s works by highlighting their separation from the more communal nature of wider manuscript production, indicating how the King’s strategies shift when appropriated by his courtiers. His texts are set apart from ‘the tradition of adaptation and re-adaptation out of which many of the lyrics in Henry VIII MS have come’; they are more specific to the moment, the court, and, above all, the author.35 This aggrandizing style also, though, distinguishes Henry from other kings and from his predecessors on the English throne; Susanne Rupp observes that, of the humanist kings of the sixteenth century, only Henry was actively ‘promoted as a composer’.36 In addition to emphasizing his status above other English composers and the talents as a composer that set him above other kings, Henry’s choices in his poetry emphasized the shift from his father’s court; his lyrics reflect the style he designed to show that he ‘would have a distinctive, novel cultural style […] that he would prefer younger persons, generally of a less austere and sober demeanour, and different kinds of recreation’.37 All of Henry’s poems in the manuscript emphasize courtly love, youth, or the pastimes associated with each. When joined to Henry’s frustrated attempts to explicitly redefine kingship, what emerges is a manipulation of these tropes to work his redefinition through other ends. Rupp observes that the manuscript is ‘embedded in a courtly world, participating in and at the same time producing this world’;38 David Lyle Jeffrey outlines the process by which ‘Henry’s courtly theatrics effectively strip the earlier French literary conventions [of courtly love] of their irony, play out polite and public flirtations as private fantasy, then legislatively normalize the consequences’.39 That is, Henry’s masques and poems use the resistant privileging of love and youth—and the often martial themes associated with their intersection—‘to define a certain social group, its practices and codes of behaviour and allow […] its members to perform their community’.40 Because Henry’s definitions are not yet the dominant ones, ‘Henry’s songs not only figured as instruments producing community, intimacy, or as a way of asserting the King’s competence and power, but they also functioned as media of social criticism and expressions of discontent’. 41 Herman encourages us to understand Henry, rather than Wyatt or Surrey, as the strongest influence in ‘using 35 36 37 38 39 40 41
Siemens, 141. Rupp, ‘Performing Court’, 107. Carlson, ‘Revels and Erudition’, 241. Rupp, ‘Performing Court’, 108 Jeffrey, ‘Courtly Love’, 527. Rupp, ‘Performing Court’, 109. Rupp, 111.
36
Gender and Position-Taking in Henrician Verse
erotic terms to talk politics’42 and to restore the centrality of the King’s verse to ‘the political imaginary of the early Henrician court’. 43 Henry not only participated in but also partly created the system of poetic position-taking which would later be used to express resistance to his ideas—or at least, codified the form this system took during his reign. While beginning with his verse risks a reification of the focus on masculine power at court, I hope to instead center Henry’s understanding of his own power as quite specifically monarchical, rather than, primarily, masculine in a sense shareable by other men—and, from there, to better engage courtiers’ position-taking as informed by positions that are aristocratic, first, and gendered second. Henry’s poetry is invested in establishing the power of the monarch to define those aristocratic positions. In ‘Thow that men do call it dotage’, Henry, by assuming the position of the lover and thus imbuing it with his royal authority, asserts that ‘the person who disdains love (and by implication the chivalric activities of the lover) has lost his place in the aristocracy: his disdain marks him as a peasant’. 44 Since the King has privileged the conventions of the courtly love lyric, to refuse participation in the chivalric expectations of the court is to refute one’s own nobility; Henry supplants the older advisers of his court by reworking tradition to privilege his own youthful activities. The poem opens by supplanting the common definition of an action with the King’s own: ‘Thow that men do call it dotage. / who louyth not wantith corage’. 45 The traditions of courtly love are both altered and made central to the definitions of appropriate courtly performance. Henry then narrows the definition of the true lover and the true courtier by claiming that ‘who loue dysdaynyth ys all of the village’. 46 The effect is doubled—the courtier who disdains love is no true courtier, and so love is ennobling. Further, though, love can be read as disdaining those of the village, and to be of the village is to disdain love, reciprocally. Thus, the rhetoric of courtly love is inherently inaccessible to those of lower rank. The first line of the poem necessitates that those who ‘call it dotage’ are inferior to the courtier who affirms the King’s system of courtly love; those who disagree with Henry’s definitions of roles and power are undermined and excluded from the power system. They prove their unworthiness as counselors when they disagree with Henry, rather neatly eliminating the 42 Herman, Royal Poetrie, 51 43 Herman, 50. 44 Herman, 29. 45 Henry VIII, Lyrics, ed. Siemens, 39–40, ‘Thow that men’, lines 1–2. 46 Henry VIII, ‘Thow that men’, line 14.
Early Verse Position-Taking in the Henrician Court
37
necessity of listening to dissenting voices. Theo Stemmler has analyzed a similar thread in ‘Whoso that wyll for grace sew’, where ‘the speaker sues for the lady’s grace. [… but] this sterile commonplace […] is followed by a lively attack on the detractors of love whom Henry denounces as envious hypocrites’.47 This process of discrediting those who do not fit into the system of courtly love is more than rhetorical posturing when the composer is a king, as is emphasized when Henry asserts his own will in the final line of ‘Thow that men’: ‘Chaunge who so wyll I wyll be none’.48 While his performance of power may, at the time of composition, face some constraints, Henry shows his resistance in his verses and clearly indicates that, as he grows into his power, he has no intention of becoming more biddable or more complacent. Henry then furthers the extension of royal power by not only defining nobility in terms of successful service but by also subtly asserting his own omnipresence and omnipotence in ‘Hey nony nony nony nony no!’49 In this early poem, Henry carefully limits the uses of his power to relatively benign observation and commiseration: he hears his lamenting subject, but ‘She had nott said. / but at abrayde. / her dere hart was full nere’.50 The King/speaker then observes the lovers’ reunion even though he claims that ‘They day thay spent. / to ther in tent. / In wyldernes alone’.51 In the poem, the lover’s sudden appearance to comfort his despondent counterpart offers ‘a sense that the royal speaker can observe his desiring subject(s) without revealing his presence, and then, through an act of will, of power, suddenly make things better’.52 In keeping with Henry’s investment in always expanding such power, Herman and Siemens also show how, in later lyrics, ‘the underlying implication [… becomes] that this power can be used for other, less benign purposes as well’.53 In these early manuscript poems, though, Henry more subtly outlines the terms of his power, returning to tropes of youthful love in his resistance to his more mature, cautious advisors. The King is an observer of all things, even of those who are alone, but his observation leads, at most, to the happy reconciliation of a disordered relationship—his expression of power is limited and benevolent. 47 Stemmler, ‘Songs and Love-Letters’, 98. 48 Henry VIII, ‘Thow that men’, line 20. 49 Siemens attributes only the short, three-line entry on the page, preceding this poem, to Henry (Lyrics, 29–30); I follow Herman in treating the longer work as the King’s (Royal Poetrie, 32–35). 50 Henry VIII, Lyrics, ed. Siemens, 30, ‘Hey nony nony’, lines 46–48. 51 Henry VIII, lines 55–57. 52 Herman and Siemens, ‘Henry VIII’, 24. 53 Herman and Siemens, 24. Herman adds to a later analysis that ‘the song […] rehearses the culture of surveillance that is beginning to develop in Henry’s court’ (Royal Poetrie, 34).
38
Gender and Position-Taking in Henrician Verse
Henry’s expressions of power are not always so subtly approached, though he uses the tropes of youth to code some of his most resistant themes. Pairing the two lyrics ‘The tyme of youthe is to be spent’ and ‘Whoso that wyll all feattes optayne’ reveals the strong thematic connections between his lyrics on youth and those on love, each of which privileges the energies and courtly recreation which separate Henry VIII’s new court from that of his father. In ‘The tyme of youthe’, Henry asserts his own good judgment even as he makes apparent concessions to the wisdom of his elders. He asserts that ‘The tyme of youthe is to be spent / but vice in it shuld be forfent’.54 In identifying vice and virtue, the kingly speaker’s next lines do not bow to the outside judgment of his anxious counselors. Rather, Henry continues that ‘Pastymes ther be I nought trewlye. / Whych one may use. and uice denye. / And they be plesant to god and man’.55 Henry himself, not some outside figure, notes those pastimes which are suitable; further, he already asserts, by association, his knowledge of God’s will, as he knows what is pleasant to Him. These acts also include some behaviors about which his counselors had expressed reservations. One of these pastimes is explicitly ‘featys of armys’;56 he also includes, more generally, ‘actyuenesss’.57 The royal speaker ends his poem with the declaration that ‘Vertue it is. then youth for to spend. / In goode dysporttys whych it dothe fend’.58 Henry asserts his royal power to declare what is virtuous, regardless of others’ arguments, while, still, acknowledging the importance of avoiding vice and seeking that virtue. The poem demonstrates a fine balance between increasing assertions of royal prerogative and still necessary assurances that the King will practice the ‘correct’ values. Although there are plenty of acts of definition in the short poem by which Henry determines the shape of courtly behavior, he still seems to speak as an act of justification to some other who might disapprove of his actions. This balance continues in ‘Whoso that wyll all feattes optayne’. As in ‘Thow that men’, the speaker asserts an interdependent relationship amongst love, courage, and nobility. The opening verse announces Whoso that wyll all feattes optayne. In loue he must be withowt dysdayne. 54 55 56 57 58
Henry VIII, Lyrics, ed. Siemens, 25, ‘The tyme of youthe’, lines 1–2. Henry VIII, lines 3–5. Henry VIII, line 7. Henry VIII, line 8. Henry VIII, lines 11–12.
Early Verse Position-Taking in the Henrician Court
39
For loue enforcyth all nobyle kynd. And dysdayne dyscorages all gentyl mynd.59
Once more, the lover who acts correctly is necessarily ‘gentyl’, and one who is gentle will necessarily act correctly. Further, this gentle lover will then be able to act in the correct and courtly manner, obtaining ‘all feattes’ and acting with appropriate courage. Henry tells his audience that ‘loue encoragith. and makyth on bold’.60 To be bold and accomplished is to be appropriately courtly, and to be these things depends upon correct performance of courtly love, as well. In all of these things, the true courtier will perform the correct terms of service. In opposition to the ideals of instruction and guidance which those counselors who kept him from altering the coronation oath put forward, Henry offers a vision of service and courtly entertainment, with himself at the center defining those who surround him. The poem continues with a particularly nuanced statement of Henry’s ideas of love and the gendered performances of love that he uses to define his court. He writes that ‘loue ys gevyn. to god and man. / to woman also. I thynk the same’.61 These two lines are complicated by Henry’s use of ‘I thynk’. In one sense, Henry’s thinking this so may make it so—that is, the King defines the terms and speaks with knowledge of God’s will, so his belief is necessarily correct. The other sense of ‘I thynk’, though, is one of uncertainty—that is, the speaker uses the construction to signal something that is actually less certain than a belief; it is instead a postulation or a possibility. Is Henry saying that love is the same for both men and women, and they may act by the same rules and be ennobled by the same terms? Or is he saying, with a slight but significant difference, that man and God act by the same terms, and that it is possible for those terms to apply to women as well? The woman of ‘Hey nony nony’ loves truly and is rewarded for her fealty, though she has little agency—she obtains no feats, but only waits and is rewarded. However, these terms of service are also acceptable in some versions of masculine performances of courtly love; patience and loyalty may be rewarded over more active pursuit. This, though, may be where Henry’s definitions create the divide that has been traditionally interpreted as troubling courtier poets like Wyatt and Surrey. According to the definitions provided by his poems, women and men may love the same; the ideals Henry espouses in his own voice and performance, though, could equally suggest that a man’s part lies 59 Henry VIII, Lyrics, ed. Siemens, 32, ‘Whoso that wyll’, lines 1–4. 60 Henry VIII, line 7. 61 Henry VIII, lines 9–10.
40
Gender and Position-Taking in Henrician Verse
in action and feats while the woman’s lies in loyal, patient chastity. When, then, Henry later demands patient loyalty without action from his men courtiers, he pushes them into positions that he himself has feminized. As we will see in later chapters, though, resistance to these less-active ideals of courtly engagement was not restricted to poets who were men, and that resistance can be read as partially, if not equally, informed by the increasing separation Henry desired between absolute power for himself as monarch and limited power, with himself as the source, for his courtiers. Together, these two poems provide clear examples of the terms by which Henry used youth and love as outlets for resistance against the traditions and limitations that were imposed on him. Siemens calls such work ‘an act of poetic self-justification, an address of the young lover that Henry really was at the time, to the aged disdainers […] opposing his actions’.62 After all, ‘in the relationship of youth and age, it is youth who is subservient; in the relationship of the lover and the disdainer who thwarts the efforts of the lover, it is the lover who is subservient’.63 However, Henry overthrows these modes of subservience; ‘he brings a political weight not typically available to the youth or the lover but only to the King, one who is truly in command of all subjects, including the disdainers’.64 Siemens observes, in this process, ‘a voice in an artificial though well-accepted discourse through which aspects of reality can be discussed’.65 The aim of Henry’s verse position-taking in this discourse is apparent enough; Henry is generally interested in gaining and centralizing power. The effect, whether intended or not, is to create a system of coding, wherein youth and love become themes entwined with resistance to larger, older, more established power. Among these position-taking verses, Henry’s ‘Though sum saith that yough rulyth me’ seems the most assertive, but the likely dates of composition indicate that this vehement assertion of personal prerogative was, in fact, a resistant act particularly suggestive of later court tensions. Despite the clear assertions of royal power and prerogative, the poem was composed early enough in his reign that these assertions on Henry’s part are not reflective of courtly practice, yet. Indeed, the so-called ‘minion crisis’—in which several of Henry’s close, boisterous friends were effectively banished from court to attempt in an effort to coerce the young King to ‘better’, calmer council—did not occur until after the supposed date of composition for most 62 63 64 65
Siemens, ‘Henry VIII as Writer’, 154. Siemens, 154. Siemens, 154. Siemens, 154.
Early Verse Position-Taking in the Henrician Court
41
of Henry’s lyrics.66 Thus, the proud declarations with which Henry opens ‘Though sum saith that yough rulyth me’ must be optimistic projections, efforts at resisting the control asserted over him. Nonetheless, this poem does represent some of Henry’s least subtle reminders that the speaker of these verses is also the King. The first lines establish both resistance and the royal voice: Though sum saith that yough rulyth me I trust in age to tarry. god and my ryght and my dewtye frome them shall I neuer vary, thow sum say that yough rulyth me.67
Henry’s insertion of his own royal motto in the third line makes the speaker unmistakable68 and aligns that speaker with the established systems of power markers in the invocation of a ‘motto’. However, Henry both makes a mitigating addition and sets that addition off in a way that minimizes the effect. By adding ‘my dewtye’ to the translated ‘Deiu et mon droit’, Henry suggests his obligations to systems outside his own conscience. He acknowledges, in fact, some of the very limitations he sought to mitigate in his alterations to the coronation oath, but he simultaneously minimizes that claim, setting it apart from the more important ‘god and my ryght’ by a pause that makes his obligation to ‘dewtye’ appear as a kind of afterthought, perhaps an obligation he will pursue so long as it is not prejudicial to his jurisdiction and dignity. Thus, even as he makes mitigatory gestures towards recognition of those limitations imposed on his will by the coronation oath and the traditions therein represented, he also offers indications that this limitation is of less significance than is his own will. The resistant tone becomes more pronounced in the second stanza, as the royal speaker levels a clear challenge at those who would limit him, asking: I pray you all that aged be. How well dyd ye yor yough carry. I thynk sum wars of ych degre. Ther in a wager. lay dar I.69 66 See Weir, The King and His Court, especially Chapter 25, and Scarisbrick, Henry VIII, 117–120. 67 Henry VIII, Lyrics, ed. Siemens, 45, ‘Though sum saith’, lines 1–5. 68 Herman, Royal Poetrie, 43. 69 Henry VIII, ‘Though sum saith’, lines 6–9.
42
Gender and Position-Taking in Henrician Verse
Some of Henry’s counselors, of course, could not honestly claim any particularly virtuous youth; this question would be particularly fraught for those advisors who discouraged Henry’s military ambitions when they themselves were young men during Henry VII’s military campaigns. At best, they become hypocrites who deny the young King the glory they themselves accrued; at worst, the line offers threatening rebuke to those pardoned for fighting against the establishment of the Tudor dynasty. However, the question and wager are also fundamentally unanswerable. Even the most virtuous advisor cannot actually respond by laying claim to greater virtue than the King—the answer must be that they were worse in degree, because any other answer claims power and precedence over Henry. This echoes the undermining effect of ‘Thow that men’—the councilors’ only possible response is to declare themselves subordinate to the King, and thus to acknowledge his greater authority. As a response to those who would counsel him to greater caution, the poem rhetorically establishes the innate superiority of kingly judgment, even as it pretends to acknowledge the subordinate position of youth. The final stanza is perhaps the strongest, in the context of a series of verses which assert Henry’s rights in forceful terms. The King, essentially, prays for those who would oppose him after he has declared an end to this debate; he then asserts his own identity and primacy, adding a sneer to his final line at the idea that something else could claim the position of rule: Pray we to god and seynt mary. That all amend and here an end. Thus sayth the king the .viii.th harry. though sum sayth that yough rulyth me.70
The marked egotism of Henry’s careful, repeated references to his own position mark the incaution of disagreeing with him too strongly. However, the very fact that he feels the need to make this assertion so often—and that he finds verse a useful outlet for these assertions—marks the relative weakness of his power at the opening of his reign. He asserts his title because others are attempting to control him despite that title. And he makes these assertions through the social force of verse rather than through the legal and governing outlets in which he has thus far been often overruled. In his verse, Henry makes the politically savvy decision to rewrite the social rules in order to be able to rewrite the rules of his own position, once his more overt efforts have been thwarted. 70 Henry VIII, lines 17–20.
Early Verse Position-Taking in the Henrician Court
43
A natural companion to ‘Though sum saith’, reflecting a great deal of the same sort of thematic concern with well-spent youth, is perhaps Henry’s most studied composition: the lyric ‘Pastyme with good companye’. In general, this poem has been understood as one of Henry’s most assertive; Stemmler characteristically remarks on Henry’s ‘intransigence and uncompromising egotism’,71 concluding that ‘the intimidating question “Who shall me lett?” might have been the motto of his life’.72 Rupp, however, provides interesting counterevidence for this interpretation. Based on the extant versions of the composition, she counters that ‘Pastyme’ ‘appears to be one of Henry VIII’s “childhood exercises”, conducted under the severe eyes of a severe tutor’.73 As such, the song is an act of king-fashioning by his councilors, not an act of self-fashioning by Henry VIII—or at least not entirely or initially. If these more controlling tutors and counselors were indeed involved in the composition of the poem, Rupp argues that they may have thought that ‘by providing the young King with this material, his innate drive to commit excess would be kept under control and firmly contained according to the rules given by the elders’.74 This argument suggests that Henry may have learned some important strategies from his early tutors, quite apart from the lessons on virtuous activity that they wished to impart. By providing a social code which dictated acceptable outlets for resistant energy, expressed in verse, the tutors aimed to control that energy while privileging and codifying their own power. When, later, these same early councilors resisted Henry’s first attempts to expand his power, he took his lessons in verse and shifted the narrative so that the privileged power became his. Rupp’s dating points to the ways in which even this early verse, composed for a more authoritative audience, could be manipulated to express resistance. The opening lines of the poem declare: Pastyme with good companye I loue and schall vntyll I dye gruche who lust but none denye so god be plesyd thus leue wyll I75 71 Stemmler, ‘Songs and Love-Letters’, 101. 72 Stemmler, 102. 73 Rupp, ‘Performing Court’, 110. If this is a childhood work, the comparison to Elizabeth’s early literary works, such as her translation of Le Miroir, may be of further interest—as are the differences in relative critical responses to the King-to-be’s and Queen-to-be’s works. 74 Rupp, 110. 75 Henry VIII, Lyrics, ed. Siemens, 20, ‘Pastyme’, lines 1–4.
44
Gender and Position-Taking in Henrician Verse
While the initial assertion of the young King’s (or perhaps, as Rupp argues, still king-to-be’s) will is clear, strong, and immovable, the final line does seem to slide into a more mitigated claim. The King loves good company, so long as God is pleased. Rupp points out the doubled interpretation that the verse leaves available: ‘If a tutor is around who wants the youngster to have some socially acceptable fun, this line would probably be read as a conditional clause […] If the severe tutor is absent, a more daring interpretation could be desirable, uttering a request along the lines of “God had better be pleased with the King’s actions”’.76 Since Henry will, eventually, come to apparently understand his own conscience as the best indicator of God’s will, the lines now read as more resistant than the original audience could have known. From the hand of the younger, less powerful Henry, though, the poem instead reflects his neat balance of self-assertion and self-justification. The final verse of the poem essentially summarizes many of Henry’s strategies in his early compositions. He asserts the freedom of choice, his own youth and energy, and his own innate virtue—but the lines also contain self-justification and seem to indicate response to some kind of external critique. The poem ends: Company is good and ill but euery man hath hys fre wyll. the best ensew the worst esshew my mynde shalbe. vertu to vse vice to refuce thus schall I vse me.77
On the one hand, the declarative ‘schall’ and the repeated claims to virtuous inclination are characteristic of Henry’s royal voice. On the other, the entire final stanza has the sense of a schoolroom promise of good behavior—every man has their choice, the King is one of these men, and he promises to choose correctly. The correct delivery, with emphasis on the self-declarative lines, could slant the poem into greater power, even as another delivery which deemphasized those lines makes the tone more pedantic and platitudinous. This lyric, then, may be read as one of the earliest of his efforts at expressing 76 Rupp, ‘Performing Court’, 110. 77 Henry VIII, ‘Pastyme’, lines 23–30.
Early Verse Position-Taking in the Henrician Court
45
resistance to the powers that currently limited him, where his dissent can be masked by something as small as the alteration of tone. In summation, in his reworking of fairly traditional love lyrics and verses on youthful pastimes to incorporate a royal voice, Henry ‘uses the persona of Youth to pre-empt or answer criticism of how he chooses to conduct his life, but goes against generic expectations of Youth by asserting his royal right to do exactly as he pleases’.78 By privileging the position of Youth or of the youthful lover, Henry slowly erodes the position of his sage advisors; by casting true courtly behavior in terms of loyal service, he undermines the ability of the courtier to express dissent. By slowly altering the social codes of the early sixteenth century to suit his purposes, Henry sets the groundwork to later be able to modify English law, removing the support system for the traditional defenses that would stand in his way. Henry shows in his early verse the shape he will later impose on the law. Conventions are used in whatever manner is most useful to him; royal power is omnipresent; the interpretation that most privileges Henry’s specific royal authority is the most correct interpretation. The responses of Henry’s courtiers work with and through the law quite differently. However, like Henry’s reactions, the lyrics of the court poets are all shaped by their understandings of their own privileges, whether poetic or political. As these poets become increasingly aware of the threats to those privileges, they also become increasingly canny in their responsive techniques.
B. Skelton and Early Henrician Courtiers’ Position-Taking Complicating the tensions of these multiple, competing definitions and social expectations was the contemporary conflict between the several English courts of law. Though difficult to apply in each specific analysis of English court verse, the debate is nonetheless relevant to contemporary understanding of what the law meant, how it functioned, and why it should be respected, all issues of paramount concern to courtiers in their interpretations of their own roles and of the King’s power. Sir John Fortescue’s influential treatise states simply that, based on the sheer age and widespread acceptance of the customs of English common law, ‘there is no gainsaying nor legitimate doubt but that the customs of the English are not only good but the best’79 because they have not required alteration and so must be 78 Herman and Siemens, ‘Henry VIII’, 30. 79 Fortescue, Laws and Governance, 27.
46
Gender and Position-Taking in Henrician Verse
nearest to ‘natural’ law. This understanding was widespread amongst an English populace that tended to take a great deal of pride in the function of their legal system. The relationship, though, of common law to ‘natural’ law was quite complicated; ‘[w]hile common lawyers sometimes spoke of natural law or the law of reason as the basis of positive law, they more often viewed English custom as its primary foundation […] Civilians saw equity, the civil law, and the canon law as direct derivations from a universal natural or divine law’.80 This resulted in different ideas of how law should be considered and evaluated in terms of personal conscience or absolute right and wrong.81 For example, the common-law jurist Christopher St. German argued in a 1528 tract ‘that conscience, the personal expression of conformity with natural law, should depart from positive law and defer to natural and divine law in cases where the positive law contravened the latter two laws’.82 This left the disturbed courtier in a complicated legal and social space—convinced, in most cases, of the validity of the King’s rule and of English law, but conflicted by the King’s own alteration of English law, partially enabled by his frequent claims to pangs of conscience. Thus, the position-taking poetry created by these courtly makers was sometimes quite radical—aiming to overthrow restrictions imposed by authority—and at other times may have seemed more conservatively grounded within a long-established tradition of speaking truth to power. That tradition heavily informed the work of early courtly makers, particularly John Skelton, whose work offers some of the clearest examples of courtly understandings of the role of verse in Henry’s court. Skelton’s work offers a clear contextual precedent for the privileging of poetic power which Surrey would later take up, though in new forms and with different motivations. Further, Skelton’s early work was, in some ways, bolstered by the King’s own poetic efforts, even though Skelton’s attempts to remain in royal favor were limited in their success. Placed in context with women’s manuscript production and alongside Wyatt’s and Surrey’s works, as explored in later chapters, Skelton’s poetry provides a starting point for a map of the role of courtiers’ poetry in Henry’s court: first, openly critical but functioning in an approved space, then, denied power and rejected, and so moving underground into more coded expressions which Surrey will, ultimately, push back out into more public and open indications of resistance against the King—ones with disastrous contemporary 80 Lockey, Law and Empire, 148. 81 Lockey, 149. 82 Lockey, 154.
Early Verse Position-Taking in the Henrician Court
47
consequences but which ultimately worked to preserve resistant voices for modern audiences. The court poets treated in later chapters were building from a framework that had been extensively shaped much earlier in Henry’s reign: partially by the King’s own insistence on courtly love rhetoric, wrapped in poetry—but also partially by courtier poets like Skelton. Walker convincingly argues that Skelton’s often explicitly critical poetic voice is actually in service of his own personal advancement,83 but W. Scott Blanchard usefully points out the real, radical energies that nonetheless exist in Skelton’s works.84 The most useful criticism likely combines both approaches: Skelton’s personal motivation is best articulated as creation of a privileged space for poetic authority, specifically in the interest of gaining a more privileged position himself.85 Skelton’s miscalculations in anticipating royal opinion in these attempts actually imbue his work with more effective critical power. Particularly later in his career, Skelton’s voice is partially aligned with the commons—albeit only partially—‘despite his conservative understanding of class structure’;86 as such, his works become potential agents for social change beyond the scope of the poet’s intentions. At the same time, as Walker claims, Skelton ‘is perhaps unique among poets in the depth of his concern for his own status and reputation’.87 The validity of the claim rests on a particular definition of ‘depth’; certainly other poets cared a great deal about their social status, particularly in terms of power and privilege, but Walker makes a fair argument that perhaps no other courtier poet cared so deeply about their personal authority as vested in their poetry, specifically. As Blanchard articulates, Skelton’s ‘traditional understanding of class structure urged him to anchor himself in the clerical estate, while the very threats to that class moved him in a more radical direction toward the autonomous and secular institution of “authorship”’.88 This problem of social identification was complicated by the fact that, particularly after his dismissal from his post as Henry’s tutor, ‘Skelton was not a regular contributor 83 Walker, Writing Under Tyranny, 53. 84 Blanchard, ‘Skelton’, 126. 85 Griffiths points to early print editions of Skelton’s work as evidence that he is ‘both more engaged with his readers and more adaptable to change than has often been suggested’, specifically seeing his use of the new medium as both achieving ‘political ends and […] defining and asserting his own poetic authority’ (‘Last Word’, 70). My argument, here, is that establishing his own poetic authority is one of Skelton’s primary political ends. 86 Blanchard, ‘Skelton’, 128–129. 87 Walker, Writing Under Tyranny, 53. 88 Blanchard, ‘Skelton’, 136.
48
Gender and Position-Taking in Henrician Verse
to Court events, nor a fixture of the social life of the Court’.89 As a result of his frustrations at this exclusion, ‘Skelton pursued in his poetry both directly and indirectly his own conflicting attitude toward dissent, a theme which lies beneath his specific complaints against a clergy reticent to enforce its traditional privileges, a nobility which had neglected its responsibilities, and a commons which was experiencing repressive measures’.90 His own dissent was naturally privileged in Skelton’s work and thought; he saw his own struggles with the ruling class as resulting, generally, from their lack of appreciation for his poetic authority. This personal dissent, though, may have also enabled Skelton to seriously view complaints and problems from other factions which he might have more easily dismissed had he ever gained the privileged position for which he aimed. This complicated balance of complete investment in and deep suspicion towards the social structures of Henrician England led Skelton to ‘the various strategies he deemed necessary to empower himself as a dissenting member of the clerical class and to distance himself from his utterances through the construction of provisional authorial personae [, which] show him struggling to define his social role’.91 Blanchard’s argument ultimately leads back to Walker’s central claim: Skelton’s struggles with authority are all essentially about his own desire for authority, offering a quite neat microcosm of the confrontation between Henry’s expansion of power and his courtiers’ reactive protections of their traditional powers. Skelton’s desire for authority is perhaps nowhere more evident than in the complicated and diverse positions from which he criticizes and praises Wolsey, which Walker characterizes as a complete, but also completely explicable, internal reversal of argument.92 These criticisms are most widely recognized in ‘Speke Parott’, ‘Collyn Clout’, and ‘Why Come Ye Nat to Courte’; the analysis here will focus on the first of these anti-Wolseyan selections. The apparent incompatibilities of the stances which Skelton uses against Wolsey (and against his ‘enemies’, more generally) are explained by Blanchard in terms of technique and strategy: Skelton’s ‘poetry is reactive, responsive to specific historical situations which demanded idiosyncratic strategies’.93 Both Walker and Blanchard agree, then, that Skelton’s ‘Speke, Parott’ is best read through a more complex lens than afforded if the work is considered straight critique. Walker explicitly argues that ‘Speke, Parott’ is a bid for 89 Walker, Writing Under Tyranny, 51. 90 Blanchard, ‘Skelton’, 126. 91 Blanchard, 126. 92 Walker, Writing Under Tyranny, 1–3. 93 Blanchard, ‘Skelton’, 127.
Early Verse Position-Taking in the Henrician Court
49
royal attention,94 an example of Skelton ‘indulging in the politician’s art of telling his intended audience what he believes that they want to hear’.95 In this situation, the satirist ‘would have to be extremely careful in treading a path across the minefield which lay between fruitless obscurity and the sort of overconfident iconoclasm which would provoke a disastrous official reaction should he overstep propriety’.96 As a result, Skelton assumes the extra distancing measure of the Parrot as narrator, spending several lines through the poem on the complex characterization of this poetic voice. The resulting voice is, indeed, clear in its condemnations. However, the strong invective that the modern reader sees in the lyric should be mitigated, in Walker’s analysis, by the understanding that Skelton is primarily aiming to please Henry through a critique of a counselor who Skelton believes to be out of favor.97 Even with that belief, Skelton’s position is tenuous enough that he finds it safer to criticize power from the additional remove of the assumed narrative voice. Jane Griffiths complicates Walker’s assessment of the essentially selfserving nature of ‘Speke, Parrot’, claiming that ‘although [Skelton] supported the traditional faction, a close examination […] indicates that his purposes in doing so were radical, rather than conservative’.98 While I would add that Skelton’s self-interest mitigates the radicalism of his positions, Griffiths’s points about his work in shoring up poetic language are nonetheless central to understanding Skelton’s position-taking work. As she argues, the critique of Wolsey is less important than the critique of the pedagogical methods he favors: ‘The new methods of language teaching championed by Skelton’s opponents—with an emphasis on imitation rather than grammar—is treated as analogous to Wolsey’s appropriation of royal authority: both are viewed as attacks on the poet’s traditional freedoms’.99 The disorienting mix of languages in the text offers Skelton an opportunity to simultaneously show off his own learning and disparage the sort of collectors’ approach to linguistic accomplishment that he sees as a flaw of the recent humanist teachings. The effect is summarized in passages like Dowse french of parryse, Parrot can lerne Pronounsynge my purpose, after my properte 94 Walker, Writing Under Tyranny, 60. 95 Walker, 82. 96 Walker, 68. 97 Walker, 67. 98 Griffiths, John Skelton, 13. 99 Griffiths, 13.
50
Gender and Position-Taking in Henrician Verse
With perliez byen, Parrot ou perlez rien With Douch, with Spanysh, my tong can agre In Englysh, to god Parrot can snpple Cryst saue Kyng Henry the viii., our Royall Kyng The red rose in honour, to florysh and sprynge100
The final line emphasizes the connection that Griffiths interrogates: the parrot, as a mimic, can speak these languages without knowing these languages, offering a kind of accomplishment without learning. Thus, when the parrot avows loyalty or fealty to the King, the absurdity is apparent. Just as there is no meaning behind the parrot’s language imitation, there is no meaning behind his speech acts. Against the advancement of mimicry, Skelton offers an implicit argument that ‘to teach by grammatical precept [is] the path to the fluent and interpretative reading necessary for a full understanding of the poet’s apocalyptic warnings, and thus for the possibility of political change’.101 In such a reading, the use of the Parrot’s voice becomes a more straight parody, a distancing choice only insofar as it distances Skelton from those he opposes. Like Blanchard’s arguments, though, this apparent disagreement with Walker works best when instead reconciled: Skelton does want a particular kind of political change, but that desired change largely focuses on a more central role for and greater recognition of his poetic voice, and he attempts to enact that change without abandoning the few protections still afforded his precarious position. Several lines in ‘Speke Parrot’ also connect this empty use of language to specifically poetic projects, suggesting that Skelton perhaps did not approve of the King’s endorsement of the imitative courtly love lyric as a demonstration of learning and nobility. In lines 235 through 262 of the poem, the Parrot, in response to Galathea’s command, produces a fairly vapid love lyric, and one specifically connected to the classical myths that provided so much of the vocabulary of the later Petrarchan imitations and translations. The parody of the courtly love tradition, wherein the speaker produces a lament on the theme of service that is also a tribute to the beloved, is a critique of new humanist imitations of Continental fashions, like Skelton’s critiques of linguistic imitation. This critique also seems more relevant to Henry’s inclinations than to Wolsey’s—a move towards the critical view 100 Skelton, Certayne Bokes, ‘Speke Parrot’, lines 29–35. Following standard editorial work with the poems, I offer the logical substitution of commas in the poem for ‘/’ in the printed text. 101 Griffiths, John Skelton, 13.
Early Verse Position-Taking in the Henrician Court
51
of the King’s more performative and youthful inclinations (inclinations particularly apparent in Henry’s own verse works) that will also be an undercurrent in Skelton’s ‘Against Garnesche’. Skelton’s critiques of Wolsey may also extend to a larger critique of inappropriate influences on the King. Such critiques seem also manifest in the Parrot’s celebration of festive disport and of, apparently, precisely the kind of cloaked verse which this project considers. At different points in the poem, the parrot broadcasts his joy in apparently frivolous pursuits: Parrot, Parrot, Parrot, praty popigay With my beke I can pyke, my lyttel praty too My delyght is solas, pleasure, dysporte and pley Lyke a wanton whan I wyll, I rele to and froo Parot can say, Cesar, aue, also102
The ‘praty’ spelling is itself rewarding, offering the parrot as a pretty and prating subject. His delight in ‘solas, pleasure, dysporte and play’ is not, apparently, endorsed by the poet 103—but the echo to near-contemporary royal works like ‘The tyme of youthe’ and ‘Though sum saith’ indicates the potentially dangerous objects for this disapproval. The final line, here, echoes the empty ‘Cryst saue Kyng Henry’ of the earlier stanzas,104 mitigating the critique of the King, perhaps, but not necessarily of the King’s judgment. Contrasting his own position to the empty vows of prating wantons, Skelton suggests the seriousness of his own avowed loyalty; in privileging seriousness, though, Skelton calls into question the worth of ‘delyghts’ that Henry has himself privileged. Among those delights is the courtly love lyric, parodied within the text for its apparent emptiness, but also necessarily included in the poem’s complicated criticisms of figurative language. The complication is both clear and unreconciled; Skelton is himself participating in the system of ‘confusion’ that the lines apparently condemn: But of that supposicyon, that callyd is arte Confuse distrubytyue, as Parrot hath deuysed Let euery man, after his merit, take his parte For in this processe, Parrot nothing hath surmysed 102 Skelton, ‘Speke Parrot’, lines 106–110. 103 Skelton, line 108. 104 Skelton, line 34.
52
Gender and Position-Taking in Henrician Verse
No matter pretendyd, nor nothyng enterprysed But that metaphora, alegoria with all, Shall be his protectyon, his pauys and his wall.105
The poem outlines, participates in, and denounces the position-taking strategies in verse that are taking root at the Henrician court. ‘Metaphora’ and ‘alegoria’, referenced in the mocking use of multiple languages that Skelton employs through the verse,106 are also the backbone of the poem itself. They are, in fact, Skelton’s ‘protectyon’; when he critiques the distancing strategies of the humanist court poets, he cannot escape an implicit critique of his own work, however much he claims to offer a more forthright vision of appropriate courtly criticism. Skelton’s aim may have seemed less radical from his own perspective, in the early court and knowing the King had been educated in the worth of the courtly poet. David R. Carlson describes Skelton’s effort as one which aimed ‘to define what it meant to be a poet, or what it ought to mean: the poet was parrhesiast, or truth-teller, no matter the circumstantial difficulties’.107 That role was meant to be accepted and traditional; Skelton was part of Henry’s early poetic project, though elements of Henry’s own early position-taking countered the aims of the older poet. The tension between the two may be captured in the contest memorialized in ‘Against Garnesche’, wherein Skelton was ‘called upon by his king […] to take part in a public slanging match at court […] against a person who stood distinctly high in the sovereign’s favour, and who represented his sovereign’s own revelrous social proclivities in ways that Skelton could not’.108 Skelton is typically vitriolic, and his most overt and outrageous criticisms are certainly leveled at his poetic competitor. However, the characteristics of Garnesche attacked by Skelton, and the values he assigns to and uses to bolster his own merit, reflect an early tension between earlier courtly ideals and Henrician performance. Carlson links Skelton’s particular concern with Garnesche not only to the man’s personal traits, but more importantly to the worth which Henry conferred upon the young courtier. To some extent, this is a function of Skelton’s ego, and the performance itself demands a voice which positions itself as superior to the competitor. Criticism of Garnesche, though, is also 105 Skelton, ‘Speke Parrot’, lines 197–203. 106 Skelton, line 202. 107 Carlson, ‘Revels and Erudition’, 240. 108 Carlson, 240. Carlson characterizes the works as a public flyting (240), but further detail is difficult as Garnesche’s half of the exchange did not survive and as Skelton is not always the most reliable self-reporter.
Early Verse Position-Taking in the Henrician Court
53
criticism of the man who values him, in the older tradition of speaking truth to power: Garnesche’s knighthood may be taken to represent the ascendance of the culture of revelry that characterized Henry’s kingship: what the king valued (and so rewarded) was costly, lavish distraction, of the sort that Garnesche helped to supply with his various associates, rather than the martial valour, say, that had ordinarily been recognized with battlefield dubbings to knighthood, or the financial prudence that had characterized Henry’s father’s regime.109
Garnesche was knighted and rewarded partly for his work on entertainment during Henry’s early French campaigns. Thus, even as Henry claimed to prioritize a kind of martial manliness, the holes in his conception were already being revealed and, Skelton implies, exploited by courtiers like Garnesche. While Henry’s verse rebelled against the advisers who would keep him at home, his actions suggested that the performance of the martial was more important than the result. Skelton critiques the King’s investment in performance in a show of resistance to power, but his aims are encoded in the power system as he understands it. While Henry may understand Skelton’s verse as resistant or, in Griffiths’s phrase, radical, Skelton himself may instead present his work as the traditional and natural voice of the critical, loyal poet-courtier. Skelton’s first entry in the series opens with mockery of faux-martial aims and, it seems, of the entire system of allusion and performance based in Arthurian tradition. After addressing his challenger, he follows with a series of allusions to ignoble or apparently invented knights: Sir Temagant, Sir Frollo de Franko, Sir Satrapas, and Sir Chesten.110 The mocking allusive names continue throughout the first poem; Skelton uses the final two lines, ‘But say me now, Sir Satrapas, what authority ye have / In your challenge, Sir Chesten, to call me a knave?’, as a refrain,111 suggesting the particular importance of his reference to Satrapas and Chesten. The knights imply the fictive nature of Garnesche’s own nobility. ‘Satrapas’ may also have the sense of deriving from satrap, perhaps indicating that Garnesche’s worth is only in his toadying to Henry. Again, this name may also call into question the merit of the figure to whom Garnesche pays court, even if the greater criticism is levied on the subservient opponent. 109 Carlson, 247. 110 Skelton, Complete English Poems, ‘Against Garnesche’, I.4–7. 111 Skelton, I.6–7.
54
Gender and Position-Taking in Henrician Verse
In the next entry, the structure is essentially maintained; here, the refrain is ‘Ye capped Caiaphas copious, your paltock on your pate, / Though ye prate like proud Pilate, beware of checkmate’.112 While the comparisons to Caiaphas and Pilate certainly escalate the criticism, these references seem also to reinforce the Satrapas/satrap correlation. Further, the implicit extension beyond these figureheads is even more fraught in this case. The ruler who controls a satrap may be good or bad; the ruler who controlled Caiaphas and Pilate must be either Tiberius—a great military mind and competent ruler—or the infamous Caligula. Although this early in Henry’s reign the latter comparison would not have been necessarily fair, there is also the possibility that Skelton, as Henry’s former tutor, knew enough of his pupil’s temperament to fear the potential aptness of the comparison. However, Skelton only opens up the possibility of this comparison, limiting his overt resistance to the choices that have led to advancement for persons such as Garnesche. Throughout, he prioritizes the choices the King makes, rather than the King himself, as the target for poetic criticism. The third poem maintains Skelton’s more general strategy: he continues to mock Garnesche as a faux-knight and dwells, characteristically, on highly physical imagery and insult. At one point, Skelton diverts into eleven lines of invective containing fourteen epithets against Garnesche: Thou toad, thou scorpion, Thou bawdy babion, Thou bear, thou bristled boar, Thou Moorish manticore, Thou rammish stinking goat, Thou foul churlish parrot, Thou grisly Gorgon gleimy, Thou sweaty sloven seamy, Thou morrion, thou maument, Thou false stinking serpent, Thou mockish marmoset 113
Many of the insults in Skelton’s work seem to prefigure the Orientalizing/ exoticizing tropes that appear in Surrey’s ‘Th Aseirian King’, pointing to the ways that the same atmosphere which fostered Henrician verse also ushered in new forms of English nationalism and racism, as Hall’s Things of Darkness 112 Skelton, II.6–7. 113 Skelton, III.162–172.
Early Verse Position-Taking in the Henrician Court
55
establishes. The adjective ‘Moorish’, here, merits particular note as an adjective that draws on a human identity, yet is part of an explicitly dehumanizing list. The poem also, though, levels some more specific insults against Garnesche. That the young man was as unattractive and unhygienic as Skelton suggests seems unlikely, certainly. Still, Carlson’s history confirms that the less overt insults Skelton lays against his rival are accurate. Skelton includes the biographical detail that: When ye were younger of age Ye were a kitchen-page, A dish-washer, a drivel, […] Ye sluffered up sauce In my Lady Bruce’s house.114
He also references Garnesche’s poor appearance at ‘Guines’ and his attempts to seduce a married lady, one ‘Mistress Andelby’.115 Carlson confirms the essentials of the biography—that is, Garnesche was where Skelton says, such people were also present, and Garnesche did indeed begin his service as a kitchen page,116 albeit likely in the traditional role of a server in a fairly high-ranking household. Skelton then avoids the issue of his own heritage, because ‘he had other claims to esteem […] When Skelton develops such indications as he can of knavery on Garnesche’s part [… ,] these turn out to be matters not of birth so much as of character’.117 In contrast to Garnesche’s low behavior—dressing inappropriately and attempting to seduce married women—Skelton offers his own achievements as evidence of his worth: A king to me mine habit gave: At Oxford, the university, Advanced I was to that degree; By whole consent of their senate, I was made poet laureate.118 114 Skelton, III.24–33. 115 Skelton, III.40 and III.56. Set against Henry’s work to privilege marital unity and stability, the choice of insult also represents a concession to the King’s imposed standards for aristocratic behavior—another balance, in Skelton’s work, of mingled acquiescence and critique. 116 Carlson, ‘Revels and Erudition’, 249–252. 117 Carlson, 248. 118 Skelton, ‘Against Garnesche’, IV. 80–84.
56
Gender and Position-Taking in Henrician Verse
In opposition to Garnesche’s ‘threadbare’ gentleness,119 Skelton offers his ‘habit’ earned by merit.120 ‘Skelton’s alternative to the Thopas-like fancy-dress foolery that he ridicules in Garnesche and Henry’s chamber—medievalistic, violent and dissolute—is classical erudition’.121 He is, self-proclaimedly, ‘advanced’ to his position,122 but that advancement is based on talent recognized by ‘whole consent’ and by a king.123 Skelton bases his worth both on popular consensus and on the judgment of a ruler appointed by divine right. Skelton also opens the poem by saying that he has received the second of Garnesche’s ‘hyime[s]’.124 First, the numbers would seem to indicate that Skelton entered the first volley in the challenge, at a minimum. Skelton’s first mention of Garnesche’s writing is in his third poem; combined with the highly similar structures of the first two entries, this may indicate that Garnesche’s first response came between Skelton’s second and third poems. Second, and most obviously, the numbering indicates that Garnesche is writing less than his opponent. Finally, Skelton wants to note this disparity; he writes it into his verse and opens with it. Thus, he privileges both his chivalric performance as challenger and his linguistic performance as poet laureate, capable of greater production than his opponent. He uses the terms of Henry’s valuation even as he critiques their application, because his goal is to advance within the space that he understands as already existing for his voice. Skelton understands his position as provided by the King, but he also implicitly argues that his position could be furthered. He argues for desert of greater power and for a system in which the rational king acknowledges and rewards desert. Even as Henry aimed to expand his control, the poet-politicians of his court responded by reinforcing their own understanding of their social role as an essential and advisory one, complicated by the contemporary debate over the role of conscience in the law. Skelton participated in the creation of that power, whether or not he fully understood his role as resistant. Although increasingly limited by confining definitions of treason and by Henry’s attempts to define nobility through loyal behavior, court poets worked to create a space in which noble behavior was instead defined through noble use of language, even in opposition to the King. At the beginning of his reign, Henry was to some extent forced to allow advising voices; by the time he 119 Skelton, IV. 70. 120 Skelton, IV. 80 121 Carlson, ‘Revels and Erudition’, 254. 122 Skelton, ‘Against Garnesche’, IV. 82. 123 Skelton, IV. 83. 124 Skelton, IV. 2.
Early Verse Position-Taking in the Henrician Court
57
tried to eradicate them, the social system he had helped shape gave them too much power to permit their complete erasure, as we will see in later chapters. Even in Henry’s last, most desperate years, Surrey’s trial would highlight the acceptance of poetic language as privileged language, never brought before the legal court despite its clear potential as evidence. Nonetheless, canny courtiers aimed to cloak their dissent with some reasonable deniability—a strategy learned from the King’s own compositions. *** In 1533, as he finally succeeded in ending his first marriage and simultaneously redefined the English church and monarchy, Henry ultimately did revise the coronation oath, using the terms he had attempted to insert over two decades before. In addition to this revision, he ‘commissioned a new coronation painting for the palace at Whitehall’, painting over the past with his new terms and definitions and redefining his role even as he tried to overwrite his first marriage.125 When Henry first attempted to revise his coronation oath and was thwarted, he turned to position-taking verse expressions to privilege his own power, gradually accumulating increased power by revising the terms of the social contract. These expressions authorized and activated key elements of the poetic systems of his court, privileging sites of often resistant position-taking embedded in courtly love, chivalry, and youth that would eventually lose both accessibility and appeal for the King. In 1533, when he revised the oath, King Henry VIII no longer needed to rely on coded expression. His courtiers took on that system and used it for criticism of their King and his court, but he himself had become, in word at least, a theoretically absolute power in the land, and versified position-taking no longer served his performance of that power. Henry’s performances of power throughout his reign and the early verse creation in which he positioned that power certainly did have gendered implications. However, this chapter establishes that the King’s verse can be read as primarily concerned not with appropriate masculine performances for all courtiers, but rather with seperating the absolute power of the monarch from the monarch-granted power of the aristocrat. Skelton’s concern with reifying traditional aristocratic power, on more even footing for criticism of and reactions against the King and other advisors, was not only available to, and did not only inform, men’s courtly verse production. Rather, Skelton’s work shows the traditional context from which courtly 125 Hunt, Drama of Coronation, 19.
58
Gender and Position-Taking in Henrician Verse
manuscript producers, aristocratic women and men, would establish the position-taking power of verse, infused with the choices established by the King for privileging and coding such position-taking.
Works Cited Anglo, Sydney. Spectacle, Pageantry, and Early Tudor Policy. Oxford: Clarendon Press, 1969. Blanchard, W. Scott. ‘Skelton: The Voice of the Mob in Sanctuary’. In Herman, Rethinking, 123–144. Brigden, Susan. New Worlds, Lost Worlds: The Rule of the Tudors, 1485–1603. New York: Penguin, 2002. Carlson, David R. ‘The Henrician Courtier Writing in Manuscript and Print: Wyatt, Surrey, Bryan, and Others’. In A Companion to Tudor Literature, edited by Kent Cartwright, 151–177. Malden, MA: Wiley-Blackwell, 2010. Carlson, David R. ‘Skelton, Garnesche, and Henry VIII: Revels and Erudition at Court’. The Review of English Studies 66, no. 274 (2015): 240–258. Clements, Robert J. and Lorna Levant, eds. Renaissance Letters: Revelations of a World Reborn. New York: New York UP, 1976. Crane, Mary Thomas. Framing Authority: Sayings, Self, and Society in SixteenthCentury England. Princeton UP: Princeton, 1993. Erasmus. ‘To Henry VIII’. 9 September 1517. Letter VI-4. In Clements and Levant, Renaissance Letters, 255–257. Erasmus. ‘To Ulrich von Hutten’. 23 July 1517. Letter I-7. In Clements and Levant, Renaissance Letters, 20–25. Griffiths, Jane. ‘Having the Last Word: Manuscript, Print, and the Envoy in the Poetry of John Skelton’. In The Oxford Handbook of Tudor Literature, 1485–1603, edited by Mike Pincombe and Cathy Shrank, 69–86. New York: Oxford UP, 2009. Griffiths, Jane. John Skelton and Poetic Authority: Defining the Liberty to Speak. New York: Oxford UP, 2006. Henry VIII. The Lyrics of the Henry VIII Manuscript. Edited by Raymond G. Siemens. Grand Rapids, MI: English Renaissance Text Society, 2013. Herman, Peter C. ed. Reading Monarchs Writing: The Poetry of Henry VIII, Mary Stuart, Elizabeth I, and James VI/I. Tempe, AZ: MRTS, 2002. Herman, Peter C., ed. Rethinking the Henrician Era: Essays on Early Tudor Texts and Contexts. Urbana: U of Illinois P, 1994. Herman, Peter C. Royal Poetrie: Monarchic Verse and the Political Imaginary of Early Modern England. Ithaca: Cornell UP, 2010.
Early Verse Position-Taking in the Henrician Court
59
Herman, Peter C. and Ray G. Siemens. ‘Henry VIII and the Poetry of Politics’. In Herman, Reading Monarchs, 11–34. Herman, Peter C. and Ray G. Siemens. ‘Introduction’. In Herman, Reading Monarchs, 1–10. Howard, Skiles. ‘“Ascending the Riche Mount”: Performing Hierarchy and Gender in the Henrician Masque’. In Herman, Rethinking, 16–39. Hunt, Alice. The Drama of Coronation. Cambridge: Cambridge UP, 2008. Jeffrey, David Lyle. ‘Courtly Love and Christian Marriage: Chretien de Troyes, Chaucer, and Henry VIII’. Christianity and Literature 59, no. 3 (2010): 515–530. Legg, Leopold G. Wickham, ed. English Coronation Records. Westminster: Archibald Constable, 1901. Letters and Papers, Foreign and Domestic, Henry VIII, Volumes 1–21. Edited by J.S. Brewer, James Gairdner, and R.H. Brodie. London: HMSO, 1862–1920. Loades, David. Henry VIII: Court, Church, and Conflict. Kew: National Archives, 2007. Lockey, Brian. Law and Empire in English Renaissance Literature. New York: Cambridge UP, 2006. Rupp, Susanne. ‘Performing Court: The Music of Henry VIII’. Journal for the Study of British Cultures 12, no. 1 (2005): 105–114. Scarisbrick, J.J. Henry VIII. Berkeley: U of California P, 1968. Siemens, Raymond G. ‘Henry VIII as Writer and Lyricist’. The Musical Quarterly 92, no. 1 (2009): 136–166. Skelton, John. Certayne Bokes, Compyled by Mayster Skelton, Poet Laureat. London: Richard Lant, 1545. Skelton, John. The Complete English Poems. Ed. John Scattergood. New Haven: Yale UP, 1983. Stemmler, Theo. ‘The Songs and Love-Letters of Henry VIII: On the Flexibility of Literary Genres’. In Henry VIII in History, Historiography, and Literature, edited by Uwe Baumann. Frankfurt: Peter Lang, 1992. 97–111. Streitberger, W.R. Court Revels, 1485–1559. Toronto: U of Toronto P, 1994. Walker, Greg. Writing under Tyranny: English Literature and the Henrician Reformation. Oxford: Oxford UP, 2005. Weir, Allison. Henry VIII: The King and His Court. New York: Ballantine, 2001.
2.
Traditions of Resistance and Verse Position-Taking Abstract: This chapter engages Henrician courtiers’ manipulation of historic traditions. Surrey and Wyatt draw on established practices in using kings of classical antiquity as allegorical stand-ins for the reigning king. Henry’s preferences in material self-presentation underline the political verse-making process that takes place when courtiers manipulate his choices of royal precedent as a tool for critique. References to tradition constitute composing choices that actively modify extant material to reflect participation in shared systems of position-taking. The limits of those shared systems are considered through Wyatt’s participation and Surrey’s apparent moments of resistance, as well as through Tottel’s editing and titling practices, which increase access but also assign and contain meaning for new audiences, emphasizing the distinct literacies of courtly position-taking. Keywords: Henry Howard; Thomas Wyatt; Henrician verse; Tudor poetry; Tottel’s Miscellany; Davidic imagery
The changed perception of Henry’s reign over the twenty-five years after he largely stopped writing poetry was such that the opinion that he ‘had degenerated into tyranny was, by 1547, a commonplace […] both at home and abroad’.1 These perceptions ‘had as much to do with the character of the sovereign and the processes of political participation he favored as with the operation of his judicial system’.2 In his early performance as king, Henry emphasized spontaneity, implying accessibility; although the element of power was always present, it was also carefully contained. Perhaps what followed was merely the inevitable course of his original ambition, perhaps 1 Walker, Writing Under Tyranny, 5. 2 Walker, 6.
Quoss-Moore, R.M., Gender and Position-Taking in Henrician Verse: Tradition, Translation, and Transcription. Amsterdam: Amsterdam University Press, 2023 doi 10.5117/9789463723534_ch02
62
Gender and Position-Taking in Henrician Verse
it was the result of several various disappointments, not least his loss of multiple children and the decline in his personal health. Whatever the case, in the 1530s, ‘when Henry, by launching his Great Matter, embarked upon a political journey a sizeable proportion […] of his advisers did not approve of, and which was evidently against the wishes of the mass of his people, he was forced to abandon […] those models of behaviour that gave his rule cultural legitimacy’.3 In light of this public resistance, Henry tried to reemphasize his willingness to accept good counsel; Greg Walker notes, for example, that ‘the proclamation proscribing erroneous books and biblical translation, issued on 22 June 1530 stressed that the King was acting, not on his own initiative but after widespread consultation’. 4 Such emphasis was only necessary in light of his subjects’ increasing anxiety that the King was, indeed, acting on his own initiatives—and in his own interests. This anxiety was in some ways a particularly English problem; Sir John Fortescue’s historic On the Laws and Governance of England takes care to emphasize that ‘the king of England is not able to change the laws of his kingdom at pleasure, for he rules his people with a government not only royal but also political’.5 A contrast is apparent to, for example, Foucault’s more generally European summary of early modern views of royal prerogative and privilege: that ‘the force of the law is the force of the prince’.6 This latter approach, though, was becoming the more typically Henrician view; Henry, later in his reign, tended to see any dissent as a direct attack on his authority and so on himself specifically.7 The oaths attached to the Acts of Succession ‘specifically stated that outward subscription to his claims was not enough: English [citizens] must swear the oaths without inward grudge or scruple’.8 Henry’s increasing inclination towards such control is also evident throughout the lives of the courtiers whose works are considered in this book: in Douglas’s imprisonment, in Howard’s fall, in Surrey’s execution. That English subjects saw these acts as reaching beyond the bounds of royal authority is reflected, then, in the poetry these 3 Walker, 13. 4 Walker, 13. 5 Fortescue, Laws and Governance, 17. 6 Foucault, Discipline and Punish, 47. 7 This is highlighted by the statutes and legal apparatus of the second half of Henry’s reign. As Walker summarizes, ‘The various Acts of Succession of the 1530s, and the 1534 Treason Act […] combined to make it an offence merely to suggest that the King was a tyrant, usurper, or heretic; while the inquisitorial apparatus of the Privy Council and Thomas Cromwell’s secretariat investigated any hint that reached them of sedition gossip or case-putting’ (Writing Under Tyranny, 24). 8 Walker, 24.
Tr aditions of Resistance and Verse Position-Taking
63
courtiers produced—and, given Henry’s own privileging of verse as a space for coded, protected position-taking, in the act of reaching for verse in their reactions, at all. The precedents of courtly poetic production necessarily informed its progress; one thread that winds through those precedents was the utility of comparison to other kings as useful creative territory for critique and consideration of court structures. The central role of precedent to English law, emphasized by the debates on common law mentioned in the previous chapter, helps to foreground the importance that early modern English citizens placed on tradition: on the idea that the past provided examples of the right way of doing things, even when widespread disagreement was common on what the right way was or what the past had been. As John N. King summarizes, ‘Tudor apologists appropriated, transformed, or rejected available cultural codes to create authoritative texts and images that selfconsciously examine the dynamics of political and religious authority’.9 The previous chapter establishes the avenues through which poetry was authorized as one of these cultural codes; also among the most important of these codes, for Henrician courtly makers, were biblical referents. The mythos of biblical kings offered ‘an important source in royal iconography for “correlative types” that recalled events in the history of ancient Israel, as publicists searched for parallels that enabled them to praise Henry VIII as a second Moses or David’.10 These sources were given further primacy as Henry himself invested apparent performative capital in depictions linking him to these Old Testament monarchs—most particularly to David. Early in his reign, Henry’s inclination towards Davidic iconography may have been related to the biblical King’s status as a poet and composer.11 By the 1530s, though, that iconography had become intertwined with the King’s Reformist efforts. King acknowledges that ‘to praise a king as another David or Solomon is not distinctively “reformist”’, as such comparisons recur frequently in medieval works; 12 nonetheless, ‘literary and artistic works that emphasize the status of David and Solomon as direct instruments of divine providence offer a powerful iconographical argument in support of the Protestant monarchs’ disestablishment of the Roman church’.13 9 King, Tudor Royal Iconography, 8. 10 King, 8. 11 See King, Tudor Royal Iconography and ‘Henry VIII as David’; White has also discussed these portrayals in her discussions of Katherine Parr’s contributions to royal image-making, which are considered later in this work (‘Literary Collaboration’ and ‘Royal Iconography’). 12 King, Tudor Royal Iconography, 56. 13 King, 56.
64
Gender and Position-Taking in Henrician Verse
Henry may also have found in David a solution to the tensions between his humanist impulses and his inherently hierarchical understandings of social structure. As Anne Lake Prescott emphasizes in her reading of the conservative energies of early modern Davidic energy, ‘if it is God who lifts the humble we can relish that thought without worrying that the lowly might rise on their own. The stress on God’s choice offers both the pleasant possibility of ascent and the sobering lack of human will in such elevation’.14 Among the many instances of such iconography in Henry’s court, perhaps the one that has drawn the most attention has been Holbein’s title page for the 1535 Coverdale. King offers an overview of the imagery of the piece, in which ‘[t]he commanding figure of Henry VIII wields the Sword and the Book at the base of the title page as a worldly manifestation of divine revelation. His authority “descends” from the Old and New Testament models for sacred kingship depicted elsewhere on the page: Moses receiving the Ten Commandments, David with his lyre, and Christ preaching’.15 Henry also endorsed such images in miniatures, in the visual signifiers of official portraits, in his masques and processions, and, extensively, in the Psalter presented to him by Jean Mallard.16 The appeals of Davidic iconography, especially, would have been numerous: in the parallel that emphasized Henry’s artistic capabilities, in the implicit reminder of forgiveness for a king’s adultery and his related sins, and, of course, in that important endorsement of the King as ‘the intermediary between heaven and earth’,17 sharing the will of heavenly authority directly with his people. In response, the Henrician court filled with similar iconographic work and particularly with references to and versions of the Psalms. Even an overview is beyond the scope of my work here.18 As one example, though, indicating the link between Davidic imagery and comparison of other figures of biblical and classical fame, Henry Parker, Lord Morley’s The Exposition and Declaration of the Psalme, Deus Ultionum Dominus is worth briefly noting.19 The psalm with which Morley works in the exposition is not one 14 Prescott, ‘Sheephook to the Scepter’, 24. 15 King, Tudor Royal Iconography, 54. 16 King, 76–80. 17 King, 59. White discusses the appeal of Davidic imagery for, essentially, royal military propaganda in her ‘Royal Iconography’. 18 White offers one general summary of this Davidic imagery in her ‘Royal Iconography’ (554). 19 Morley’s valorizing work here is either especially unexpected or especially tragic in its historical context. The title page of Morley’s exposition reads ‘1534’, apparently as a date of publication. The Oxford DNB, though, and the book’s listing on EEBO indicate a publication date of 1539. In 1536, Henry executed Morley’s son-in-law George Boleyn; six years later, in 1542, Morley’s daughter Jane Parker was executed for her part in Katherine Howard’s affairs. Notably,
Tr aditions of Resistance and Verse Position-Taking
65
usually attributed to David, though the larger association of psalters with David’s king/poet voice is nonetheless relevant. Morley’s primary theme is in the interpretation of the psalm as biblical support for Henry’s break from the Catholic Church; in his exposition, his justification frequently rests on comparisons of Henry to righteous kings from history—or, at least, to those who Morley interprets and presents as righteous. At several points in the exposition, Morley offers quite explicit comparisons between Henry and ruling figures of antiquity, including Judith.20 Morley’s work is not metaphoric or veiled; even in the potentially trickily gendered comparison to Judith, his praise of Henry is always explicit and explained. England will say of Henry, as the ‘Jewes dydde to Judith: You are our beautie, you are oure honour, you are our glorie’.21 The King is always as good as, or better than, his many royal and right predecessors; ‘Scipio the Affrican dyd moche for the Romayns, Codrus for the Atheniens, Epaminondas moche for the Thebans, Themistocles moch for the Grecians, Cirus moche for the Persians, Salandine moche for the Egyptians, and yet all these compared with your hyghnes, may seme almost to have done nothing at al’.22 When Morley wants to explicate the goodness of Henry and the correctness of the path on which he leads the country, he turns to the most orthodox comparisons by Henry’s own definitions: a king should be compared to other kings, because, as a fundamental point in Morley’s exposition, a king is different from other men. In its comparisons, the text offers several specific connections to David, likely informed by Henry’s preferences. Among the most extended and explained is the passage ‘Lyke as the excellente kynge and prophete David, greatly meruaylynge, dydde deamunde, who shulde ryse with hym, to subdewe euyll doers, workers of wyckednesse, so may our kynge saye, who ought not to rise with me, to the vanquishyng of this monstruous hydra’.23 Here, Morley’s text takes on a process of definition not dissimilar to the work of Henry’s early verse. The King used his poetry to define noble, courtly behavior as that which aligned with his iconographic projects of courtly though, there are no indications that Morley ever acted against Henry; indeed, Morley ‘was the only peer to have an unbroken record of participation’ in the trials of peers in Henry VIII’s reign, including those of his family members (Oxford DNB). 20 Morley, Exposition, 3r. White offers contextualization for this choice in her overview of the ‘war-related prayers’ in William Tylotson’s Prayers of the Bible (1544), which included a ‘Prayer of Judith’ (‘Royal Iconography’, 560–562). 21 Morley, 3r. 22 Morley, 3r–3v. 23 Morley, 16r.
66
Gender and Position-Taking in Henrician Verse
love; Morley’s exposition defines appropriate behavior in the English subject as aligned with royal iconographic projects connecting Davidic imagery to the establishment of the Church of England—and, as Micheline White’s work establishes, to Henry’s policy agendas more widely.24 Morley offers some particular efforts in clearly aligning this imagery with specifically English concerns, as when God defends (among a long list) ‘ David frome Golyas, […] Henry the fyfte frome the frenche men, the wyse Henry the seventh, from the tyrant kynge Rycharde’.25 Morley asks that God offer that same defense to Henry. While Morley’s exposition is just one example, it is nicely representative of courtly work that endorsed Henry’s Davidic imagery unironically and that promoted the royal project in placing David within an English royal tradition. Moreover, Morley’s work explicitly places Davidic comparison alongside other biblical, historical, and classical precedents, demonstrating the associative networks of comparison that could grow from Henry’s own choices. We will return to courtiers’ position-taking work with Davidic imagery and the Psalms in the fourth chapter, here, where I consider the authorizing function of this endorsed imagery in both Katherine Parr’s royally endorsed publications and Wyatt’s potentially more resistant verse. For Wyatt and Parr, as for early modern English writers, more generally, ‘dealing with David and his sin is an early modern mode of poetical and political analysis’.26 Their entries in that analytical tradition are characteristically dense. Surrey’s biblical work with the Psalms and Ecclesiastes are, equally characteristically, at an extreme on the available spectrum of tribute and critique. Walker offers an excellent reading of these works,27 and I am less interested here in revisiting or revising that explication than in exploring the relationship between these visions of David and, in this chapter, Wyatt’s and Surrey’s distinct approaches to the strategies both men would have learned in the courtly communities of manuscript production. Both poets place David within a ‘poetics and politics of royal guilt, blame, and restoration’,28 but Surrey’s vision is less nuanced and makes less use of potential deniability or distance. To summarize, Walker identifies ‘the references to princely acts, the building of conduits and water-courses, and to delicious gardens 24 White’s work includes an overview of some of Parr’s figurations of David, which she shows were created in collaboration with Henry (‘Royal Iconography’, 573). White’s analysis is fundamental to the work with Parr in the fourth chapter, here. 25 Morley, Exposition, 5v. 26 Kilgore, ‘Politics of King David’, 413. 27 Walker, Writing Under Tyranny, 400–412. 28 Kilgore, ‘Politics of King David’, 413.
Tr aditions of Resistance and Verse Position-Taking
67
and royal sports [as deliberate echoes of] Henry VIII’s current building plans’.29 These are, to Surrey, all ‘part of the same grotesque royal will to self-indulgence’.30 When Surrey offers translations on biblical kings, his themes are on their greater culpability, rather than on their elevation or righteousness: Surrey’s Solomon, like David, was […] first and foremost a royal sinner, and it was his royal office that gave his sins their political importance. And it is hard to avoid the conclusion that it was Henry VIII’s example that the poet had in the forefront of his mind as he wrote: that he was indeed recklessly taunting the King—among the small circle who read his poems in manuscript at least—with his own failings, and defying him to redeem himself.31
That such a conclusion is hard to avoid typifies the distinction between Surrey’s and Wyatt’s works. Both poets combine elements of the position-taking energy available through poetic coding and the authority offered by biblical referent. In Wyatt, a specific conclusion is often difficult to pin down, rather than difficult to avoid; his work is, in this way, more nearly aligned with the coding strategies of the larger manuscript tradition, where audiences and writers create meaning together. Surrey takes the resistant energies of that tradition and instead aims to move them into more aggressive and overt forms of critique, forms which tend to prioritize a more individualized and authorial voice. As shown in the first chapter, Henry’s investment in courtly love lyrics eventually offered greater authority to his courtiers’ versified visions of service; in both their revision of the courtly love lyric and in their application of classical and biblical precedent, courtiers met the King on the metaphorical grounds he laid out, but they used those images very much to their own purposes. Siemens says of courtly verse that ‘the presentation and circulation of Henry’s lyrics in […] a public arena rendered their sentiments readily identifiable targets for anti-court satire’.32 Henry’s later iconographic projects were even more public. Henry’s iconographic self-comparisons codify David, specifically, but also classical kingship, more generally, as symbolic of his reign. Like the systems of the courtly love lyric, these symbols 29 Walker, Writing Under Tyranny, 402. 30 Walker, 403. 31 Walker, 401. 32 Siemens, Lyrics, 14.
68
Gender and Position-Taking in Henrician Verse
then become particularly useful for courtly critique; this use of specific figurations and systems against the same authorities that have imbued them with power is a definining feature of courtly verse position-taking. When, then, courtiers explicitly utilize kingly comparisons as critique or condemnation, rather than as valorization, the energies of such comparisons have more potential to move beyond the more coded iterations of verse position-taking and into open resistance. Moreover, because of the greater ambiguity of the tradition, the use of classical kings to critique Henry may be a more dangerous strategy, with less potential to obscure the writer’s agency in the interpretation.
A. Wyatt and Contextual Position-Taking Likely the most dangerous of Wyatt’s poems are so because of their context as much as their content; the clearest example may be ‘Who lyst his welthe and eas Retayne’. Considering this work within its ostensible frame as a ‘translation’ from Seneca’s Phaedra centers the potential of the reference work of the poem. While the court poets often have quite loose definitions of translation, though, Wyatt drops the tenuous translation work after the first two stanzas.33 Here, the reference is less clearly between Henry and one specific, well-known king and seems, more broadly, to offer a comparison of the treacheries of the Henrician court with the instabilities of this GrecoRoman precedent. Nonetheless, the poem does offer a potential reading of Henry as a Theseus figure, misled by a self-preserving Phaedra into attacking those loyal to him. Potentially, the poem works as a preservative act by Wyatt—whose own complex relationship to Henry’s affairs makes his voice readable not only as the commenting Chorus of the original, but also as the partially innocent Hippolytus, caught up in the fall of a lustful woman.34 However, this reading only works so far, as Hippolytus can also be read as partly guilty, in his willingness to step into his king-father’s rightful place. The broader indictment of the court atmosphere as inherently, invasively corruptive may nonetheless work to offer a sort of mitigation of the crimes committed under its influence. For Wyatt’s own self-preservation, though, the important balance lies in countering the apparent condemnation of the 33 Muir and Thomson, eds., Collected Poems, 415. 34 This reading might be especially available to Henry and his ministers, who, in the aftermath of the trials and executions around Anne Boleyn, had received letters from Wyatt detailing her lustfulness and his own lack of blame, one of which is discussed in the conclusion here.
Tr aditions of Resistance and Verse Position-Taking
69
thundering throne against the sense of mitigation for the King’s actions: Theseus-like, Henry responds to the corrupt actions of courtiers who ought to have accepted their subject roles and stayed within the bounds of ‘natural’, courtly behavior. The contemporary context opens up readings which are important for their plausibility to Wyatt’s original audience, even when those readings are not the simplest available. Reading the poem in the specific context of the contemporarily recent Treason Act of 1534 highlights a more cautious tone than a more generally political approach acknowledges. Speaking of what has been lost, the poetic voice laments, ‘My lust, my youth dyd then departe, / And blynd desyre of astate. / Who hastis to clyme sekes to reuerte’.35 Negative energy towards the thundering throne is not entirely dispersed in a more conservative, more general reading, but the stanza also emphasizes the injudicious haste of the climber. The speaker has lost something, but the emphasis is not clearly on injustice; there is also an implication the climber has set value on the wrong things. The quintain that most firmly places the ‘translation’ in Wyatt’s contemporary context also further complicates the placement of blame: The bell towre showed me suche syght That in my hed stekys day and nyght; Ther dyd I lerne out of a grate Ffor all vauore, glory or myght, That yet circa Regna tonat.36
The reference to the Tower, echoing the strokes of the recent executions, invites a reading of the poem as among Wyatt’s most scathing remonstrances against Henry and his system of justice. The reiteration that favor, glory, and might belong only to the monarch does partly level a charge of inconstancy; the sense that favor can be withdrawn, perhaps capriciously, is underscored by the willful connection to contemporary events. However, the emphasis also serves to highlight the incautious behavior of those subjects who would attempt to claim power or privilege rather than acknowledge their dependency. The connection to Phaedra underscores the culpability of the King’s subjects. In this less conventional reading, the last lines of the poem are the easiest to read in Henry’s favor. When the speaker warns that ‘Wyt helpythe not 35 Wyatt, Collected Poems, ed. Muir and Thomson, 187–188, ‘Who lyst his welthe’, lines 12–14. 36 Wyatt, lines 16–20.
70
Gender and Position-Taking in Henrician Verse
deffence to yerne, / Of innocence to pled or prate. / Ber low, thereffor, geve god the sterne’,37 he does not make any reference to the validity of the pleas for defense or of innocence. In fact, dependence on wit or self-defense is set in opposition to giving control to God. By aligning the avoidance of royal displeasure with a surrender of control to God, the speaker implies a connection between royal and divine favor.38 To return to the classical reference, Phaedra, Hippolytus, and the nurse are all both guilty and not culpable. Phaedra’s unnatural lust is explicitly paralleled with her mother’s, a fate ordained by the gods. Hippolytus blames himself for his stepmother’s lust, which he cannot actually control, while also being blinded by his misogyny, which he can. Phaedra’s nurse is the most clearly damaging actor in the play—encouraging Phaedra, scheming to condemn Hippolytus—but she also acts to perform the will of her queen.39 Particularly in the context of the ongoing negotiation of appropriate courtly performance, this last set of action and motivation presents some difficulty. If the role of the ideal Henrician courtier is to perform their monarch’s will, rather than to critique or guide, then Phaedra’s nurse may be partly correct in her actions: she does as her mistress wills and acts to protect her mistress. But she can only ever be partly right, because her actions work against the King and against the appropriate social order. When Theseus acts to restore that social order, this reading would run, he is correct in his punishments, regardless of the excuses that might be available to those being punished. Of course, such a reading is hardly the most typical or the simplest interpretation of this particular lyric. The importance of the reading, though, is that it is possible. Wyatt’s first strategic choice is to cast the lyric as a translation, however loose; he mitigates that distancing technique, though, by the fairly specific contextual atmosphere of the poem. Equivocation then becomes his best defense, if he wishes to create a work that speaks against tyranny without inviting any inescapable consequences. According to the Treason Act, to even imply that Henry VIII was a tyrant was to invite death. However, to imply that all subjects should stand in dread of royal power is not necessarily to level a charge of tyranny. Henry himself, after all, was working to emphasize the importance of acknowledging his absolute power. While Wyatt’s fairly frequent imprisonments indicate that the energy of the 37 Wyatt, lines 22–25. 38 The series of layered associations between surrender to God and appropriate courtly service or kingly behavior pulls on similar threads as Wyatt’s work in the Penitential Psalms and to Parr’s frequent comparison, if not conflation, of Henry and God, both further considered in Chapter Four. 39 Seneca, Phaedra.
Tr aditions of Resistance and Verse Position-Taking
71
less charitable reading was perhaps accessible, enough creative negotiation exists to not only excuse the poet, but even to keep him in royal employ. This balancing act may explain why some of Wyatt’s most leveled critique of Henry gained its authority not because of the poet’s own strategies, but because of the projects of his admirers after his death. W.A. Sessions analyzes Surrey’s participation in Wyatt’s literary canonization as an attempt to privilege the power of poetic language; Walker’s similar analysis focuses on those privileging efforts as work to create a political tool. Sessions argues that special authority for Wyatt and Surrey in early modern England results from the ‘particular interventionist act’ of Surrey’s publication of his elegy on Wyatt and his support of Leland’s 1542 Naeniae.40 Sessions characterizes Surrey’s and Leland’s responses to Wyatt’s death as attempts to articulate the recent changes in their social world. 41 Specifically, Surrey wants to claim a special authority for poetic language as language that authorizes and mobilizes ‘noble’ values. 42 That ennobling poetic language may originate in, and be centered on, the court, but it both transcends the court and moves beyond it into the wider world of England. 43 As characterized by Surrey’s efforts, ‘Wyatt’s labor is […] poetic as well as courtly and political; and only the two labors of text and court, language and politics in dialectic, could build new communities’. 44 Through specifically contextualizing Wyatt’s work as reformative, Surrey both privileges his own poetic voice and reaffirms the political and cultural worth of poetic voices, generally.
B. Surrey and Self-Authorization Walker argues that Surrey then moved from this valorization into the potential it opened up to legitimize his own projects. He advances the claim that ‘like Wyatt, Surrey deplored the increasing despotism of Henry’s reign and the moral and physical decline of the King himself […] Unlike Wyatt, however, Surrey was in a position to do something more substantial about them’.45 A qualification is useful, here: history shows us that Surrey certainly thought he could ‘do something more substantial’; whether he would have met with more success had he been less driven by personal ego is an open 40 Sessions, ‘Surrey’s Wyatt’, 169. 41 Sessions, 173. 42 Sessions, 175. 43 Sessions, 176. 44 Sessions, 189. 45 Walker, Writing Under Tyranny, 392.
72
Gender and Position-Taking in Henrician Verse
question, but he was not, perhaps, any more efficacious in his resistance than was Wyatt. In fact, the same position that allowed Surrey to immortalize his and his fellow courtier-poets’ resistance likely contributed to his personal doom. Surrey’s proclivity to emphasize his own importance was not unlike Henry’s, and in the early to mid-1540s he embarked on a series of projects based on his own importance, including an ostentatious house emphasizing royal heraldry in the glasswork and ‘grand, self-glorifying portraits’. 46 The result clearly reflects the dangers of straying too far into violation of Henry’s new laws and social codes; as Walker laconically states, ‘such ostentatious displays of personal and familial ambition would have been dangerous at any time, but in the last years of an obviously dying King they were particularly ill-judged’.47 This particular series of poor judgments would cost Surrey his life. Surrey’s arrest and trial were an exercise in the further performance of royal power, a ‘symbolic humiliation’ in which Surrey was pointedly subjected to public display and denied the usual rights of the aristocracy.48 We can apply, here, Foucault’s articulation of the relationship between the political/personal body of the king and the political/personal body of the condemned;49 the goal of the public trial is to eradicate the potential threat to the body politic of the diseased, traitorous, condemned body. Surrey’s trial illustrates Henry’s embrace of this philosophy, however he might have articulated that understanding; the trial and punishment indeed ‘aim[ed …] not so much to re-establish a balance as to bring in play, as its extreme point, the dissymmetry between the subject who has dared to violate the law and the all-powerful sovereign who displays his strength’.50 If Surrey had aimed at treasonous levels of self-importance, the solution was to erase his aristocratic importance entirely and so to reactivate the King’s power.51 In this case, the reactivation of power was hardly as complete as it might have otherwise been; within ten days, the King who had ordered Surrey’s execution was dead himself. The precise charges brought against Surrey—and limitations therein— again emphasize the complicated question of the privileges afforded to poetic voice in Henrician England. Despite the clearly transgressive content of much of his work and its wide availability, ‘in all the testimonies and all the evidence brought against Surrey at his trial in 1546, at no point did 46 Walker, 383. 47 Walker, 383. 48 Walker, 385. 49 Foucault, Discipline and Punish, 28–29. 50 Foucault, 48–49. 51 Foucault, 59.
Tr aditions of Resistance and Verse Position-Taking
73
anyone allude to his poetry’.52 Walker further points out that ‘his poetic manuscripts [… and] the printed text of his elegy for Wyatt [offered] ample material from which to fashion charges against him’.53 Certainly, the elegy on Wyatt was published and read; further, the prevalence of Surrey’s work in Songes and Sonettes, a decade later, would seem to belie any particular dearth of his work or difficulty of access. Though a comprehensive list of the manuscript copies of his work exceeds the scope of this chapter, there is ample evidence that Surrey’s lyrics were widely circulated.54 The answer to the absence of such evidence in Surrey’s trial, under these circumstances, may rest in the still present, if increasingly tenuous, privilege of poetic voice even in the late Henrician court. Surrey had, at earlier points in his career, demonstrated an understanding of the balance that allowed for the protection of verse position-taking. His Aeneid and his references to Troy in ‘So cruel prison’ offer some of his more restrained potential uses of comparison as critique. When Surrey contextualizes his work in the Virgilian tradition, though, his references to royalty tend to legitimize his own position, as explored in the fourth chapter. Priam may be the King of the impressive and royal Troy, but Troy has fallen. While the implicit threats to Henry are still potentially quite dangerous, Surrey’s focus seems more on self-valorization—a valuation that undermines Tudor legitimacy, but one whose focus is not necessarily on Henry’s worthiness. The biblical translations and paraphrases, instead, seem to focus the question of legitimacy on the worth of the reigning king, and a similar focus underlies references to other historic kings in Surrey’s corpus. Among Surrey’s apparent use of kingly references to critique Henry, ‘Dyvers thy deathe’ demonstrates both, on the one hand, Surrey’s knowledge of the meaning-making community in which he was taking part and, on the other, his drive to push the strategies of that community beyond the bounds of protection. Echoing the repetitions and linked motifs of manuscript productions, the poem offers strong reference to Wyatt’s ‘CEsar, when that the traytor’; Tottel, strikingly, places the work directly after a poem on Wyatt’s David.55 The placement reinforces the established practice of 52 Walker, Writing Under Tyranny, 386. 53 Walker, 386. 54 May reflects, of Wyatt and Surrey, ‘that manuscript collections of their work circulated widely enough that f ive different printers had obtained and published some of their canon before 1557’ (May, ‘Popularizing Courtly Poetry’, 421). A similar point is made in May and Marotti (‘Manuscript Culture’, 86). 55 Tottel, fol. 16v. Wyatt’s ‘Ceaser, whan the traytor’ is also copied in the Devonshire (70r), further indicating its place in the communal position-taking of the manuscript tradition.
74
Gender and Position-Taking in Henrician Verse
kingly-comparison-as-critique and the communal nature of both critique and practice. Surrey’s specific work in ‘Dyvers thy deathe’, though, moves to condemn the larger community and to cement one interpretation from the more flexible visions of multiplicity in Wyatt’s work. Where Wyatt’s Ceasar, as explored in the following chapter, is presented in complicated juxtaposition alongside nuanced presentations of famously feigned emotion, Surrey’s reference is less open to multiple interpretations: ‘Some, that in presence of thy liuelyhed / Lurked, whose brestes enuy with hate had swolne, / Yeld Ceasars teares vpon Pompeius hed’.56 Where Wyatt balances this reference with his lines on Hannibal, Surrey instead offers only a more ghoulish group who ‘watched with the murdrers knife, / With eger thirst to drink thy giltlesse blood’.57 In the first few lines, the poem moves to indict those who falsely mourn Wyatt; ‘it aims both to excite tears and to incriminate by tears: apparent distinctions between diverse ways of grieving for Wyatt collapse, as both groups of “some” in the first two quatrains turn out to be identical’.58 Both groups are then contrasted with Surrey’s noble ‘I’, who ‘knewe what harbred in that hed: / What vertues rare were temperd in that brest’59 and who mourns Wyatt ‘As Pyramus dyd on Thisbes brest bewail’.60 Simpson calls this ‘a suicidal poem, not only because the reference to Caesar implicates the king among those falsely mourning Wyatt, but more powerfully because the last line expresses a powerful, even erotic desire for suicide’.61 Surrey’s insistent ‘I’, though, resists the potential self-erasure implicit in suicide. Linked, here, are Surrey’s anxieties about production in a communal code and his projects of promoting individual genius and worth. These verses are printed in 1542, moving several of the strategies of manuscript court poetry into the public eye and, in so doing, changing the terms of coded poetic creation.62 On one hand, ‘Surrey’s sonnet unquestionably reveals the self-fashioning subject, since the courtiers who weep for Wyatt are revealed 56 Surrey, Songs and Sonettes, ed. Tottel, fol. 16v, ‘Dyvers thy deathe’, lines 2–4. 57 Surrey, lines 5–6. 58 Simpson, ‘Henrician Ovidianism’, 326. 59 Surrey, ‘Dyvers thy deathe’, lines 9–10. 60 Surrey, line 14. 61 Simpson, ‘Henrician Ovidianism’, 326. 62 The appearance of Surrey’s work in print may have additional resistant potentials in context of the 1542 Religion Act. Though ostensibly focused on eliminating false doctrines, Simpson’s analysis of the Act’s scope outlines its reach beyond religious texts; according to the language of the Act, ‘discursive legitimation is wholly a matter of the king’s arbitrary desire’ (342).
Tr aditions of Resistance and Verse Position-Taking
75
as actors in a theatrics of grief with its own complex and deceptive code’.63 In this way, the poem functions within the traditions of position-taking in manuscript production. However, Surrey is less interested in accessing the potential of that complex code and more interested in exempting himself from the multiple interpretations that have become entwined with courtly social performance and with courtly verse production—that is, from what he often simplifies as deception. As Simpson argues, by the final lines of the poem, ‘the only true tears are those of Surrey himself, whose act of publishing a statement of communal grief for Wyatt recoils into a repudiation and exposure of communal grief’.64 There is some irony in the ways that Surrey’s insistence on his own fitness to interpret and mourn Wyatt erases the nuance of Wyatt’s own work; Surrey’s focus on his own poetic project in fact obscures the earlier poet’s accomplishments, replacing the complex balance of Wyatt’s Caesar with the easily condemned figure of Surrey’s reference. That tension and erasure are, though, typical of Surrey’s determination to prioritize his individual voice, which often comes at the expense of communal expression and often refuses to engage fully the position-taking strategies in which the courtly poet necessarily takes part. Surrey’s works are often condemnatory of the larger public and of the communal codes of courtiership, but they are, most dangerously, ‘shot through with barbed allusion to the falsity and cruelty of kings’.65 Walker argues that Surrey could not for long imagine iniquity of any sort without seeing before him the image of his King, the antithesis of all his own ideas of true nobility and masculinity. Hence, his sonnet on the suicide of Sardanapalus, that most conventional of tyrants, seems motivated by a particular loathing for a king far closer to home, not least in its stress on the subject’s inaptness for military exploits.66
That stress merits particular consideration alongside Surrey’s martial imagery in, for example, his translation of ‘Loue, that liueth, and raigneth in my thought’, with its tense depiction of a conquering but cowardly love.67 ‘Th Aseirian King’ offers no mitigating visions of successful military incursions 63 Simpson, 327. 64 Simpson, 326. 65 Walker, Writing Under Tyranny, 400. 66 Walker, 409. 67 See Chapter Three.
76
Gender and Position-Taking in Henrician Verse
to balance the object’s cowardice. This king is an irredeemably unfit lord who ‘in warre that should set princely hartes on fire: / Did yeld, vanquisht for want of marciall art’.68 Surrey’s images of the King’s cowardice are functionally all-encompassing, as is his condemnation. The ‘should’ of the third line indicates the implicitly prescriptive element of the poem. Surrey’s vision of appropriate noble behavior is as commanding as that of Henry’s early verse, but he defines that behavior almost entirely by negative contrast. For this King, The dint of swordes from kisses semed strange: And harder, than his ladies side, his targe: From glutton feastes, to souldiars fare a change: His helmet, farre aboue a garlands charge.69
He is ‘Feble of sprite, impacient of pain’.70 Because of all these failures, he is barely a man. Implicitly, the ‘better’ monarch—the better man—would be well-acquainted with swords, shields, helmets, and campaign food; he would bear up under pain and be bold or strong or determined of spirit. Surrey insistently masculinizes and militarizes his vision of a ‘good’ king, suggesting not only his tensions at performing courtly service, but, moreover, the potential misogyny of his resistance to understanding his own work within the manuscript system of coded position-taking. The subtlety that characterizes that system is essentially abandoned in this poem. The only ‘solution’ the poem proposes, once a king has ‘lost his honor, and his right’71 is the ultimate act of self-erasure; the only ‘manful dede’ remaining to this king is his suicide.72 The counter-balance of the suicidal vision that Simpson sees at the close of ‘Dyvers thy deathe’ might initially seem to offer potential mitigation of this harsh sentence,73 but the dramatic differences of the poems undercut that mitigation. First, and not to be dismissed, is Surrey’s pattern of self-exemption; he is, as he understands and presents it, different from other people and subject to different rules and interpretive parameters. Moreover, Surrey’s ‘I’ in ‘Dyvers thy deathe’ contemplates suicide only metaphorically, and even the metaphor stops short of Pyramus’s final action, imaging only his mourning 68 Surrey, Songes and Sonettes, ed. Tottel, fol 17v, ‘Th Aseirian King’, lines 3–4. 69 Surrey, lines 5–8. 70 Surrey, line 11 71 Surrey, line 12. 72 Surrey, lines 13–14. 73 Simpson, ‘Henrician Ovidianism’, 326.
Tr aditions of Resistance and Verse Position-Taking
77
cries. Finally, that mourning is called into being by the failures of others—a noble reaction to the ignobility that surrounds the speaker, rather than an ignoble end that offers the only escape from one’s own failures. In his vision of a failed and slothful king whose only further potential is death, Surrey moves well beyond the coded, deniable position-taking of manuscript production, generally, or of Wyatt’s similar work in translation and allusion, specifically. A further analysis of the extreme strategies of dissent which Surrey employed can be offered through his vitriolic ‘Eche beast can chose his fere according to his mynde’, which a recent edition of Tottel’s characterizes as ‘perhaps his most dangerously rebellious poem […] certainly his angriest’.74 Walker follows likely the widest contemporary court reading of the poem in analyzing it as a volley in a continuing power struggle between the Howard and Seymour factions.75 The incident apparently recounted is Lady Anne Seymour’s refusal of a dance; Surrey’s response in verse is characteristic of his increasingly volatile responses to any perceived slight to his honor. As Walker characterizes Surrey’s actions of the 1540s, ‘he seems both to have sought desperately to win back a role at the political centre and paradoxically to have signaled his resentment at that exclusion with ever more self-marginalizing acts of bravado’.76 The poem in question, though, may be an even more extreme act than Walker argues. Certainly, Surrey is unlikely to have taken Lady Anne’s refusal with any uncharacteristic grace. However, the poem also reaches beyond the anger of this particular noble feud. If taken in the tradition of lyrics where the lover’s resentment of the lady is analogous to the courtier’s resentment of the sovereign’s control, the poem matches Surrey’s oft-expressed anger at his perceived devaluation within Henry’s court and by Henry, specifically. Opening with the story of the wolf’s rejection of the lion’s chivalrous selfpresentation, the speaker then characterizes the lion’s reaction: ‘I might perceiue his noble hart much moued by the same. / Yet saw I him refraine and eke his wrath aswage, / And vnto her thus gan he say when he was past his rage’.77 The lines that follow, however, do not sound like the product 74 Holton and MacFaul, Tottel’s, 379. 75 Walker, Writing Under Tyranny, 388. The lion represents Surrey himself, following the Howard heraldry; the wolf represents the Lady Anne Seymour, Countess of Hertford, Anne Stanhope before her marriage. The wolf characterization rests both on the Stanhope’s wolf badge and her Seymour husband’s Wulfhall residence. 76 Walker, 381. 77 Surrey, Songes and Sonettes, ed. Tottel, fol. 14r–14v, ‘Eche beast can chose’, lines 24–26.
78
Gender and Position-Taking in Henrician Verse
of assuaged wrath; instead, they are loaded with implied threats and the outrage of wounded dignity: How can ye thus entreat a Lion of the race, That with his pawes a crowned king deuoured in the place: Whose nature is to pray vpon no simple food, As long as he may suck the flesh and drink of noble blood.78
If no privilege existed to protect the poet, or if Surrey abandoned either of the doubled pretenses of animal vision and lyric on rejected chivalric love, these lines would be a death warrant. Although the ‘crowned king’ mentioned is almost certainly the Scottish James IV,79 the implicit threat to Henry is strengthened by the fact that this King had been his brother-in-law, even if it was at the head of Henry and Katherine of Aragon’s army that Surrey’s grandfather gained this triumph. The poem further ennobles the family history as the speaker recounts and romanticizes the fate of Surrey’s uncle, Thomas Howard, in the lines ‘for loue one of the race did end his lyfe in woe / In tower strong and hie for his assured truth’.80 In privileging his relative’s choice, Surrey does not directly confront royal authority, but the challenge is certainly implicit. Given the cultural importance of the Douglas-Howard affair to the court, as demonstrated by the Devonshire Manuscript and explored in the last chapters here, the challenge is likely also easily understood by most of Surrey’s audience. Within this context of challenge that expands beyond Seymour to royal authority, the condemnation of much of the poem takes on a riskier aspect. Condemning the wolf’s haughty response, the lion argues that ‘I seke my foes: and you your frendes do threaten stil with warre’.81 Given the King’s increasing paranoia and his violent responses to any perceived threats, the charge works far more clearly if leveled at Henry than against the wife of a member of an opposite faction, even accepting the metaphor for the keen insult to Surrey’s not inconsiderable pride. Finally, in keeping with the implied threat of the poem entire, the lion ultimately tells the wolf that he ‘shall be glad to fede on that that would haue fed on me’.82 Such invective reflects Surrey’s greatest strength and 78 79 80 81 82
Surrey, ‘Eche beast can chose’, lines 29–32. Holton and MacFaul, Tottel’s, 379. Surrey, ‘Eche beast can chose’, lines 35–36. Surrey, line 48. Surrey, line 68.
Tr aditions of Resistance and Verse Position-Taking
79
weakness as a poet and politician: although he distances himself from the content with some minor strategies, he still refuses to restrain his voice in the ways demanded by Henry’s increasing strict legal and social codes. The legacy of manuscript poetry in print collections, particularly in the influential Songes and Sonettes, showcases the successes and failures of Surrey’s approach to the position-taking strategies he learned at court. The poems in Tottel’s survive, where Surrey did not, but they survive in a form that often mitigates their resistant energies, as considered in the following chapters. I conclude this first section with this poem because Surrey’s function as an exception helps to define the parameters of position-taking verse work in the Henrician court. Surrey’s use of allegory activates even riskier spaces than the other traditional tropes, as with those of biblical precedent or of historical kings. Certainly, this is a less protected space than that of translation or transcription; in allegory, Surrey the poet clearly assigns meaning, and so he cannot claim the same deniability available in the communal productions of, for example, the Devonshire. Like Henry himself, Surrey moves away from many of the preservative strategies used by other courtly poets; he instead privileges his own authority and exceptionalism. Nonetheless, he relies on similar spaces of reader and writer interaction to create meaning and energy for his work. Not least, Surrey seems to explicitly reference his departed uncle, a f igure who frequently activates the communally created interpretive potential of verses considered throughout this work and who, himself, participated in this tradition. Surrey’s work also exemplifies the shifts from the courtly context to the broader access of print; Tottel’s editorial title of ‘A song wrtiten by the Earle of Surrey by a Lady that refused to daunce with him’83 simultaneously offers a clear example of an attempt to ‘open the closed communications of an elite to a wider audience’84 and depoliticizes the content of the verse. 85 Even as Tottel preserves the work and expands access, likely contributing to our own access to Surrey, today, he also begins the process of assigning and containing meaning. When we push Surrey back into the contexts which he understood and worked within, we must restore the role of the reading and writing community, and so open up the broad, shifting, and multiplicituous interpretations through which verse position-taking thrived. 83 Tottel, ed. Songes and Sonettes, 14r. 84 Marotti, English Renaissance Lyric, 196. 85 Warner, Making and Marketing, 148–149.
80
Gender and Position-Taking in Henrician Verse
Works Cited Bates, Catherine, ed. A Companion to Renaissance Poetry. Hoboken, NJ: Wiley, 2018. Clements, Robert J. and Lorna Levant, eds. Renaissance Letters: Revelations of a World Reborn. New York: New York UP, 1976. The Devonshire Manuscript. 1525?–1567? MS BL Add. 17492. British Lib., London. Fortescue, John, Sir. On the Laws and Governance of England. Edited by Shelley Lockwood. Cambridge: Cambridge UP, 1997. Foucault, Michel. Discipline and Punish: The Birth of the Prison. Translated by Alan Sheridan. New York: Vintage Books, 1995. Herman, Peter C., ed. Rethinking the Henrician Era: Essays on Early Tudor Texts and Contexts. Urbana: U of Illinois P, 1994. Holton, Amanda and Tom MacFaul, eds. Tottel’s Miscellany. By Sir Thomas Wyatt; Henry Howard, Earl of Surrey; et al. New York: Penguin, 2011. Kilgore, Robert. ‘The Politics of King David in Early Modern English Verse’. Studies in Philology 111, no. 3 (2014): 411–441. King, John N. ‘Henry VIII as David: The King’s Image and Reformation Politics’. In Herman, Rethinking, 78–91. King, John N. Tudor Royal Iconography: Literature and Art in an Age of Religious Crisis. Princeton: Princeton UP, 1989. Marotti, Arthur F. Manuscript, Print, and the English Renaissance Lyric. Ithaca: Cornell UP, 1995. Marotti, Arthur F. and Steven W. May. ‘Manuscript Culture: Circulation and Transmission’. In Bates, Renaissance Poetry, 78–102. May, Steven W. ‘Popularizing Courtly Poetry: Tottel’s Miscellany and Its Progeny’. In The Oxford Handbook of Tudor Literature, 1485–1603, edited by Mike Pincombe and Cathy Shrank, 418–433. New York: Oxford UP, 2009. Morley, Henry Parker, Lord. The Exposition and Declaration of the Psalme, Deus Ultionum Dominus. London: Thomas Berthelet, 1539. The Oxford Dictionary of National Biography Online. Oxford UP, 2015. Web. https:// www.oxforddnb.com/. Accessed 31 May 2021. Prescott, Anne Lake. ‘The 2011 Josephine Waters Bennett Lecture: From the Sheephook to the Scepter: The Ambiguities of David’s Rise to the Throne’. Renaissance Quarterly 65, no. 1 (2012): 1–30. Seneca. Phaedra. In Seneca: Tragedies, Volume I. Translated by John G. Fitch. Loeb Classical Library 62. Cambridge, MA: Harvard UP, 2018. Sessions, W.A. ‘Surrey’s Wyatt: Autumn 1542 and the New Poet’. In Herman, Rethinking, 168–192. Siemens, Raymond G., ed. The Lyrics of the Henry VIII Manuscript. Grand Rapids, MI: English Renaissance Text Society, 2013.
Tr aditions of Resistance and Verse Position-Taking
81
Simpson, James. ‘Breaking the Vacuum: Ricardian and Henrician Ovidianism’. Journal of Medieval and Early Modern Studies 29, no. 2 (1999): 325–355. Songes and Sonettes, Written by the Right Honorable Lorde Henry Howard Late Earle of Surrey, and Other. Ed. Richard Tottel. London: Tottel, 1557. Walker, Greg. Writing under Tyranny: English Literature and the Henrician Reformation. Oxford: Oxford UP, 2005. Warner, Christopher J. The Making and Marketing of Tottel’s Miscellany, 1557: Songs and Sonnets in the Summer of the Martyr’s Fires. Burlington, VT: Ashgate, 2013. White, Micheline. ‘Katherine Parr, Henry VIII, and Royal Literary Collaboration’. In Gender, Authorship, and Early Modern Women’s Collaboration, edited by Patricia Pender, 23–46. Palgrave: 2017. White, Micheline. ‘The Psalms, War, and Royal Iconography: Katherine Parr’s Psalms or Prayers (1544) and Henry VIII as David’. Renaissance Studies 29, no. 4 (2015): 554–575. Wyatt, Thomas, Sir. Collected Poems of Sir Thomas Wyatt. Edited by Kenneth Muir and Patricia Thomson. Liverpool: Liverpool UP, 1969.
3.
Translation and the Position-Taking Verse Tradition Abstract: This chapter focuses on translation in the courtly love lyric tradition, bridging the intersection of translation and engagement with tradition as position-taking strategies and emphasizing the importance of understanding the manuscript as a space shared by courtier poets of all genders. The chapter contextualizes the uses of translation as a genre that codes emotions and positions; makes them socially acceptable, expressible, and exchangeable; and lends these positions legitimacy through an attempt at universalizing the poets’ experiences. As the most prominent discourse, translation of love poetry marks an important access point; as only one of the available discourses, the practice can be used to explore what elements transfer between strategies and how each kind of poetic work influences others. Keywords: Tudor translation; courtly love lyric; Devonshire Manuscript; Tottel’s Miscellany; Henry Howard; Thomas Wyatt
Translation has been so often and so thoroughly understood as a sort of code for courtiers that a section on the practice may seem superfluous or to imply that I aim here only to extend the established critical practice to new works. Certainly, this chapter partly establishes the current conversation about translation as a jumping-off point for understanding other courtly verse. I want also, though, to draw attention to how establishing as ‘conventional’ this reading of translation work as coded can obscure some important facts about the practices of the creation of the verse. This framework may by now have become so accepted that our interpretation of it has become traditional and even conservative: we see the same tropes, and so we miss the resistant and even shocking nature of the content. In practice, writers at court were able to spend years using translation as coded commentary on contemporary events without ever becoming repetitive in that commentary.
Quoss-Moore, R.M., Gender and Position-Taking in Henrician Verse: Tradition, Translation, and Transcription. Amsterdam: Amsterdam University Press, 2023 doi 10.5117/9789463723534_ch03
84
Gender and Position-Taking in Henrician Verse
In our efforts to restore reading and writing networks, we can try to restore the immediacy and urgency of the work this kind of verse was doing. In fact, the tradition for the practice was important in its contemporary setting, rather than devaluing the technique through any sense of tired repetition. Position-taking verse in the Henrician court gained energy by positioning itself as and within a specific poetic technique and tradition. As established in the first section of this work, courtiers relied on the reading of their actions as ‘traditional’ to justify and, to a limited extent, protect their works. The use of Petrarchan verse translations aimed at Henry VIII’s particular vanities. The poetry itself was part of the pan-European humanist tradition of which Henry so very much wanted to be considered the pinnacle. Given how ancillary England was seen as being by the usual Continental power players, the amount of praise Henry garnered as an ideal humanist prince is worth noting.1 Further, that attention shaped the expectations of the courtiers who served him, who understood their role as both artistic and advisory. These artists believed that part of their duty was to shape the politics of their country, even—or perhaps especially—when that meant criticizing the King.2 The Henrician courtier-poets were placed in a particularly difficult position by Henrician policies that essentially criminalized dissent. The writers of Henry’s court also often used translation to explore their socially stratified and gendered positions, as well as to examine what service to king and country meant. Translation may have been particularly useful to such exploration because it tacitly allowed comparison of England’s politics and norms with those of the country in which the poem was originally written or in which other translations were already available. It opened up a space for imagining other ways of being. Equally important to poetic production, though, was the circumscription of such possibilities—the meaning-making process through which audiences received and revised position-taking verse. Henry’s courtiers often use the love lyric to work through frustrations at the 1 As discussed in the first chapter; see also Herman, Reading Monarchs and Royal Poetrie. 2 See, as discussed in the first section of this work, Walker, Writing Under Tyranny and Lockey, Law and Empire; Pincombe and Shrank offer a neat summary of the situation when they connect Tudor writers’ ‘awareness of the wider society’ to a characterization of the sixteenth century [as] the age of ‘commonweal’. Coined during the second half [of] the fifteenth century, the term gained rapid currency and came to dominate sociopolitical discourse in Tudor England, employed (often simultaneously) both as a synonym for ‘the realm’ and to denote a state in which government serves the interests of the polity as a whole. It inscribed the obligation of governors to rule for greater good, but it also imposed responsibilities on citizen-subjects to play their part. (‘Travails of Tudor Literature’, 6)
Tr ansl ation and the Position-Taking Verse Tr adition
85
increasingly subordinate, and potentially negatively feminized, nature of appropriate courtly performance as monarchical power and centralization increased. Though Wyatt and Surrey are the most well-known actors in this genre, the Devonshire Manuscript preserves the work of other courtiers in this mode, as well. The key intervention introduced by the work of the Devonshire is that many of the writers of the manuscript are not men. Placing these works alongside Wyatt’s and Surrey’s allows for more complicated readings of that tension between ideal courtly performance and idealized interiority, suggesting the weaknesses of understanding the Henrician system as only a dichotomy between masculinized and feminized performances. The association of service and subordinate positions—and frustration at that association—was not limited to use by courtiers who were men, though gender identity does inform poets’ uses of and reactions to the trope, perhaps particularly in translation work. Henrician poets borrowed from their Continental predecessors an appreciation for and understanding of the ability of the courtly love lyric to convey political frustration.3 By joining translation to similar position-taking techniques, such as those used in transcription and in original composition, we can highlight some consistent systems of political and social resistance at play across poetic genres. The joint analysis also highlights the ways that coding works differently in translations than through other outlets. In translation, this system functions with more clarity in its critiques and so more risk, as the topic may explicitly be a corrupt ruler, and as even the more shielded love poetry was widely understood as poetry about service at court. However, the nature of translation also offered a certain kind of deniability that could protect the poet/translator. That clarity and deniability were both bound up in the tradition of the thing. Across early modern Europe, Petrarch’s Canzoniere ‘was considered [… a] repertory, a sort of commonplace book where poets could find ready-made materials and a common, transnational poetic language per figuris shaping emotions and feelings for a widespread western European aristocratic milieu’. 4 That language—the available vocabulary—functioned on a huge scale as one kind of coding, but the English courtiers invested their own particular contextual anxiety into their restructuring of these works. Petrarchan poetry serves as a useful outlet for such expression ‘because of 3 See, for example, Domenichelli, ‘Wyatt’s Translations’ and Greene, ‘Petrarchism’. Much of my work and the work of the current critical conversation is, of course, indebted to the foundations laid in Marotti’s English Renaissance Lyric and Greenblatt’s Renaissance Self-Fashioning. 4 Domenichelli, ‘Wyatt’s Translations’, 66.
86
Gender and Position-Taking in Henrician Verse
its engagement with such political issues as the distribution of power among agents, the assimilation of difference, and the organization of individual desires into common structures of action and reaction’.5 This last point, on the organization of individuals into a common structure, is important for any analysis of courtly verse, but particularly here, where I wish to emphasize the uses of such poetry as part of a social structure and, simultaneously, as a force which aims to shape that social structure. In fact, their training in understanding translation as an adaptive commentary may have helped shape courtiers’ abilities to adapt other genres into commentary. The habit of accessing alternative meanings could be applied to other forms, once learned—in much the same way that I want to use our established understanding of the multiple meanings of translation to open up new understandings of other courtly poetry. Translation becomes a point of access for these practices not only in our modern understanding and criticism, but also in the original historical practice. The focus of work on Petrarchan translations at Henry’s court has often been on decoding the emotions being expressed and on relating those expressions to specific events. I want to add to specific interpretations a larger consideration on the uses of translation as a genre that codes those emotions; makes them socially acceptable, expressible, and exchangeable; and lends these emotions legitimacy through an attempt at universalizing the poets’ experiences. In short, I want to look at translation as one element of a system of position-taking, rather than assigning a position to each poem for its own sake. Roland Greene expresses the appeal of ‘love poetry [a]s one of the available discourses—for sheer volume in this period, perhaps the most available—in which the concerns of power, selfhood, and difference can be figured with shadings of particularity and ambiguity’.6 As the most prominent discourse, translation of love poetry marks an important access point; as only one of the available discourses, the practice can be used to explore what elements transfer between strategies and how each kind of poetic work influences others.
A. Titles, Tottel’s, and Verse in Context The poems considered as Wyatt’s and Surrey’s, here, were all included either in Tottel’s Songes and Sonettes or in other, later print collections, and I 5 Greene, ‘Petrarchism’, 131–132. 6 Greene, 149.
Tr ansl ation and the Position-Taking Verse Tr adition
87
generally use the versions of their work that appeared in Songes and Sonettes.7 Although the versions in Tottel’s collection often differ significantly from the manuscript versions that first circulated at court, there is value in seeing the forms these poems took in their first presentation beyond the court in terms of understanding how originally resistant political energies might be dispersed.8 Generally, even overtly critical elements of the poems are maintained, and the only mitigation added is Tottel’s titles.9 I consider the imposition of titles across all of these works as an extension or result of the system of coding inherent in such work. Arthur F. Marotti notes that Tottel’s practice ‘signals the recoding of social verse as primarily literary texts in the print medium’10 and contrasts this ‘workmanlike activity’ with Tottel’s presentation of himself as ‘a class mediator taking advantages of print technology’s ability to open the closed communications of an elite to a wider audience’.11 Steven W. May points out that these titles ‘seldom offer insight into the motives or contexts of composition, but instead summarize what can be deduced from reading each poem’.12 This point could, perhaps, be joined to Christopher J. Warner’s argument that Tottel has in mind an idea of a collaborative space as a ‘pragmatic and an idealistic response to 7 On the one hand, Songes and Sonettes, first printed in 1557, was an invaluable repository for Wyatt’s and Surrey’s works, among others. On the other, as I explore throughout, Tottel’s investment in legitimizing forces may have led to the erasure of some authors and to the cloaking, intentional or accidental, of the more overtly political content in the verses he chose to include. Warner’s Making and Marketing particularly foregrounds the socioeconomic influences traceable in the text’s production; Heffernan’s Making the Miscellany offers excellent arguments on the distorting weight particular interpretations of Tottel’s have had on our field’s formation. 8 Edwards argues, in fact, for superior readings, in many cases, in Tottel’s renderings of Surrey’s lyric, though not any ‘invariable superiority’ and further points out that, in several cases, Tottel’s provides the earliest available version of a poem (‘Print and Manuscript’, 35). 9 Warner offers an overview of instances of censorship in the collection, reviewing Rollins and Hamrick, as follows: They number merely five, two of which are political and two religious in nature, while the ground for the first one cited by Rollins is obscure: (1) the deletion of Wyatt’s reference to a ‘Kitson’ in R126/M136, possibly the London sheriff Sir Thomas Kitson or the bookseller Anthony Kitson; (2) removal of Wyatt’s reference to corruption at Rome in R125/M135; 16 (3) deletion of antipapal stanzas from the epitaph for William Gray, R255/M239 [stanzas which Warner doubts, in any case]; (4) deletion of John Heywood’s reference to Mary in R199/M168 (‘complimentary though they were’,[…]); and (5) removal of a reference to the younger Wyatt in ‘A praise of Audley’, an anonymous poem celebrating the career achievements of a captain at Guisnes that included his helping to put down the Wyatt rebellion, R205/M175. (Making and Marketing, 169) 10 Marotti, English Renaissance Lyric, 218. 11 Marotti, 294–295. 12 May, ‘Popularizing Courtly Poetry’, 430.
88
Gender and Position-Taking in Henrician Verse
the religious, political, and social upheavals’ of 1557.13 That is, one way of understanding the titles is as an effort to, if not de-politicize (as an effort towards class mediation is not apolitical), then perhaps de-polarize the contents of the poems—often by imposing Petrarchan themes in the absence of cues from the poem. However, the lack of titles in the original works would have emphasized their flexibility. The context of the courtly love tradition did not enforce a strict interpretation of all lyrics as love lyrics, in the way that the later projection of titles might suggest.14 Instead, this process suggests that the work of the original authors was successful in its construction of deniability and alternative interpretations, but that greater work must be done to consider the socio-historical context and the compositional context—of manuscript culture, multiple versions, and courtly exchange. Indeed, the first example of Wyatt’s poetry which I wish to examine includes also one of the most incongruous titles: Wyatt’s ‘CEsar, when that the traytour of Egypt’, which Tottel titles ‘Of others fained sorow, and the louers fained mirth’.15 The title added by Tottel may function in two distinct ways: either as a misreading or as an additional layer of subterfuge. For what reasons does Tottel assume that this poem, which seems most overtly about the hypocritical performances demanded of those in the public eye, applies to or speaks of a love affair? Why does Tottel speak only of others’ ‘sorrow’, when Wyatt’s poem balances between Caesar’s feigned tears and Hannibal’s assumed cheer? Partly, Tottel may be understood to appeal to the aristocratic tradition of courtly love as an authorizing force—that is, he positions the poem within the larger courtly love tradition for much the same reason that the preeminent name on the title page is Surrey’s, whose position, title, and bloodline had far outstripped Wyatt’s. Additionally, though, the person who created this collection would necessarily have access to at least a measure of the interpretive force of the frustration, resentment, and often anger which these courtier-poets expressed through their verse. As such, these titles may also serve as an attempt to gain security for the text by masking its challenges to authority. While my argument here is somewhat distinct from his focus, I think this can be reconciled quite easily with Marotti’s arguments about the class mediation work in the text, as the coding of 13 Warner, Making and Marketing, 3–4. 14 While the Petrarchan originals, particularly, may have been understood as being ‘about’ love, they were also, extensive scholarship has shown, understood as not only about love; the variety of the translations themselves are indicative of the readings available to Henrician poets. See, especially, Marotti, ‘Love is Not Love’. 15 Tottel, ed., Songes and Sonettes, fol. 21r.
Tr ansl ation and the Position-Taking Verse Tr adition
89
such challenges to authority does not translate neatly between the social spheres of court and the city, and the impulse towards security certainly could result from the push towards decreased ideological disagreement that Warner identifies.16 Such work does not remove interpretive possibilities for courtiers with ‘insider’ knowledge; it simply does not offer a clear transfer of those possibilities for the wider audience.17 If the titles do attempt to preserve the ability of the poems to mean more than they say, they simply extend part of the project of Wyatt’s poems. Peter Sacks makes the point that sonnets like Wyatt’s ‘may come into being in large part to construct or repair enough sense of interiority […] so that they can address and maybe redress whatever has shamed them into being’.18 That is, the poems construct a particular embodiment of performativity, a voice created to both respond to real conditions and to be read as constucted. Greene characterizes the objects of such poetry as ‘celebrated and deplored as epitomes of the speakers’ and their societies’ polarized values’.19 Such a characterization applies, too, to the kings and ministers who were often the veiled objects of such poetry. Henry, particularly, seems an appropriate target for such doubled reference, celebrated as the perfect humanist prince and deplored as an archetypical tyrant. Such multiplicities are apparent in ‘CEsar, when that the traytour of Egypt’, where Wyatt expresses ‘the incommensurability of human insides and outsides, and the helplessness of the first person within a warped system of communication’.20 Greene’s reading accesses Wyatt’s expression of frustration at the limitations of self-expression, both those that are natural to the human state and those imposed by the courtly system. We can then add the explicit acknowledgement that such frustrations are inherently political in the Henrician system, where limitations and rules of self-expression are at least ostensibly determined by the monarch. Wyatt seems to use Caesars as frequent analogues or stand-ins for Henry, as in ‘Whoso List to Hunt’ and ‘Circa Regna Tonat’; the relationship is of 16 Marotti, English Renaissance Lyric, and Warner, Making and Marketing. 17 Megan Heffernan’s Making the Miscellany, released while I was working on revisions for this book, considers the ‘parcels’ of Songes and Sonettes in terms strikingly similar to my own approach to the Devonshire (30–38); Heffernan identifies, for example that ‘the design of the compilation stitched extant writing together into new material and conceptual forms that exceed the original limits of any individual poem’ (Making the Miscellany, 31). Heffernan is most interested in these parcels as a distinctive feature of print compilation; I see, there, an outgrowth or new implementation of manuscript practice, though certainly one that is being used in a new way. 18 Sacks, ‘Face of the Sonnet’, 28. 19 Greene, ‘Petrarchism’, 133. 20 Greene, 145.
90
Gender and Position-Taking in Henrician Verse
course strengthened by the understanding of ‘caesar’ as simply ‘king’. And in this depiction of a king who rejoices over the death of his noble adversaries while pretending sorrow, Wyatt could have thought of any of a dozen incidents in Henry’s own reign. The entire first two quatrains of the sonnet seem most likely to apply to Henry, as emphasized by the use of commanding historical heads of states for analogous examples. Further, the first quartet initially seems to cut off any potentially sympathetic interpretation. Historically, understandings of Caesar were more complex—he could be both villain and hero. The opening of Wyatt’s poem, though, takes a particular and decided stand: CEsar, when that the traytour of Egypt With thonorable hed did him present, Couering his hartes gladnesse, did represent Plaint with his teares outward, as it is writ.21
Caesar does not truly weep for Pompey, and his sorrow is all show. Since Caesar’s sorrow is show, he is allied with the traitorous ‘other’ of the Egyptian.22 Indeed, one particular identity for the ‘traitor’ could be Pothinus—certainly one of the central plotters—whose association with the feminized luxury of the ‘East’ and whose status as a eunuch would emphasize the unmanliness of Caesar’s actions. The implication then reaches much further, casting the speaker’s lot with one particular ideological interpretation of history. Caesar is not a hero, these first lines imply, and perhaps a tyrant who overthrows popular rule cannot be a hero, or even a proper man. Kim F. Hall’s foundational Things of Darkness enables this reading; as one example, Hall writes of the intersecting signs of Egypt, the Nile, and Cleopatra that figure ‘overwhelming sexuality and social disorder […] the kind of overflow and excess characteristic of the female grotesque’.23 The ‘othering’ of Egypt implicit in the lines speaks also to a particular type of English nationalism, one that privileges English tradition and likely ‘masculinity’, especially as expressed in language and appropriate gendered performance. The lines on Hannibal, though, are less plainly critical; the inside and outside of the man still do not match, but now the reader understands the 21 Wyatt, Songes and Sonettes, fol. 21r, ‘CEsar, when that’, lines 1–4. 22 Much of the most thorough work on Tudor-Stuart historiographies of Egypt has been in Shakespeare studies. See, for example, Park, ‘Discandying Cleopatra’; Crane, ‘Roman World, Egyptian Earth’; and Loomba, ‘Space of the Other’. 23 Hall, Things of Darkness, 157.
Tr ansl ation and the Position-Taking Verse Tr adition
91
separation as a sign of courage. The ‘Eke’ that begins the fifth line emphasizes both the shift and the connection, as the next example then commences, ‘Hannibal, when fortune him out shit / Clene from his reigne, and from al his entent / Laught to his folke’.24 This description emphasizes another element of the lines on Caesar—in both cases, the mismatch between show and emotion is actually advantageous. In opening with the hypocritical Caesar, Wyatt apparently critiques this show, but he quickly shifts to an example that can be understood as sympathetic, even admirable. Further, the reader’s sympathies are, in some sense, misaligned according to the values of the day. Wyatt offers two examples of masculine performance, both celebrated generals, but one in the flush of victory and the other in the agony of defeat. There seems, in the poem, more imagined opportunity to play the man in the face of martial failure than in the bloom of success, particularly given the presence of the likely-feminized Egyptian traitor. This opportunity complicates the understanding of the poem’s potential commentary on Henry’s reign—certainly he could be the hypocritical Caesar, but he is also a king oft-disappointed in war, aligning him with the defeated Hannibal. Further, Pompey had been the head of the conservative faction. Interpreting the conservation as one of power for the elites, rather than government by the common man, creates a quite different effect—Caesar’s show of sorrow is appropriate and efficacious, and there may be little shame in only pretending to mourn an enemy of the people. The criticism is both decreased and turned against the norms of a culture which demands such performance. The shift between the second quatrain and the following tercet opens the possibility of a change of the orientation of the critique from personal— albeit political in its object—to societal. The speaker gathers, from his earlier examples, the conclusion ‘that euery passion / The minde hideth by colour contrary, / With fained visage, now s[ad], now mery’.25 Initially, in context, these lines may seem another lament about the falseness of society. However, the final tercet interrupts this set of expectations as well—the social critique now applies to the speaker of the poem: ‘Wherby, if that I laugh at any season: / It is because I haue none other way / To cloke my care, but vnder sport and play’.26 The speaker is as duplicitous as anyone else, but that duplicity is also a utility—a way of preserving interiority. Sacks supposes that the question underlying this ‘critique of dissembling’ may still be ‘to 24 Wyatt, ‘CEsar, when that’, lines 5–7. 25 Wyatt, lines 9–11. 26 Wyatt, lines 12–14.
92
Gender and Position-Taking in Henrician Verse
what extent could such an appearance, however ‘honest’, serve the inner life it is supposed to lay bare?’.27 Our consideration of verse position-taking as a communal practice extends this reading; if social expectations create the parameters of communication, the speaker’s and the objects’ practices of dissembling may be the best way to protect and to project the individual. Further, because the importance of those historical figures of the opening lines is not erased, the critique also serves to collapse social boundaries. Wyatt’s problem, the charge his speaker faces, is the same as that faced by kings—and kings across time, moreover. Even as the poem critiques the court system, it also suggests an inevitability to these limitations and frustrations. From overt critique, the lines turn to statement of fact, a simple description of the world within which Wyatt exists. Such a description offers the reader an important guideline—trust nothing on the surface. Whatever a poem seems to be ‘about’, the poem is as much a part of the courtiers’ performance as their dress, their play, or their speech, and the poem suggests that its lines should be similarly distrusted and dissected according to the context, just as the ‘real’ reactions of both Caesar and Pompey can be contextualized through a consideration of the larger history. The poem’s position as a translation both emphasizes its connection to some kind of desired historicity and, simultaneously, encourages the reader to consider the way established texts can be adapted to consider new situations, drawing the reader’s attention to a balance between authenticity and performance that adheres to the poetic act, itself, as much as to the narrative content produced. Wyatt’s ‘I Finde no peace, and all my warre is done’ demonstrates the interpretive importance of such context. Tottel uses the title ‘Description of the contrarious passions in a louer’, and such a title does speak to the place of the poem in a tradition of love poetry based in contrasting extremes.28 The poem begins with six lines, each containing a contrast, followed by a contrasting couplet, then proceeds with another six lines of internal contrasts. Part of the context, then, is this sort of balancing verse as a feature of courtly love lyrics. In the larger system of verse position-taking, that association with the courtly love lyric does not limit the meaning; the balancing lines activate one context, through a trope of courtly love, and the larger court is activated as a further context, partly because Wyatt’s own oeuvre of verse complaint often extends to courtly service. Seen through this lens, the poem takes on clearer position-taking energies, even as the poem’s 27 Sacks, ‘Face of the Sonnet’, 22. 28 Tottel, ed., Songes and Sonettes, fol. 21v–22r.
Tr ansl ation and the Position-Taking Verse Tr adition
93
position as a translation protects and codes those energies. From the first line, the speaker’s arguments can be seen as resistant to the contemporary strictures of service—but the nature of translation allows Wyatt to trouble any simplification of his relationship to that speaker. In this context, the first line—‘I Finde no peace, and all my warre is done’—may in fact be a double complaint; the space of martial achievement has been cut off, and the speaker is also not left in peace.29 The increasing centralization of the court could produce such a complaint in any number of aristocrats; a class who traditionally gained honor in combat was now expected to instead pay constant court to the King as the new source of honor.30 The third line, ‘I flye aloft, yet can I not arise’, similarly expresses a doubled tension of a system within which power is held so extensively by the King and those he favors at a given moment.31 Wyatt himself would have experienced this problem any number of times in his career—favored enough to be sent frequently abroad as an ambassador and at times admitted to Henry’s most intimate social circles, but never raised in any permanent way above the status his father had achieved, and always subject to Henry’s tempers and suspicions.32 Indeed, the line that follows—‘And nought I haue, and all the world I season’—seems particularly appropriate to exactly the work of an ambassador, sent throughout Europe as a power broker but always under the strict and often impossible dictates of his monarch.33 The following four lines then unite around a theme of carcerality, another topic with strong connections both to themes of courtly service and to Wyatt’s own experience: That lockes nor loseth, holdeth me in prison. And holdes me not, yet can I scape no wise: Nor lettes my liue, nor dye, at my deuise, And yet of death it geueth me occasion.34
The speaker complains of a force that imprisons him in misery without any apparent, ‘real’ physical constraints. Such lines offer a clear case of the ways 29 Wyatt, Songes and Sonettes, fol. 21v–22r, ‘I Finde no peace’, line 1. 30 See Walker, Writing Under Tyranny, particularly ‘Tyranny Condemned’, 296–334. 31 Wyatt, ‘I Finde no peace’, line 3. 32 For details of Wyatt’s biography, see especially Muir, Life and Letters, and Brigden, The Heart’s Forest. 33 Wyatt, ‘I Finde no peace’, line 4. 34 Wyatt, line 5–8.
94
Gender and Position-Taking in Henrician Verse
in which Petrarchan tropes were revitalized by the complaints of Henry’s courtiers. Love as a prison was a poetic trope; Henry’s rule proved, though, that acts of love or service improperly performed could lead to much less metaphorical constraints and punishments. One line stands out from the rest for its physicality and for the relative lack of analogues in other contemporary love lyrics. The freezing and burning, the torture and delight are fairly common tropes, but, following the prison lines, the speaker claims that ‘Without eye I se, without tong I playne’.35 A blind and mute speaker, readable as a result of some kind of disfigurement, is a more visceral image than that of the rest of the poem, or than is typical of the larger body of work. In this moment, I suggest, the poem itself becomes the speaker. Analogous riddles were highly popular at court; they filled the same kinds of exchanged manuscripts and commonplace books that would have been the first homes of much of Wyatt’s verse. The thing that sees and records the actions of the unjust master/mistress and the tortured servant is the poem. What that poem records is that the speaker’s self-loathing is intimately bound up in his service to another, explicit in the line ‘I loue another, and I hate my selfe’ and the complaint that ‘my delight is causer of this strife’.36 If a doubled interpretation is opened up by the conflation, in the riddling reading, of the lover’s and the poem’s tortured bodies, wherein the speaker is lover and poem, then the ‘delight’ becomes similarly doubled, both the beloved and, potentially, the act of writing or of self-creation, in a sense. The poem’s desire to die, then, complicates the speaker’s desire to die; if the poem is an act of performance of self, the speaker’s contradictory desires can potentially be read as possible if he can escape this space of performance. While Tottel’s title encourages an interpretation of this service as that of unrequited or tortured love, the context of Wyatt’s extensive work suggests that the ‘other’ could as easily be the person behind the imposition of courtly service under which he so often chafed. The final line may codify Tottel’s decision to treat the lyric as love poetry, but Wyatt’s long and complex history of court service suggests any number of alternatives. While ‘I Finde no peace’ may perhaps offer some cues towards consideration as a love lyric, ‘MY galley charged with forgetfulnesse’ aligns more definitely with ‘CEsar, when that the traytour’ in the absence of any clear suggestion of such a context. Nonetheless, Tottel adds just such a distinction through his title: ‘The louer compareth his state to a shippe in perilous 35 Wyatt, line 9. 36 Wyatt, lines 11 and 14.
Tr ansl ation and the Position-Taking Verse Tr adition
95
storme tossed on the sea’.37 Again, the insistently limited metaphor of the title suggests that Tottel assigned it with some distinct aim in mind, but the exact authorizing action—beyond a ‘connection […] to the traditional fictional world of love experience’38—is difficult to assess. Nonetheless, if the title insists that the poem is casting Love or the heart as the speaker’s cruel lord and torturer, then Love (or the speaker’s heart) forces the speaker to do what his rational mind does not want. Given that the courtier is also in a sense a ‘louer’ of the monarch, the title may not, ultimately, shut off the alternative interpretations available. In the poem, lines three and four emphasize the applicability of the poem to Wyatt’s service at Henry’s court. ‘Twene rocke, and rocke: and eke my fo (alas)’ both continues to chart the journey of the galley/speaker and aligns the foe with the rocks and ‘sharpe seas’ around which the soul must navigate.39 The enjambment emphasizes the surprise of the next line: ‘That is my lord, stereth with cruelnesse’. 40 The foe is also the lord, and also the one at the helm of the journey. This lord steers his craft with cruelty. The analogy of ship to country springs easily to mind, and, in this case, the captain is a hard one. The comparison is further emphasized by Henry’s own role in establishing the Royal Navy, perhaps even veering near a dangerous reference to Henry’s naval defeats against the French early in his reign. The language is further suggestive of a lost path, throughout, which calls to mind the changing perceptions of Henry and of his kingship. A rayne of teares, a clowde of darke disdaine Haue done the weried coardes great hinderance, Wrethed with errour and with ignorance, The starres be hidde, that leade me to this paine, Drownde is reason that should be my comfort: And I remaine, dispairing of the port. 41
The speaker cannot see the stars or revive the reason which should guide his galley; the tack of his ship has been worn down through tears, disdain, error, and ignorance. As a result of all these complications, the speaker no longer trusts the end of his journey. Here, that end may be of the soul, whose fate 37 Tottel, ed., Songes and Sonettes, fol. 22r. 38 Marotti, English Renaissance Lyric, 218. 39 Wyatt, Songes and Sonettes, fol. 22r, ‘My galley’, line 3. 40 Wyatt, ‘My galley’, line 4. 41 Wyatt, lines 9–14.
96
Gender and Position-Taking in Henrician Verse
has been called into question by the Reformation, or of the physical body, whose fate, most courtiers now understand, may be subject to the capricious jealousies of their King. As such, the galley is ‘charged with forgetfulnesse’ in, at minimum, the senses of needing to forget what was expected in order to go forward, of being loaded down by the weight of what others have forgotten, and of being charged by the captain/foe to forget—to move only forward, leaving the mistakes of the past as irredeemable. 42 By emphasizing the ‘charge’ of forgetting, the speaker paradoxically encourages a process of recalling, as the reader tries to sense what it is that must be forgotten.
B. Language Choices and Communal Position-Taking Practice From these examples of Wyatt’s contextualized practice, I want to move into a new comparison of Wyatt’s and Surrey’s versions of Petrarch’s Rime 140, considering each specifically as an expression of its poet’s position-taking ideals. Moreover, I want to contextualize these poems not only in their political and courtly context, where rich traditions of critical work have already been established, but also in the context of translation’s durable and poetic function and within an understanding that this function was not created or used solely by men. This last point, particularly, encourages us to restore gender to our readings of the poems, recognizing that the position ‘man’ matters as much to a writer as the position ‘woman’—and, at the same time, rethinking when and how such positions matter. Wyatt was, perhaps, the more cynical of the two men, in terms of his belief in the efficacy of language to effect change. However, Surrey’s belief in the efficacy of his own efforts is linked to a particularly arrogant strain of naiveté—a misapprehension of his own importance and power, inflected by his class and by his gender. If Wyatt is less certain that the world will turn out all right by his lights, he seems also more sympathetic in his frustrations and his worldview. Surrey’s own view of what a ‘good’ world would be is less appealing to our modern audience, and his sense of his own infallibility can be frankly off-putting. Wyatt’s poetry tends to speak to a wish that his world placed more value on talent, that goodness and/or talent might be enough to secure some sense of place and surety of safety, and that goodness and talent might be more easily reconciled—that is, that courtiers of talent, such as himself, did not so often feel that they could only fully exercise those talents through compromise of personal moral codes. In contrast, 42 Wyatt, line 1.
Tr ansl ation and the Position-Taking Verse Tr adition
97
Surrey’s poetry booms with a belief that morality belongs to a particular brand of nobility, that talent will almost always follow the inborn worth of the nobleman, and that Surrey himself is preeminently noble, preeminently talented, and so, of necessity, preeminently right. The third and fourth points are inarguable; Surrey’s rich history of assault and rash action makes his conclusion impossible. 43 Mario Domenichelli argues of Wyatt that ‘the kind of English he uses, the kind of expression he looks for […] bears witness to a lower, more colloquial linguistic choice’; he contrasts these choices with the ‘higher’ language of Surrey.44 The particular projects of the two poets were indeed different—but their belief in the importance of language was shared, even if they brought to that belief disparate ideological preconceptions and aims. In this separation, the stakes were on not only language, but ‘also self-perception: the right way of feeling shaped by the naming of the figures of the mind’s life’. 45 It follows that when Wyatt chooses the more common language, he makes an evaluative statement about how one should be and of what sort of people the court should be composed. Regardless of the ostensible object of a particular poem, Wyatt’s very language makes every poem a political act, arguing for a space for the more middling sort of English person as a central figure at the court, the site and seat of government. While Wyatt’s vision may not be entirely radical—that his language is more common does not make it the language of, for example, rural farmers—his vision is nonetheless several degrees more comprehensive than Surrey’s. Domenichelli catches the poetic philosophical significance; I want to extend that point to more fully engage the sociopolitical implications thereof. Wyatt’s poetry works in Wyatt’s world; Domenichelli comes nearest to the importance of this with his point that ‘Wyatt’s idea of poetry, his own implicit ars poetica […] must be understood in the same key of mediocritas struck by Castiglione’. 46 Indeed, such arguments for meritocracy indicate the very reason that Henry could be both denounced as an upstart and celebrated 43 An early reader highlighted that my interpretations might seem to reify ‘the conventional reading of the two men—Surrey the stuffy aristocrat; Wyatt the true genius’. Acknowledging that risk, I hope that placing Wyatt in this larger context of communal practice effectively troubles identif ications of any author as a ‘true genius’, a designation that likely reinforces individualized, masculinized misconstructions of our literary canon and history. I also hope that my critiques of Surrey’s work read as critiques of institutionalized, learned classism and misogyny—an engagement with the implications of his content, rather than a refutation of his poetic ability. 44 Domenichelli, ‘Wyatt’s Translations’, 74. 45 Domenichelli, 74. 46 Domenichelli, 75.
98
Gender and Position-Taking in Henrician Verse
as the coming of the new Ideal. Humanism espoused investment in ability over tradition—and so promised that persons of ability could rise. Henry himself would very likely not have understood his own power or fame in these terms, and it is possible that subjects like Wyatt might not have, either, particularly given Wyatt’s father’s loyalty to the Tudor dynasty. 47 But the fact remains that Henry’s investment in raising England up is related to his project of establishing a government of semi-meritocracy. Surrey, seemingly, understood Henry’s reign pretty much exactly that way: it was quite all right for his beloved Wyatt to be raised to a certain status based on merit and talent; it was quite another to put a king on the throne based on anything other than hereditary right, and Surrey was not, his later actions indicate, completely convinced of the greater Tudor claim to the throne. 48 There is, further, a suggestion here of the influence of women’s manuscript poetry on Wyatt’s composition, and, perhaps, on Surrey’s, as well. Terttu Nevalainen and Helena Raumolin-Brunberg have produced extensive study of linguistic shifts in the sixteenth century, with an especial emphasis on the role of gender and class as informing writing—or, conversely, on writing as performing gender and class.49 While their study is more specifically applied to the later work here on the Devonshire, one generalization of their research is particularly key: courtly women and ‘lower class’ people of all genders adopt linguistic shifts more rapidly. Wyatt’s own uses, this data suggests, are likely partly informed by his socially aspirant status—but they are also almost certainly informed by his participation in courtly manuscript production and, thus, by the patterns used by women courtier-poets in those same manuscripts. What we see in the recorded ‘common’ language of Wyatt’s published work are the traces of language use within the court—language use highly informed by social prescription, by self-preservation, and by participation in a larger system of position-taking. Wyatt’s language, then, does privilege a system based on talent and innovation, and Domenichelli is right to observe that, to some extent, ‘Wyatt’s lingua d’uso seeks to identify with the very language of truth 47 In her outline of some of Henry Wyatt’s key acts of loyalty to the Tudors during the War of the Roses, Brigden emphasizes that Henry Wyatt’s ‘heroic suffering [was] revered in family memory’—and that Henry Wyatt himself ‘would not allow his son and grandson to forget’ that recent history (The Heart’s Forest, 66–67). 48 While many biographical details of Surrey’s and Wyatt’s lives are widely known and, I would argue, interpretation of their personalities undergirds most criticism of their works, my own interpretations are particularly informed by Brigden’s New Worlds, Walker’s Writing Under Tyranny, Weir’s The King and His Court, Muir’s Life and Letters, and Childs’s Last Victim. 49 Nevalainen and Raumolin-Brunberg, Historical Sociolinguistics.
Tr ansl ation and the Position-Taking Verse Tr adition
99
against “newfickelness”, hypocrisy, and courtly life itself, which is seen as a huge lie’.50 When, though, Domenichelli’s summative statements describe Wyatt’s ‘strong and direct language’,51 we may move away from Wyatt’s evident awareness of his medium as one which is inherently dependent on interpretation and affect. That interpretive work is meant to convey a ‘deeper’ truth, yes, but the poet is always pointing the reader to the artifice, because only by recognizing the artificial surface can the reader be inspired to dig into the code to unearth the meaning—or cued to do so, as courtiers knew to expect, engage, and move beyond the artifice of the constructed communal code. In contrast, as outlined in the previous chapter, Surrey shows a great capacity to simultaneously condemn the artifice of others and to explain away any such taint to his own honorable actions. This refusal to fully acknowledge his own artifice—and his resistance to the patterns of language use adopted by women and the lower classes—are partly indicative of Surrey’s refusal to participate fully in the highly effective, but potentially feminizing, work of contemporary manuscript production and coded position-taking. The titles which Tottel assigns the two versions of Rime 140 instantly evoke the differences between the two poets, and each one’s approach to the translation. Surrey’s poem is ‘of a louer rebuked’, while Wyatt’s ‘louer for shamefastnesse hideth his desire’:52 ‘Complaint of a louer rebuked’ LOue, that liueth, and raigneth in my thought, That built his seat within my captiue brest, Clad in the armes, wherin with me he fought, Oft in my face he doth his banner rest.
‘The louer for shamefastnesse hideth his desire within his faithfull hart’ THe long loue, that in my thought I harber, And in my hart doth kepe his residence, Into my face preaseth with bold pretence, And there campeth, displaying his banner,
50 Domenichelli, ‘Wyatt’s Translations’, 75–76. 51 Domenichelli, 76. I align more with Simpson’s summary of Wyatt’s style; in his analysis of ‘Each man me telleth’, he concludes ‘The plain style of this poem [… is itself] expressive of a plainness that cannot be direct’ (‘Henrician Ovidianism’, 346). 52 Tottel, ed., Songes and Sonettes, fol. 4v and fol. 19r.
100
Gender and Position-Taking in Henrician Verse
She, that me tought to loue, and suffer paine, My doutfull hope, and eke my hot desire, With shamefast cloke to shadowe and restraine, Her smiling grace conuerteth straight to ire. And coward loue then to the hart apace Taketh his flight, wheras he lurkes and plaines His purpose lost and dare not shewe his face, For my lordes gilt thus faultlesse bide I paines, Yet from my lorde shall not my foote remoue Swete is his death, that takes his end by loue.53
She that me learns to loue, and to suff[er] And willes that my trust, and lustes negligence Be reined by reason, shame, and reuerence. With his hardinesse takes displeasure, Wherwith loue to the hartes forest he fleeth, Leauing his enterprise with paine and crye, And there him hideth and not appeareth. What may I do when my maister feareth, But in the field with him to liue and dye, For good is the life, ending faithfully.54
Even setting aside the added titles, the poems themselves begin with difference: Surrey’s ‘LOue, that liueth, and raigneth in my thought’55 is quite distinct from Wyatt’s ‘THe long loue, that in my thought I harber’.56 As Sacks has observed, in contrast to Wyatt’s ‘long loue’,57 ‘neither Petrarch nor Surrey says anything directly about the length of love’s residency’.58 As Sacks has also pointed out, Wyatt actually emphasizes his addition through the meter, ‘with two immediately consecutive strong accents […] involving a reversed, and caesura-severed, second foot’.59 Sacks argues that Wyatt uses this stress on duration to ‘augment […] inwardness [and] contrast the length of love and the brief social moment in which it will be made manifest’;60 the point is analogous to the problems and limitations 53 Surrey, Songes and Sonettes, ‘LOue, that liueth’, 4v. 54 Wyatt, Songes and Sonettes, ‘THe long loue’, fol. 19r. 55 Surrey, Songes and Sonettes, ‘LOue, that liueth’, 4v, line 1. 56 Wyatt, Songes and Sonettes, ‘THe long loue’, fol. 19r, line 1 57 Wyatt, Songes and Sonettes, ‘THe long loue’, fol. 19r, line 1. 58 Sacks, ‘Face of the Sonnet’, 23. 59 Sacks, 23. 60 Sacks, 23.
Tr ansl ation and the Position-Taking Verse Tr adition
101
of court service, as well. Performance is the only indication of interiority, and so loyalties and disloyalties can be masked or determined based on the actions of only a moment. Here, the social system that is apparently accepted by the speaker of ‘‘CEsar, when that the traytour’ is called into question for its reactionary impulses, even as assumed performance is again emphasized as an appropriate protective measure. Wyatt adds another element to his poem which is not present in Surrey or Petrarch’s poems: an explicit mention of ‘faith’.61 Added to Wyatt’s line emphasizing the role of ‘reason, shame, and reuerence’, multiple and even competing interpretations are not only possible, but immediately apparent.62 The poem becomes a space for philosophical valuation: what kinds of service are best, if service to king, Church, and beloved are not commensurate? Functioning within love becomes functioning within an extant social order, one in which the poet is frequently constrained by the orders of masters of several different levels. Detailing the Old English/Latin clash of ‘lustes negligence’ and the combined use of Old English suffix -nes and Latinate suffix -ence, Sacks argues that ‘the sonnet could almost be seen as a struggle, enmeshed with translation itself, between the two, aligned as they are with a native impulse colliding with an external code’.63 The contemporary political and social spheres of England mirror this tension. England is a country where imperialist ideals of the divine right of kings have been transplanted.64 Though the social structure was still rigidly hierarchical, there was still an important difference in the philosophies of the systems. In some ways, the humanist privileging of demonstrated talent could be understood as a return, rather than a new idea—a different sort of revolution. In others, the ideals of the Reformation that joined to humanist thought undermined completely the established belief systems of an entire country. Wyatt’s poem creates a space that privileges service even within a tradition that often chafes at its requirements. The self is created in reaction to the external—particularly when that external is demanding conformity and obedience. One must be like, without equaling, the King. It may be in failures of imitation, though, that more of the self can be found, like the significance of manipulated interruptions in pentameter or the alteration, 61 Sacks, 23. 62 Wyatt, ‘THe long loue’, line 7. 63 Sacks, ‘Face of the Sonnet’, 25–26. 64 Adjacent to the apparent tension of hierarchy and humanism, for Henry, would have been his own understanding of his royal power, as outlined in the first chapter.
102
Gender and Position-Taking in Henrician Verse
through transcription, of an accepted and well-known poem. In the final lines, ‘the turn to the question of agency is made inseparable from an explicit turn toward the reader’.65 This turn involves, implicates, and educates the reader. As the speaker questions the reader as to the appropriate action, ‘the poem produce[s] the potential for many faces—its own repeating face and […] the prospect of any readers’ pictured faces’.66 First, those many faces may draw in the community of manuscript writers; communal production and protection are key to the larger system, and Wyatt participates in that system (where Surrey prefers insistent individuality). Also, though, the poem becomes a kind of simulacrum, replicating a speaker who does not exist and a code that is dependent on the recognition of its subject and object as not real, or at least as a performance which can be adopted and adapted by anyone. This kind of replication extends to the use of Petrarchan love lyrics throughout the Henrician court. The ‘nonexistence’ of the idealized Laura must first be understood; then, the poems must be about more than a ‘beloved’, since the beloved is not and cannot be real. The replications, further, become commentaries on themselves—each performance which comments on performance critiques, modifies, pastiches what has come before. At times, when the poems are read through in a collection like Tottel’s or in one of the contemporary manuscripts, the effect becomes nearly absurdist; the repetition of scenarios and of the words related to the theme—love, fleeing, serve, etc.—become so layered that the lack of realized object becomes inescapable, like the widely known effect wherein a repeated word seems to lose meaning—or to lose its original meaning and become instead a symbol of the arbitrary nature of language. This, then, is the effect of the repetition of Petrarchan tropes; the verse becomes a symbol of the arbitrary nature of social performance and socially assigned roles and of the extent to which those tropes have also become the ‘natural’ order of things.67 While the practitioners who worked with these tropes certainly knew that no literary convention served as a strict reflection of reality, they seem to stretch these particular conventions to the limit. I argue that this stretching is partly meant to draw the readers’ attention to the effort of the performance and so the possibility of alternate performances. 65 Sacks, ‘Face of the Sonnet’, 28. 66 Sacks, 29. 67 As Lorna Clymer outlines in her introduction to Ritual, Routine, and Regime, ‘repetition [can] be understood as an attempt to impose continuity on incongruities’, including in efforts to establish the shared identity practices through which national (or, I would add, social) communities are delineated (3).
Tr ansl ation and the Position-Taking Verse Tr adition
103
The strictures of these roles are more evident in Surrey’s version of the poem than is any particular ability to overcome or manipulate such tropes. This regularization is apparent even in the meter, as Sacks sees in the greater correspondence to form in ‘Surrey’s immediate march, after the first foot, into the lockstep of normative iambic pentameter’.68 The regularity of form is matched by a regularity of tone; Domenichelli sees a contrast between ‘Surrey’s […] courtly language of politeness and indirectedness [and] Wyatt’s […] directedness, and even outspokenness’.69 However, part of the art of Wyatt’s version is actually in the indirection—in the delay of clause endings, for example. In fact, both poets tend towards different sorts of directness and outspokenness; Surrey’s work is often far from any masterpiece of polite indirectness. The difference is partly in the clarity and rank of the objects of their critiques, and the strategies match the men’s positions and ambitions. Wyatt aims to level the playing field. His language is simpler, his critique is of the social system, and he intentionally participates in communal position-taking practices. Surrey aims to maintain his place at the top—and so his language is more individal and elevated and his critique is of those who oppose him or who he feels undermine the aristocratic system. Domenichelli helpfully outlines the word choices that make Wyatt’s love more ‘common’ than Petrarch’s or Surrey’s;70 Surrey’s love reigns on a throne, setting forth arms and banners, where Wyatt’s harbors in a residence, pressing forth, pretentiously.71 Surrey’s Love is nonetheless cowardly,72 and, if we consider Love to be the parameters and restrictions of the contextual social system, following the established relationship between love and service, this makes perfect sense. Surrey sees the structure as fallen away from earlier ideals. Love’s martial unfitness aligns with one of the major problems Surrey sees in Henry’s court, as emphasized in his ‘Th Aseirian king in peace, with foule desire’.73 Love is a coward, but the social rules require that the speaker follow his ‘master’—perhaps as Surrey views the King as martially cowardly, but must follow his rules and restrictions nonetheless. There is, ultimately, an energy in Wyatt in contrast to a lack of vivid agency in Surrey—Wyatt is declaring allegiance to particular social constructs, while Surrey is merely decrying the failures of the extant construct. In Surrey’s conclusion, ‘the speaker is a captive, defeated by a conquering 68 Sacks, ‘Face of the Sonnet’, 23. 69 Domenichelli, ‘Wyatt’s Translation’, 80. 70 Domenichelli, 81. 71 Domenichelli, 81. 72 Surrey, Songes and Sonettes, ‘LOue, that liueth’, 4v, line 9. 73 Surrey, Songes and Sonettes, fol. 17v. For more on this work, see the previous chapter.
104
Gender and Position-Taking in Henrician Verse
interloper who has built his own throne, from which he now reigns’.74 The parallel to Henry, given Surrey’s views, is irresistible. Love represents both the King and the social order he has introduced and reinforced. Surrey’s speaker in his poem does still see his role as to fit in to some degree, but we also see the resistant fissures that would lead to the poet’s death. Love ‘is no guest in Surrey’s version; he is an invader who now maintains allegiance by sheer force of arms’.75 This invasion sharply figures ‘[t]he urgent drive of the passions beyond rational human restraint, in complete and tyrannous disregard of any willed self-control’.76 Harris agrees with Thomson that Surrey’s poem departs from ‘the big moral issues’,77 but, from Surrey’s perspective, the poem may offer a commentary on just such issues. The speaker has been subjected to a force he resents, a force that he must serve, but that he sees as ignoring the rules of decent, noble conduct. The correlation speaks to Surrey’s resentment of Henry, and his conviction that this was because he was morally more ‘right’ than his King. The modern discourse on toxic masculinity encourages an extension of this reading of Surrey’s resistance and resentment. The military figuring of love that Surrey so emphasizes reflects a different understanding of both politics and human affairs than that in Wyatt’s verse. There is no comfortable, established fealty or allegiance, here; to be conquered and occupied is to be always conquered, always resistant. Love still wears his armor,78 offering both a show of his martial victory and a suggestion of the threat presented by the resistant speaker. The beloved is more fickle—where Wyatt’s lady more mildly ‘takes displeasure’,79 an apparently consistent position against the speaker’s injudicious displays of love, Surrey’s moves ‘straight’ from ‘smiling grace’ to ‘ire’.80 Most telling is the speaker’s claim that ‘For my lordes gilt thus faultlesse bide I paines’. 81 This insistence offers a troubling echo of the trope of performed masculine helplessness at the hands of desire—the implication that the temptation presented by a woman is insurmountable and that men are subject to ungovernable, irresistible drives. While the martial imagery of both poems may be considered masculine, by an early modern understanding 74 Harris, ‘Surrey’s Creative Imitation’, 300. 75 Harris, 300. 76 Harris, 301. 77 Thomson, ‘First English Petrarchans’, 91; Harris, ‘Surrey’s Creative Imitation’, 301. 78 Surrey, Songes and Sonettes, ‘LOue, that liueth’, 4v, line 3. 79 Wyatt, ‘THe long loue’, line 8. 80 Surrey, ‘LOue, that liueth’, line 8. 81 Surrey, line 12.
Tr ansl ation and the Position-Taking Verse Tr adition
105
of gender, Surrey’s choices emphasize both that masculinization and its most troubling manifestations. Surrey’s poem contains a contrast and separation similar to Wyatt’s, but to a very different effect. Like Wyatt, Surrey pits Love-as-master and the mistress as completely separate, though he excludes the vague elements of religious devotion that color Wyatt’s poem. Indeed, Love and the mistress are so separate that, as Harris observes, the mistress seems unaware of Love’s responsibility for the speaker’s actions.82 We might, of course, also question this surrender of agency on Surrey’s part—the claim, as it were, that he can’t help himself. Harris argues that in this conflict, between the interior driving force and the exterior controlling element, Surrey ‘recreates that conflict between the inner and outer existences of the passion-driven man’;83 I would add that, more specifically, Surrey recreates the conflicts of the man constrained by a highly structured social system—and, most particularly, of the man in a position of sufficient privilege to interpret limitation or constraint as oppression and enslavement. The specific intersection of that social system and privilege represent the point of tension for Surrey; the rules are changing, and the result is less power for the speaker. The poem dramatizes a problem of allegiance, but within a different set of problems than that of Wyatt’s version—the system that the speaker understands to be proper, at odds with the greater force of the more inappropriate, domineering ‘love’. This ‘Love in defeat is a blusterer turned coward’;84 because of the limitations of such a lord ‘no loving relationship between liege and yeomen has been developed’.85 Ultimately, Harris sees the end of the poem as a pledge of allegiance to the sensual, reasonless tyrant, but I see, in the final lines, a more resistant, resentful reading. If the speaker claims that ‘Yet from my lorde shall not my foote remoue’, because ‘Swete is his death, that takes his end by loue’, an interpretation is available that sees this love, lowercased and not personified, as loyalty to the ‘better’ system.86 The speaker can potentially be turned against the usurping Lord, in an act of disobedience that will claim his life. Where Wyatt aims for preservation of self in a system where conflicting allegiances cannot be reconciled, Surrey’s speaker must be turned either against the mistress, against his master, or against the system of which he complains. 82 83 84 85 86
Harris, ‘Surrey’s Creative Imitation’, 302. Harris, 302. Harris, 304. Harris, 304. Surrey, ‘LOue, that liueth’, lines 13–14.
106
Gender and Position-Taking in Henrician Verse
To those works of Wyatt and Surrey’s included in Tottel’s, I want to add some contemporary translation from the Devonshire, indicating the multiple platforms available for these kinds of resistant writing and to emphasize the connection between the published work and its original manuscript contexts. The works discussed here from the Devonshire are both in unidentified or contested hands, Hands 1 and 8.87 Hand 8 frequently transcribes works by Wyatt; the poem here described was included in Tottel as Wyatt’s and has been included in later edited collections of his works. Hand 1 uses a translation from Pietro Bembo, the figure to whose ideals Domenichelli compares both Wyatt’s and Surrey’s language,88 and in the translation creates a sustained metaphor of the forces of love on the speaker as warring elements of fire and water, each keeping the speaker from the death that would represent release. The work offers opportunities to analyze what can be lost as poetry is regularized. In the same way that Tottel’s added titles constrict the meaning of the poems in his collection, modern editions of these poems can overwrite the strategies of the manuscript writers. As an example, I would transcribe the closing lines of the first stanza thusly, following the transcription of A Social Edition: ‘Whatt maye I more / sustayne alas that dye wuld fanne / and cane not dye for panne’.89 Heale’s addition of punctuation in her edition of the poetry, moving ‘sustain’ to the preceding line, setting off ‘alas’ with commas, and joining all of the lines with a closing question mark, clarifies the meaning, but risks a correction where an analysis can be especially rewarding.90 The rhetorical effect of ‘What may I more sustain?’ is quite different from the alternative opened by movement of the enjambment, which potentially reads ‘What may I more?’ with the answer ‘sustain, alas, that die would fain / and cannot die for pain’.91 In the edited version, the 87 Heale, ed., Devonshire. Throughout, my hand reference agrees with the numbering system used by A Social Edition and so, generally, with Heale. In later chapters, I explore my use of ‘she’ when referring to Hand 1, but the gender reference is not particularly important, here, except to reinforce the necessity of avoiding default to a masculinized interpretation of Henrician verse. 88 Domenichelli, ‘Wyatt’s Translations’, 65–86. 89 MS BL Add. 17492, 4r–4v, ‘At last withdraw’, lines 7–9. I have confirmed my transcripts against the original; however, I am indebted to the groundwork and paleographic samples created by the compilers of A Social Edition and to the interpretive background laid by Heale’s modernized edition. 90 Heale, ed., Devonshire, 56, ‘At last withdraw’, lines 7–9; MS BL Add. 17492, ‘At last withdraw’, lines 7–9. My point here is not a disagreement with Heale, whose contributions in the modernized edition are invaluable; all editorial decisions around this manuscript will necessarily circumscribe meanings. 91 MS BL Add. 17492, ‘At last withdraw’, lines 8–9.
Tr ansl ation and the Position-Taking Verse Tr adition
107
rhetorical question is essentially limited to a question of the ability of the speaker to continue in pain. In contrast, the reading of the line as question and answer creates a stoicism in the poetic voice. The speaker knows their fate and accepts the inevitability of the speaker/lover’s subjection to the beloved object. In either case, death is presented as an escape for the speaker, but an unavailable escape. Heale further corrects the ‘he’ of Line 28—‘he wolde haue my death’—to ‘ye’,92 deemphasizing, but not eliminating, the potential of the poem both to refer rather more explicitly to Henry VIII and to disrupt the traditional, though hardly uniform, gendered roles of Petrarchan tropes. The gender ambiguity of the poem may be undercut by the final, defiant lines: ‘no man alyve nor I / of doble dethe can dy’.93 However, the ‘man’ here seems more likely to work as a referent to humankind. The speaker here charges the object of the poem with keeping them alive only because death would end the speaker’s torment and object’s pleasure. Read as the voice of a courtier under Henry’s reign, though, the lines also call attention to a defiant streak which claims perseverance in the face of the King’s unpredictable, inconstant, and threatening behavior. The second and third stanzas of the poem offer an especially elegant example of the manuscript’s interest in contraries—a theme that applies both to the King’s inconstant behavior and to the courtier’s balance of performance and preserved interiority. The metaphor of balanced heat/ dryness and cold/wetness is conventional, of course, but the use here is nicely sustained and complex. The fire of the beloved should burn the speaker’s thought and desire—the ‘though’ for ‘thought’, in ‘my though and mye desyre’ marks into the spelling a sort of foreshadowing of ‘shulde’ in the following line.94 What should happen does not; the beloved sends ‘a stormy rayn’ that both quenches the burning fire before it can destroy the speaker and becomes the speaker’s tears.95 The new threat is of drowning, but a potentially ambiguous heat then saves the speaker; the line reads ‘The heate’,96 allowing the first word to be read as the article or as the possessive pronoun. Conventionally, of course, and given the earlier clarif ication that ‘ye burne / my though and mye dysyre’,97 the burning heat and the 92 Heale, ed., Devonshire, 57, ‘At last withdraw’, line 28; MS BL Add. 17492, ‘At last withdraw’, line 28. Heale does agree that the reading is ‘he’ in the monograph, in her footnote to the line. 93 MS BL Add. 17492, ‘At last withdraw’, lines 35–36. 94 ‘At last withdraw’, lines 11–12. 95 ‘At last withdraw’, line 14. 96 ‘At last withdraw’, line 21. 97 ‘At last withdraw’, lines 10–11.
108
Gender and Position-Taking in Henrician Verse
drowning rain are both attributed to the contrary lover. Nonetheless, the form of these particular lines offers a reading of this last heat—the heat which saves the speaker from drowning but keeps the speaker in the hurt to which they stoically submit in the first stanza—as not necessarily within the beloved’s control. The effect of these elements together is to suggest a kind of resistant resilience, emphasized by the defiant tone of the final stanza. Regardless of the mercurial demands of the person who controls their service, the speaker will carry on, since they can only once be subjected to the ultimate punishment of death. The other selection examined here from the Devonshire is not a strict translation of any known poem, but draws heavily on a poem by Aquiliano. Partially because of this association, the poem has often been attributed to Wyatt, who frequently reworked Aquiliano’s poems, but the attribution is not certain.98 While the defiant tone echoes the themes of the earlier poem, the escape here is in a potentially embittered freedom, rather than in the ownership of acquiescence offered in ‘At last withdraw’. The poem Hand 8 writes details, in each stanza, tropes of the spurned Petrarchan lover, but responds to each situation with the mocking refrain ‘But ha, ha, ha full well is me, / for now I am at liberty’, undercutting the destructive potential of the beloved’s power by claiming an already achieved escape.99 Muir and Thompson acknowledge the potential of the ‘ha, ha, ha’ of the refrain as a referent to the poems of Aquiliano, but the widespread use of monograms for the royal couples may open up an additional interpretative outlet.100 Given the use of acrostics throughout the manuscript, the repeated, joined letters could easily reference such initials. The HA, in this reading, may emphasize through repetition the union of Henry and Anne—and, potentially, depending on the position of the reader, Henry and either Anne.101 If written by Wyatt, the poem may refer to his supposed interest in Anne Boleyn, particularly given the clearly gendered object ‘she’ in Lines 24–25. In these terms, the ‘ha’ refrain takes on the sarcastic, bitter tone we 98 Muir and Thompson include the work in their Collected Poems (227), where they acknowledge Aquiliano as a potential source, citing Koeppel (426). Heale cites Muir and Thompson’s note on the potential source text, but does not engage the attribution, while the Social Edition notes mention R.A. Rebholz (Sir Thomas Wyatt: The Complete Poems, 1978) for both attribution and source identification. 99 MS BL Add. 17492, 79v, ‘Tanglid I was’, lines 5–6. 100 Muir and Thompson, eds., Collected Poems, 426. 101 Anne of Cleves’s brief marriage to Henry in the first part of 1540 likely occurred later than most of the composition of the manuscript, but the later entries indicate that readers after the first burst of entries in the 1530s might still have had access to this interpretation. Such a reading would be especially relevant to the discussion here on women’s courtiership.
Tr ansl ation and the Position-Taking Verse Tr adition
109
associate with Wyatt’s wit, and which his contemporaries would likely have associated with his compositions as well. Even if not written by Wyatt, the courtly context makes the reference to an embittered, disappointed suitor of Boleyn available: stories of such lovers were current in the court and, as my analysis of manuscript production throughout emphasizes, stories of disappointed lovers shaped the ways that courtiers reproduced and received all of their peers’ poetic creations. However, if copied by any of the women courtiers, the relative isolation of that gendering may work to leave the poem open to a somewhat unusual application of the Petrarchan tropes to Henry VIII—perhaps moving beyond typical applications of love poetry to the service of courtiership and instead speaking simultaneously about love and courtiership. Certainly, following Anne Boleyn’s and Jane Seymour’s meteoric rises, any woman courtier who caught the King’s eye could realistically entertain further ambitions. Drawing on contemporary proverbs, the stanzas repeat traditional warnings about the dangers of courtiership—dangers that were particularly marked for women courtiers. This reading of the poem then frames terms of escape in less cynical or sarcastic terms; the escape of a woman from Henry’s attentions may be a genuine relief, even if ambition temporarily numbed her to the dangers of her potential rise. This possibility also underlines the relative unimportance of gendering specific hands or speakers once women’s larger place in manuscript production is restored. Because the poetry was produced, or at least reproduced, in a work so aligned with the women courtiers of the time, any poem that could be understood as being in a woman’s voice always had that interpretation readily available to the original audience. This original audience—a community of cooperative writers and readers—is largely what my work here attempts to emphasize. Placing Wyatt’s and Surrey’s published works alongside these selections from their contemporaries highlights the larger system of coding at play in manuscript position-taking work. Tottel’s imposition of titles on the love lyrics he published both indicate the potential reach of that system, as they may reflect an effort to redirect subversive or resistant energies, and indicate the potential misinterpretations of Henrician verse when the system is understood as one created exclusively by courtly men. The tendency to impose what have been understood as contemporarily normative gendered relationships on verses that are more ambiguous has partly contributed to the erasure of more rewardingly messy readings. Integrating work with manuscripts where women’s strategies predominate encourages us to reconsider the strategies and concerns of the Henrician court poets.
110
Gender and Position-Taking in Henrician Verse
That court poets used Petrarchan love lyrics to express frustrations at courtly service is still important, however well-established, but from here we need to move further and forward. Because these authors understood that elements of their work would be interpreted in certain ways by particular readers, we can begin to recognize elements of a system of coding that allowed courtiers to communicate to one another their discomfort, discontent, and disillusionment with the shifting tensions of the Henrician court. While these expressions sometimes included alignments that we would now see as essentially conservative, as in Surrey’s privileging of aristocratic entitlement, the expressions were often radical in that they expressed a challenge to the King’s agenda—and in that they used patterns valued by the courtly system he had established for that expression. That these same patterns are used in the manuscript production of the time emphasizes the likelihood that courtiers used the strategies that have been so well-examined in translation in their original composition and in their copywork, as well, and that the transfer was not simply in one direction. This likelihood, and the various strategies used in manuscript production first explored in the introduction, will be more thoroughly engaged in the final section of this work. For now, we move into a further contextual consideration: how might we recontextualize the way gender functions within translated verse and re-interpret the position-taking work of translation?
Works Cited Brigden, Susan. New Worlds, Lost Worlds: The Rule of the Tudors, 1485–1603. New York: Penguin, 2002. Brigden, Susan. Thomas Wyatt: The Heart’s Forest. London: Faber and Faber, 2012. Childs, Jessie. Henry VIII’s Last Victim: The Life and Times of Henry Howard, Earl of Surrey. New York: St. Martin’s, 2007. Clymer, Lorna. ‘Introduction’. In Ritual, Routine, and Regime: Repetition in Early Modern British and European Cultures, edited by Lorna Clymer. Toronto: U of Toronto P, 2006. Crane, Mary Thomas. ‘Roman World, Egyptian Earth: Cognitive Difference and Empire in Shakespeare’s Antony and Cleopatra’. Comparative Drama 43 (2009): 1–17. The Devonshire Manuscript. 1525?–1567? MS BL Add. 17492. British Lib., London. Domenichelli, Mario. ‘Sir Thomas Wyatt’s Translations from Petrarch and the “Lingua Cortigiana”’. English Studies in Italy 15, no. 1 (2002): 65–86.
Tr ansl ation and the Position-Taking Verse Tr adition
111
Edwards, A.S.G. ‘Print and Manuscript: The Text and Canon of Surrey’s Lyric Verse’. In In the Prayse of Writing: Early Modern Manuscript Studies. Essays in Honor of Peter Beal, edited by S.P. Cerasano and Steven W. May, 25–43. London: British Library, 2012. Greenblatt, Stephen. Renaissance Self-Fashioning. Chicago: U of Chicago P, 1980. Greene, Roland. ‘Petrarchism among the Discourses of Imperialism’. In America in European Consciousness, 1493–1750, edited by Karen Ordahl Kupperman, 130–165. Chapel Hill: U of North Carolina P, 1995. Hall, Kim F. Things of Darkness: Economies of Race and Gender in Early Modern England. Ithaca: Cornell UP, 1995. Harris, William O. ‘“Love That Doth Raine”: Surrey’s Creative Imitation’. Modern Philology 66, no. 4 (1969): 298–305. Heale, Elizabeth, ed. The Devonshire Manuscript: A Women’s Book of Courtly Poetry. Toronto: Center for Reformation and Renaissance Studies, 2012. Heffernan, Megan. Making the Miscellany: Poetry, Print, and the History of the Book in Early Modern England. Philadelphia: U of Pennsylvania P, 2021. Herman, Peter C. ed. Reading Monarchs Writing: The Poetry of Henry VIII, Mary Stuart, Elizabeth I, and James VI/I. Tempe, AZ: MRTS, 2002. Herman, Peter C. Royal Poetrie: Monarchic Verse and the Political Imaginary of Early Modern England. Ithaca: Cornell UP, 2010. Hunt, Alice. The Drama of Coronation. Cambridge: Cambridge UP, 2008. Lockey, Brian. Law and Empire in English Renaissance Literature. New York: Cambridge UP, 2006. Loomba, Ania. ‘The Theatre and the Space of the Other in Antony and Cleopatra’. In Shakespeare’s Late Tragedies: A Collection of Critical Essays, edited by Susanne L. Wofford, 235–248. Upper Saddle River, NJ: Prentice Hall, 1996. Marotti, Arthur F. ‘“Love Is Not Love”: Elizabethan Sonnet Sequences and the Social Order’. English Literary History 49, no. 2 (1982): 396–428. Marotti, Arthur F. Manuscript, Print, and the English Renaissance Lyric. Ithaca: Cornell UP, 1995. May, Steven W. ‘Popularizing Courtly Poetry: Tottel’s Miscellany and Its Progeny’. In Pincombe and Shrank, 418–433. Muir, Kenneth. Life and Letters of Sir Thomas Wyatt. Liverpool: Liverpool UP, 1963. Muir, Kenneth and Patricia Thomson, eds. Collected Poems of Sir Thomas Wyatt. Liverpool: Liverpool UP, 1969. Nevalainen, Terttu and Helena Raumolin-Brunberg. Historical Sociolinguistics: Language Change in Tudor and Stuart England. New York: Pearson, 2003. Park, Jennifer. ‘Discandying Cleopatra: Preserving Cleopatra’s Infinite Variety in Shakespeare’s Antony and Cleopatra’. Studies in Philology 113 (2016): 595–633.
112
Gender and Position-Taking in Henrician Verse
Pincombe, Mike and Cathy Shrank, eds. The Oxford Handbook of Tudor Literature, 1485–1603. New York: Oxford UP, 2009. Pincombe, Mark and Cathy Shrank. ‘Prologue: The Travails of Tudor Literature’. In Pincombe and Shrank, 2009. 1–17. Sacks, Peter. ‘The Face of the Sonnet: Wyatt and Some Early Features of the Tradition’. In Green Thoughts, Green Shades: Essays by Contemporary Poets on the Early Modern Lyric, edited by Jonathan F.S. Post, 17–40. Berkeley: U of California P, 2002. Simpson, James. ‘Breaking the Vacuum: Ricardian and Henrician Ovidianism’. Journal of Medieval and Early Modern Studies 29, no. 2 (1999): 325–355. A Social Edition of the Devonshire Manuscript MS (BL Add. MS 17492). Wikibooks, The Free Textbook Project. https://en.wikibooks.org/w/index.php?title=The_Devonshire_ Manuscript& oldid= 3469218. Accessed 1 July 2022. Songes and Sonettes, Written by the Right Honorable Lorde Henry Howard Late Earle of Surrey, and Other. Ed. Richard Tottel. London: Tottel, 1557. Thomson, Patricia. ‘The First English Petrarchans’. Huntington Library Quarterly 22, no. 2 (1959): 85–105. Walker, Greg. Writing under Tyranny: English Literature and the Henrician Reformation. Oxford: Oxford UP, 2005. Warner, Christopher J. The Making and Marketing of Tottel’s Miscellany, 1557: Songs and Sonnets in the Summer of the Martyr’s Fires. Burlington, VT: Ashgate, 2013. Weir, Allison. Henry VIII: The King and His Court. New York: Ballantine, 2001.
4. Men’s and Women’s Approaches to Translation and Authorityin the Late Henrician Court Abstract: This chapter focuses on translation from classical and ancient sources, engaging Surrey’s Aeneid and Wyatt’s Psalms and considering Katherine Parr’s uses of biblical and political authority. Parr’s work offers a bridge into an examination of the use of translated work from verse into prose by Elizabeth I and Jane Lumley. These works are considered, together, as translations of texts that had already been assigned cultural importance and authority, thus increasing the interpretive cultural framework. This section on translation emphasizes the interpretive frameworks audiences were expected to bring to their reading and offers us practical demonstrations of how power—inflected by gender, status, and age—is played out on the page. Keywords: Surrey’s Aeneid; Penitential Psalms; Katherine Parr; Elizabeth I’s Miroir; Jane Lumley; Tudor translation
This chapter continues investigation into the functions of translation as a position-taking strategy, but the focus here shifts to translation from classical, ancient, and royally authorized sources, engaging translations like Surrey’s Aeneid and Wyatt’s Penitential Psalms alongside work by Katherine Parr and two prose translations created by young Englishwomen out of foreign language verse: Elizabeth Tudor’s Miroir and Jane Lumley’s Iphigeneia. On one hand, the canonical texts from which Surrey, Wyatt, and Lumley work may offer greater constraints than the Italianate love poetry examined in the previous chapter. These canonical authors had, after all, already famously highlighted particular concerns and themes; while the same can be said of Petrarch, certainly, the weight of the traditions were simply different. These works were often approached as more sacrosanct and
Quoss-Moore, R.M., Gender and Position-Taking in Henrician Verse: Tradition, Translation, and Transcription. Amsterdam: Amsterdam University Press, 2023 doi 10.5117/9789463723534_ch04
114
Gender and Position-Taking in Henrician Verse
less open to entirely new interpretations.1 This argument was particularly foregrounded in translation of religious texts, and further emphasized by the contemporary controversy on whether vernacular translation of religious texts was even appropriate. On the other hand, though, this offered a cultural store of response which writers might manipulate in particularized ways, either working through their own tensions as mirrored in the classical texts or manipulating their content to match or frustrate perceived desires and expectations. While a similar cultural resource was accessed when writers translated courtly love lyrics, the sheer weight of interpretive tradition associated with Virgil, the Psalms, or Euripides increases the translator’s interaction with, and reaction to, the cultural mythos surrounding each work. Some elements of translation as position-taking work transfer fairly directly from the previous chapter; as Danielle Clarke summarizes, ‘the act of translation itself may be a highly coded political or ideological intervention where the guise of another’s words can help to evade the authorities’.2 In fact, through the translation of culturally important texts, authors could not only evade the authorities, but also gain legitimacy through official endorsement by those same authorities who might resist similar ideas had they been presented as entirely new works. Nonetheless, as Margaret Tudeau-Clayton notes, ‘to translate carried significant sociopolitical and ideological stakes’.3 Clarke acknowledges the useful counterbalance to these stakes—the ‘relative marginality or slipperiness of ownership [which] can be exploited as a form of agency to figures who otherwise lack it’. 4 This obfuscation of ownership, already examined in the previous chapter, was likely particularly important to the young women whose translations are examined in the last section here. This last section of the chapter, on two young women’s translations, is included to emphasize the contemporary awareness of the resistant energies of verse translations. Surrey’s Aeneid and Wyatt’s Psalms both offer the poets imaginative access to positions of power. While both poets critique 1 Tudeau-Clayton outlines translators’ work to represent the (contradictorily figured) complete and perfect whole of Virgil’s work, as one example of this concern with interpretation (TudeauClayton: ‘Supplementing the Aeneid’); Zaharia’s overview of early modern theories of translation outlines such concerns as the obligation to the original author (‘Theories of Translation’, 15), scrupulous and literal translation (17), and the particular difficulty of translating sense in certain cases, as with Aristotle or the Bible (18). 2 Clarke, ‘Translation’, 169. 3 Tudeau-Clayton, ‘Language, Verse, and Politics’, 390. 4 Clarke, ‘Translation’, 170.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
115
the performances of power by their leader/subjects, they both also align themselves with the authority of these men’s positions by underscoring their shared masculine performance. In addition to conferring authority on the critiques and valuations the two men outline in their translations, this authority may also be intended to legitimize the linguistic innovations Surrey and Wyatt add to their versions of the texts—and through which they seek further authority as masterly courtly makers. In contrast, both Elizabeth and Jane Lumley choose to work away from the verse versions they translate, moving their work to prose.5 Parr’s first two texts—the translated Psalms or Prayers taken out of Holye Scripture and the translated and hybridized Prayers or Medytacions—offer an especially interesting bridge between the translated verse work explored thus far and the prose translations of the younger, courtly women. All three women offer visions of rhetorical authority for their women narrators or characters, but that authority is bounded by appeals to a greater masculinized power. In a court that understands the potential of verse translation as resistance to authority, the young translators choose to instead protect their voices through greater appeal to the system, choosing not to activate the position-taking energy implicit in verse production. Partly, Parr’s texts offer a model for the younger translators, but her text is also considered here as a complex hybrid of manuscript and print, of verse and prose strategies, and of communal and individual authorities.
A. Surrey’s Aeneid Why does Surrey choose to translate specifically Books Two and Four of the Aeneid? Of course, Surrey’s execution at only thirty may have put an end to any plans for a larger project, and indeed proposed dates of composition 5 The later dates of these women’s translations should also be considered in context of the shifting ideas about translation. Clarke considers the possibility that this feminization of the concept of translation as a form of submission arises in order to differentiate such activity from the apparently more productive textual encounters with antecedent texts through imitation and copia, but […] perhaps also suggests some notion of suitability based on homology, where the cultural function of both translation and women is to reproduce and transmit cultural and economic capital. (‘Translation’, 172) I would consider that this shift might also arise precisely to contain the energy of texts like Lumley’s, which, as Uman argues, articulates ‘the contradictory goals of a humanist education for those who are praised for silence and obedience’ (Women as Translators, 77). As translation as a position-taking strategy is accessed and used by women, that is, patriarchal structures become more invested in feminizing that act, thus, in their systems, mitigating its power.
116
Gender and Position-Taking in Henrician Verse
have ranged from 1539 to 1545.6 The later date would place the translation fairly immediately before Surrey’s final fall from favor, but Florence H. Ridley convincingly argues for the earlier end of these possible dates, showing that Surrey’s recorded travels and the apparent influences for the text favor a late 1539 to early 1540 date.7 In either case, though, no evidence has ever been offered that indicates that Surrey intended a complete version of the work, whether he had eight years or only two for such consideration. At a minimum, we can proceed from the assumption that the evidence suggests Surrey thought translation of these books, together, was the more immediately important or satisfying project. Both David A. Richardson and Ridley see this choice as a stylistic one, tied to Surrey’s ambition to make high art of the English language. Richardson points out that Books Two and Four contain ‘about five times as many’ epic similes as Books One and Three;8 he connects Surrey’s interest in these books to ‘Surrey’s intent to elevate his diction through figurative language’.9 Richardson partially acknowledges the political element of such elevation, but he constrains his argument largely to the politics of language. To take the argument further, the connection to Surrey’s larger, often essentially egoistic, project is essential. A stylistic choice is not an apolitical one, and one of Surrey’s major ideological drives was not only the worth of language, as Richardson suggests, but also, by association, the worth of the man who can use language correctly. Although Richardson primarily studies the linguistic, and peripherally politico-linguistic, implications of Surrey’s writing, several of his points emphasize the humanist values which Surrey espouses by implication in his text. Richardson acknowledges that Surrey is ‘not among the humanists who read the classics as literature of knowledge rather than as literature of power’10 and ‘does not represent a sterile tradition of neoclassicism among the humanists’.11 Instead, as shown in his linguistic choices, his ‘ideal of neoclassical imitation transcends didactic commentary and mechanical translation’.12 The outcome of these choices and ideas, though, is more than an achievement for the English language. Surrey’s project of promoting power for his native language is an important leg of his larger political platform. Precisely because Surrey understands these classics as ‘literature of power’, 6 7 8 9 10 11 12
Ridley, ed., ‘Introduction’, The Aeneid, trans. Surrey, 1–4. Ridley, 3–4. Richardson, ‘Humanist Intent’, 215. Richardson, 216. Richardson, 217. Richardson, 218. Richardson, 218.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
117
his adaptation of the materials should be understood as a move in the larger game of political position-taking—an understanding that his act of verse translation would have activated for the courtly audience. Given the poetic tradition and humanist ideal of speaking truth to power through art, Surrey seems likely to understand Virgil’s project as simultaneously propagandistic and critical. As such, Surrey’s choice of books needs to be approached on the terms of the content as much as by the style. In the separate books of the Aeneid, Virgil introduces complications and counters to the Roman values that the piece as a whole more generally espouses.13 When certain books are selected from the whole, some of the problematized values can be more clearly endorsed, and some of the mitigated understandings of vices are simplified. By choosing Books Two and Four, Surrey emphasizes a tragic love story and a specific performance of masculinity. Surrey’s propaganda is both for the power of the language, as Richardson and Ridley argue, and for the worth of noble men; he may also gesture towards the necessity of following a larger destiny at all costs. Surrey’s criticism is of false oaths and of getting carried away with passion—particularly when that passion becomes an impediment to effective rule. By bounding this propaganda and criticism within the translation, Surrey protects what would otherwise be open resistance; as a position-taking text, the translation has greater appeal to tradition, a greater potential for further distribution, and greater protection for the poet than a more overt personal screed. Unlike his work with the more malleable Italianate love poetry, Surrey’s translation of the epic poetry of the Aeneid does not so often introduce new elements of critique or valuation as instead emphasize selective threads from the original piece through deletions, isolation of these particular books, and careful word choice—highlighting some of the strategies that courtly manuscript transcription, as further considered in the final two chapters, adapted from and contributed to work in translation. Throughout Surrey’s translated books, the events recorded by Virgil seem to shadow strong parallels in Surrey’s own life. I suggest that rather 13 Tudeau-Clayton offers a consideration of the ‘parts’ of Virgil in early modern English translation in her ‘Supplementing the Aeneid’. Carlson offers a similar conclusion to my own with regard to Surrey’s selections where, he writes, Surrey elected to concentrate on both the prototype and the antitype—Troy and Carthage—of Vergil’s Roman imperium sine fine, power without end, built on the ruin of these others. So Surrey’s elections tell his purposes: Aen. 4 counts the costs of erotic furor, personal and political, and the costs of the pursuit of hegemony, for the absolutist as well as those touched by his ambition. The unexpected Aen. 2 revels in power’s fall. (‘Henrician Courtier’, 155)
118
Gender and Position-Taking in Henrician Verse
than being purely coincidental, or, more precisely, naturally present only because of Surrey’s own affinity for the text, these parallels are emphasized because of Surrey’s abridged approach to the text, which necessarily alters the narrative arc, truncating context, and by Surrey’s linguistic choices. These linguistic choices, as Ridley carefully emphasizes, are themselves partially the product of Surrey’s use of Gawin Douglas’s Scots Eneados. Ridley traces notable influences from Douglas in ‘more than 40 per cent of [Surrey’s] lines’;14 she notes that in Book Two these influences are particularly marked in passages several lines long, whereas in Book Four ‘they tend to take the form of striking single words or brief phrases’.15 In a text with such a long history, choosing close work with a translation already nearer Surrey’s own native language could suggest a greater drive towards tailoring the text. The Scots translation may have also served Surrey’s investment in valorizing the languages of Britain. Throughout Surrey’s version of the text, the selections he makes, whether drawn or departing from Douglas’s work, reliably underscore particular patterns of valorization of nobility, suspicion of the common people, and indictment of rulers who fail to live up to their potential or promises. Given Surrey’s relationship with Henry Fitzroy, particularly as characterized in his other poetry, the moment of Hector’s visitation to Aeneas is one of the first in the text to take on an element of biographical or political refraction. Indeed, in ‘So cruel prison’, Surrey specifically evokes the parallel, with the lines ‘a kinges sonne my childishe yeres did passe / In greater feast than Pr[ia]ms sonnes of Troy’.16 Further, Surrey must have seen some affinity between his own experience, with his dear friend’s death in the summer of 1536 followed immediately by demanding military service against the Pilgrimage of Grace in the autumn and winter, and the mixture of grief and martial fury expressed by Aeneas throughout his recitation to Dido’s court. Thus, when Hector first appears, there is an intermingling of the personal loss of a friend, of the country’s loss in the death of the king’s most promising son, and of the gory despoilment of the battlefield. The passage, like Virgil’s text, simultaneously glorifies Hector’s past achievements—his victory against Achilles, his naval success—and mourns the degradation of his body through the wounds inflicted in similar martial conflict. The balance is further emphasized in Surrey’s language: Aeneas views Hector with ‘rufull chere’, while Hector’s hair is both ‘crisped lockes’ (that is, curled in a very 14 Ridley, ‘Introduction’, 42. 15 Ridley, 42. 16 Surrey, Songes and Sonettes, fol. 6v–7v, ‘So cruel prison’, lines 3–4.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
119
intentionally structured manner) and ‘clustred with his blood’.17 While in the period, cheer would still primarily mean disposition or expression of that disposition, the connotation of ‘good’ cheer was already current;18 thus, Surrey’s word choice still emphasizes the balance of contrasts that characterizes his treatment of this section. His language emphasizes Hector’s nobility—and so the importance of trusting his judgment—and Hector’s death, both as an end to his potential line and as a result of the tragic situation. The figure, further, is not only there to warn Aeneas, but also to affirm Aeneas’s value as the last scion of Troy. Of course, this emphasis is not only present in, but also essential to, the themes of Virgil’s epic, but the appeal to Surrey’s own well-documented ego seems very strong. As such, Hector’s speech emphasizes the most politically radical strain of Surrey’s writing, generally; again and again, he comes back to allusions that stop just shy of suggesting his own appropriateness as an heir to the throne.19 Once more, reference to ‘So cruel prison’ provides further evidence, as Surrey claims that between Richmond, the King’s only acknowledged living son until 1537, and himself, ‘ech of us did pleade the others right’.20 Here, Hector’s message to Aeneas is one wherein the dead prince transfers responsibility for the survival of his noble line to his friend: O Troyan light, O only hope of thine: […] Sighing hy sayd: flee, flee, O Goddesse son, And saue thee from the furie of this flame. […] Troye commendeth to thy charge Her holy reliques and her priuy Gods. Them ioyne to thee, as felowes of thy fate.21
The themes of these lines are all essential to the Aeneid: the tearing down of something old, the building of something new, and the transfer of the old 17 Surrey, Aeneid, II. 344; Surrey, II.355. 18 OED, s.v. ‘cheer’. 19 The allusions outside of his verse, though, may have contributed to Surrey’s ultimate fate; as Irish summarizes, ‘Surrey was accused of thinking exactly what the Arundel portrait seems to suggest he may have thought: that the Howards, through both birthright and service, were natural guardians of a royal minor, a role they had played in the past’ (Emotion in the Tudor Court, 86). Irish links this belief to ‘Surrey’s ambivalent, envious affinity with the royal family, a dynamic first apparent in his complex friendship with the lost Richmond’ (85). 20 Surrey, ‘So cruel prison’, line 12. 21 Surrey, Aeneid, II.359–377.
120
Gender and Position-Taking in Henrician Verse
traditions to the new line. Such themes, though, were as politically important in Henrician England as they had been in the early Roman Empire. Like Virgil, Surrey wrote his version of this story in a world under siege from internal changes. Like Virgil, he faced the problem of negotiating varying claims of what was ‘traditional’ as even major cultural shifts laid claim to a base in older systems. Surrey also dramatizes the transfer of power along nontraditional lines: Hector was Priam’s son and heir, but he does not appear to one of Priam’s immediate house for his appeal. Rather, Hector looks to a different noble branch as the line of renewal. Priam’s reign is done, and the town of Troy is fallen, but Hector commends Troy to Aeneas’s charge; the larger sense of the civilization will survive while the ruling family and established capital are allowed to pass into history. And, unlike Virgil’s, Surrey’s lines do not have to only valorize the newly established line which has risen to power; his work can as easily be read as a suggestion that the current ruling family should be displaced by another, more vital line. Surrey’s translation further suggests some of the terms by which he wanted to define and value an appropriate ruling lineage. Surrey’s disdain for London and Londoners, famously expressed in both his 1543 rampage through the city’s streets and the ensuing ‘London hast thou accused me’, suggest something of the poet’s investment in abandoning a corrupt urban space.22 Though the common people’s poor choices are a common theme across most versions of the Aeneid, Surrey does seem to take a particularly vicious stance against the commoners of Troy throughout his translation, and his choice to focus much of his verse on the destruction of the city may serve as a kind of fantasized elimination of those voices, preserving only the noble, martial men who sail with Aeneas. That this narrative in Book Two also largely eliminates women, as Aeneas’s wife is one of the victims of the mobs of Greeks, may also appeal to and simplify certain concerns for Surrey, whose choices when dealing with women in the text often emphasize the tensions already present in the original text and in the culture’s convoluted, often contradictory, depictions of idealized womanhood. Inscribing that kind of contradiction, a kind of defense of women interrupts Aeneas’s revenge fantasies against the archetypal ‘bad’ woman: Helen. The blame Aeneas heaps on Helen is interrupted by his mother, Venus. First, the hero of the work spies Helen hiding in the ruins of Troy: 22 Childs offers an overview of the event and a reading of the associated poem (Last Victim, 189–197). As Susan Brigden summarizes, ‘The night of misrule he justified as a warning to the citizens of London, sunk in their seven deadly sins, in idolatry and spiritual blindness, of their impending doom’ (Oxford DNB).
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
121
Then boyld my brest with flame, and burning wrath, To reuenge my town vnto such ruine brought. With worthy peines on her to work my will.23 […] for though on wemen the reuenge Unsemely is, such conquest hath no fame: To geue an end vnto such mischief yet My iust reuenge shal merit worthy praise.24
Although Aeneas acknowledges the dishonorable bent of his desire for revenge, he argues himself into an act of revenge as praiseworthy in the extreme situation. The focus of the passage is largely on Aeneas’s internal justification, rather than on Helen’s desert or actions. Further, in this initial passage, the language used seems to support Aeneas’s argument: he is bold, the work would be worthy, and revenge would be just and praiseworthy. However, in a striking juxtaposition of the virtuous mother and traitorous wife, Aeneas’s divine matriarch interrupts his justification: With furious minde while I did argue thus, My blessed25 mother then appeard to me, […] Disclosing her in forme a Goddesse like, […] thus did she say. Son, what furie hath thus prouoked thee To such vntamed wrath? what ragest thow? Or where is now become the care of vs?26
Interrupting his claims of honorable vengeance against a woman, Venus steps in to remind her son where his concerns should lay, and her argument suggests that the honor and preservation of his immediate family should take precedence over revenge against those he sees as responsible for his country’s 23 For the Loeb edition, Fairclough translates the lines thus: ‘there comes an angry desire to avenge my falling country and exact the wages of her sin’ (333). Surrey, then, emphasizes Aeneas’s independent will, removing the more passive construction. In this choice, he draws attention to the hero’s agency, as ‘my will’ becomes a choice in a way that the angry desire which seems to descend externally is not. I offer occasional comparisons to the Loeb edition throughout this section, chosen for the series’ emphasis on literal translation. 24 Surrey, Aeneid, II.755–77. 25 Fairclough chooses ‘gracious’, here (333); Surrey’s choice then heightens the divine goodness of the interceding goddess. 26 Surrey, Aeneid, II.775–785.
122
Gender and Position-Taking in Henrician Verse
destruction. The characterization that confers her authority, though—her form as a ‘Goddesse’ glistening in light—is linked to her beauty,27 even as Helen is figured as the destructive beauty who has brought about the ruin of Troy.28 While Surrey’s replication of Virgil’s equivocal approach generally participates in the larger cultural narrative of dangerous women, Venus’s intercession suggests that a noble man always has larger concerns than revenge against even the most perfidious woman, offering a further criticism of Surrey’s King without necessarily participating in any real mitigation of feminine threat. The double-sided approach to depictions of women carries over to the fourth book, where Surrey’s choices emphasize the figure of Dido as both a vengeful and capricious woman and as an essentially honorable woman whose choices are misguided by ill counsel. While Dido initially claims refuge in her sacred oath, she seems fairly easily persuaded by her sister’s speech, which Surrey captures in an odd cadence that is both persuasive and quietly suggestive of the tragedy that will follow this choice: O sister, dearer beloued than the lyght: Thy youth alone in plaint still wilt thou spill? Ne children sweete, ne Venus giftes wilt know? Cinders (thinkest thou) minde this? or graued ghostes? […] Comes not to mind vpon whoes land thou dwelst On this side, loe the Getule town behold, A people bold vnuanquished in warre, Eke the vndaunted Numides compasse thee Also the Sirtes, vnfrendly harbroughe: On thother hand, a desert realme for thrust The Barceans, whose fury stretcheth wide.29
Although Anna seems unaware of the problems of her proposal, even her counsel suggests the conflicts that will arise, as the proposed peace between rival kingdoms through marriage is undermined by a selfish drive toward an unsuitable mate. The balance of the language of her speech is much more 27 Surrey, II.778–779. 28 While the Loeb translates similarly beatifying language for Venus’s appearance, the parallel between the two women is undermined by the repeated reference to Helen as a ‘hateful thing’ and ‘unholy thing’, removing her femininity, identity, and beauty. 29 Surrey, Aeneid, IV.39–55.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
123
about the threats surrounding Dido than the suitability of Dido’s choice or any real justification for departing from her bond. Further, Dido will, ultimately, spill her youth, not knowing a child, and attempt to find refuge in those evocative ‘cinders’. Surrey’s rhythm and structure here shows Anna’s persistent de-centering of the subjects of her questions, moving beyond the normal, inverted subject/verb interrogatory structure. While the ‘thou’ of line 40 works as the implied subject of the two questions of line 41, by line 42 Dido is no longer the subject of the questions. More strikingly, her presence in even that question has become parenthetical. In Anna’s efforts to elide the true question at hand, she also begins to erase the proper subject of those questions, just as Dido’s eventual surrender to her sister’s logic will begin to erase her political and personal selves. When Dido yields to her sister’s valorization of selfishness, she steps beyond the bounds of honorable conduct. Perhaps reflecting his own values, Surrey emphasizes only Dido’s misconduct, rather than Aeneas’s, in most of his translation. However, his gendered depiction of the dangers of a monarch guided by lust may be taken more as an effort to feminize Henry, as he does in ‘Th Aseirian King’, than as an indication that the ruler who is a man should be held any less culpable. While he partially borrows this inclination from Virgil, Surrey’s choice to translate only two of the books also does necessarily emphasize this affair. Kim F. Hall identifies Dido as one ‘cautionary tale of the threat of female sexuality to colonial expansion’ and, more broadly, to ‘dynastic politics and royal authority’.30 A prince who fails to escape or contain that threat associates himself with this threatening, feminine foreignness. This gendering is reflected in Surrey’s description following the two rulers’ night together: Ay me, this was the first day of their mirth And of their harmes the first occasion eke. Respect of fame no longer her witholdes: Nor museth now to frame her loue by stelth. Wedlock she cals it: vnder the pretence Of which fayre name she cloketh now her faut.31
Although the first two lines use the plural pronoun to indicate that what follows will affect both Aeneas and Dido (and, by extension, the two rulers’ kingdoms), the dishonor of the misrepresented marriage belongs entirely to 30 Things of Darkness, 153. 31 Surrey, IV.217–222.
124
Gender and Position-Taking in Henrician Verse
Dido. The Latin does end similarly, with ‘hoc praetexit nomine culpam’.32 However, earlier in the same passage, in the Latin, the phrase ‘fulsere ignes et conscius Aether / conubiis’ is used to reference the joining of the two leaders, which the Loeb translates as ‘fires flashed in Heaven, the witness to their bridal’.33 If Surrey takes from the Latin the idea that the sin is Dido’s, he nonetheless undermines the sense that she may not be mistaken by mitigating that marital language. While Surrey does translate ‘wedlock’ for ‘conubiis’ in his corresponding line, he de-personifies Heaven, eliminating the witness necessary to ratify a marital contract.34 In the final line of this passage, the spelling used by the printer for the word choice usually accepted as ‘fault’ is a key example of the kind of slippery multiple meanings that were more easily incorporated in the early modern period. While more characteristic of manuscript writing, generally, than of translation, specifically, the word’s double function as fate/fault is important to the line and to the larger understanding of Dido’s character in the work. The slipperiness of meaning offers a doubled interpretation: Dido’s fate is caused by her fault, or her fault is fated. Lust becomes a particularly dangerous fault that can alter a queen’s fate, inviting a commentary on the situation of Surrey’s contemporary court, where the official rhetoric held that one queen had already fallen victim to lust and the associated fate,35 and where the fate of the country seemed increasingly hostage to the King’s shifting caprices of lust. The declamation against the lustful ruler is momentarily interrupted by the long passage on the damaging force of rumor, which concludes that ‘This monster blithe with many a tale gan sow / This rumor then into the common eares’.36 The role of rumor is again tinged with Surrey’s resentment of the common people. In the Loeb translation of IV.173–197, Rumor is more clearly personified, a ‘foul goddess spreads here and there upon the lips of men’.37 The earlier removal of Heaven’s personification more clearly condemns Dido’s error in judgment; the removal here of rumor’s personification instead pushes greater culpability on the common people, no longer subject to a ‘foul goddess’. On the one hand, Dido and Aeneas are at fault: each has set aside a larger purpose to instead pursue lust, and 32 Virgil, Aeneid, IV. 172. 33 Virgil, IV. 167–168; Fairclough, trans., Aeneid, 433. 34 Surrey, Aeneid, IV. 221. 35 Anne Boleyn’s trail and execution predate even the earliest proposed compositional dates; a slightly later date might also encompass Katherine Howard’s 1542 execution. 36 Surrey, Aeneid, IV.243–244. 37 Fairclough, trans., Aeneid, 435.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
125
each has broken vows to do so. However, in just these three lines, rumor becomes a clearer villain than either character: specifically a ‘monster’ that adds to the truth ‘things don as that was neuer wrought’.38 While the poet/ translator offers indications throughout that Dido and Aeneas’s actions are not fully acceptable or honorable, the lines nonetheless contain the sense that they are unfairly judged by a commons which only condemns their acts on the basis of exaggerated claims and which is not fit to condemn such heroes in any case. The condemnation of Dido and Aeneas, though, is brought forcefully to the front of the translation just four lines later: ‘And that the while the winter long they passe / In foule delight, forgetting charge of reigne / Led against honour with vnhonest lust’.39 The echo here of Surrey’s ‘Th Aseirian King’ is strong, indicating again that part of the appeal of these particular books lay in their potential to shore up many of the beliefs which Surrey already held about those behaviors essential to noble conduct. The charge, though, is importantly not only to do with submission to lust; the focus is not on the religious values that Surrey would likely have also endorsed, but rather on the dishonorable conduct of those who choose personal comfort over the charge of their reigns. One element of this charge is emphasized above all in the text: the importance of a child’s inheritance. Destiny is emphasized as important in terms of what a man can pass on, and for that destiny a queen must be abandoned. When Mercury chastises Aeneas, he focuses above all on the reproach of robbing Iolus of his birthright. His language touches on the importance of personal honor and fame, but the larger concern is on Aeneas’s ‘seed’ and ‘heir’.40 Aeneas further emphasizes this concern in his response to Dido’s pleas for explanation or reprieve. Having acknowledged Dido’s nobility—and not, strikingly, having implied any condemnation of her actions to this point, whatever the energies of the larger text—Aeneas then seeks to explain his own actions, coming ultimately to the point pressed by ‘The wronged hed by me of my deare sonne, / Whom I defraud of the Hisperian crown, / And landes alotted him by desteny’.41 As the earlier condemnation of Aeneas and Dido similarly emphasizes, the relevant concern is one of duty. If a noble queen must be put aside to ensure that the right inheritance goes forward, then Aeneas’s abandonment is noble; indeed, Surrey’s specific choices in 38 Surrey, Aeneid, IV.243–245. 39 Surrey, IV.249–251. 40 Surrey, IV.354. 41 Surrey, IV.462–464.
126
Gender and Position-Taking in Henrician Verse
the surrounding passages offer relatively little resistance to a reading of Aeneas as purely right in his actions towards Dido at the end of Book Four. However, any suggestion that perhaps Henry himself is right to set aside noble queens in service of the preservation of the line is undercut by Surrey’s characterization of Aeneas himself. By focusing his translation on Books Two and Four, Surrey is able to offer a clear endorsement of a particular vision of Aeneas: martial, determined, and right in his ultimate decision to focus on the restoration of the Trojan line. Written under a king who Surrey saw as insufficiently determined and unfit for battle, and who was famously plagued by the problem of securing a line of inheritance for his kingdom, the work suggests the possibility of an entirely new heir as a solution. 42 Surrey’s work emphasizes the necessary intersection of political positiontaking in verse and choices in verse formation. His Aeneid insistently constructs dichotomies that privilege his own positions: masculine over feminine, aristocratic over common, and traditional over progressive—except linguistically. His larger linguistic project, though, elides the history of the modes of writing he seeks to promote; as discussed in earlier chapters, Surrey is concerned with privileging individual authorial voices, just as he privileges Aeneas’s individual worthiness. Surrey constructs this masculinized translation using the strategies of communal manuscript creation, strategies that manifest, for example, in his slippery spelling choices and in his decisions in excising or expanding a given section. The themes of the work, though, reflect the distortions that Surry’s larger poetic project and, perhaps more pressingly, foundational work with Skelton, Wyatt, and Surrey in English studies would impose on this history of position-taking in communal verse production, masculinizing and individualizing poetic authority.
B. Wyatt’s Psalms In earlier chapters, I have established that Wyatt is more comfortable with communal meaning-making than Surrey, and his choices in translation reflect his willingness to embed his work in extensive, shared traditions. When Thomas Wyatt translated his version of the seven Penitential Psalms, he offered an entry in an even more extensive tradition than that associated with the courtly love lyrics and Petrarchan verse for which he is perhaps more well-known. The Penitential Psalms had become a foundational text 42 Walker offers a thorough overview of Surrey’s ambitions and resentments in the chapter ‘Invention of Resistance’ (Writing Under Tyranny, 377–413).
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
127
for essentially every prominent version of European Christianity by the middle of the medieval period, and their influence only grew. The origins of considering these disparate psalms as one piece are unclear, as outlined in Claire Costley King’oo’s research on the Psalms as a united text in late medieval and early modern England. 43 Further, their purpose shifted across time and sect. What remained consistent, though, was that the ‘seven psalms often circulated together’. 44 This tradition, manifested in ‘material embodiments of the unified nature of the sequence’ like illustrated manuscripts and popular prayer books,45 ‘had the effect of fostering certain metonymic or synecdochic modes of interpretation: any particular psalm in the series might be explicated by association with any other psalm, or by reference to the greater whole’. 46 Thus, by the time Wyatt created his particular translation, the potential of the psalms to speak to one another or to outline a particular process of penitence was already well-established. The psalms were an integral part of religious life across Europe; as King’oo emphasizes ‘however (and by whomever) penance was undertaken in the West, the Penitential Psalms were almost always woven into its procedures’.47 Hannibal Hamlin adds that ‘because of the central place of the Psalms in English daily life, and their vital functions within the body of English culture, they were […] in a powerful if peculiar sense, English works’. 48 One of the defining characteristics of early modern understandings of the psalms was the understanding that ‘all of them had been composed by King David’, 49 a figure whose appeal to Henry has already been discussed in the second chapter. The monarchically endorsed use of Davidic imagery at court can be united with the humanist traditions of the courtier as a critical adviser and of the emerging English tradition of poetry as an expression of that advice and criticism; to wit, as Hamlin expresses the connection, ‘the fact that David, divinely appointed king and prophet, had written poetry provided Renaissance poets with a crucial precedent and justification for their own poetic vocation’.50 We can add, to that sense of authority and critique, a natural interest on the part of courtly poets in the variously understood 43 King’oo, Miserere. 44 King’oo, Miserere, 13. 45 King’oo, 13. 46 King’oo, 13. 47 King’oo, 14. 48 Hamlin, Psalm Culture, 6, italics in original. 49 Hamlin, Psalm Culture, 3. 50 Hamlin, 14. Anne Lake Prescott has offered an argument for more conservative interpretations of the Psalms (‘Sheephook to the Scepter’), but I focus here on Wyatt.
128
Gender and Position-Taking in Henrician Verse
roles of Nathan as an adviser to David. Hamlin quite clearly connects poets’ psalm projects to the early modern period’s ‘cultural movement founded on the enterprise of translation’.51 King’oo acknowledges the various re-workings of the psalms through translation as ‘sometimes subversive, sometimes reactionary’.52 Because of the ‘generic indeterminacy’ of the psalms, Hamlin argues, they were a particularly rich outlet for a variety of translations.53 As Hamlin summarizes, then, ‘what often seems most interesting about the Psalms in this period is the way in which they are transformed—adapted and assimilated’.54 Wyatt’s translations of the Psalms are most distinctive because of his insertion of a frame narrative for the works. Though Alexandra Halasz and Richard G. Twombly have both pointed out that Wyatt borrows this feature from Aretino, the narrative is nonetheless unusual, and Wyatt’s version is unique.55 In the opening prologue, Wyatt’s narrator seems particularly to valorize the voice of the resistant advisor and to condemn the king who falls prey to the sin of lust. Specifically setting the action following the seduction of Bathsheba, the narrator chronicles that when he sawe, that kindeled was the flame The noysome poyson, in hys harte he launced So that the soule dyd tremble wyth the same […] So that he forgotte, the wysdom and forecaste Wh[ic]he woo to realmes, when that the [Kinge] dothe lacke Forg[ett]inge eke, goddes Maiestye as [fast] Yea, and hys owne56
Like Surrey’s Dido and Aeneas, Wyatt’s David is guilty of sin which here rests partially, if not primarily, in his abandonment of royal duty. The confused pronouns of the first few lines mix up the personified lust and the sinning king, suggesting that David’s very identity has been consumed by his sin. His particular kingly traits—wisdom and foresight—have been forgotten, and his personal failure is a failure of his kingdom. While the Penitential Psalms were often portrayed as following the events of David’s affair, Wyatt’s choice 51 Hamlin, Psalm Culture, 1. 52 King’oo, Miserere, 3. 53 Hamlin, Psalm Culture, 3. 54 Hamlin, 11. 55 Halasz, ‘Wyatt’s David’, 321–322 and Twombly, ‘Wyatt’s Paraphrase’, 351. 56 Wyatt, Certayne Psalmes, 3r–6r, ‘Prologe’, lines 9–20.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
129
to add this more explicit outline of events is not likely to have gone unnoticed, particularly given Henry VIII’s own penchant for portrayal as David. However, Wyatt’s own life offers another interpretation, both for current readers and for his contemporary audience. One of the issues over which Wyatt frequently found himself at odds with the King and his advisors was Wyatt’s own mistress.57 The possibility for the narrative to refer to Wyatt’s own penitential journey adds several significant dimensions to the work. First, the self-reference offers an additional screen for Wyatt; not only is the work a translation, it is also a translation which he can be understood to undertake out of penitential self-interest. Second, though, that self-referential interpretation aligns Wyatt with the King—both David, the penitent poet, and Henry himself, the infamous adulterer who often figured himself with that same Davidic imagery.58 As Surrey can imagine himself as Aeneas, defender of noble values and lynchpin in the shifting dynastic future of his people, so Wyatt can figure himself as both David and advisor. Third, and in an important distinction, Surrey’s figuration focuses on exceptionalism—on how his identification with the character of Aeneas sets him apart. Wyatt’s figuration instead focuses on commonality—on how his own struggles might compare with others and be mapped onto this text of communal penitence. Fourth and finally, Wyatt’s perspective is given extra authority by his own image as the sinning but redeemed king. This odd balance of penitence and redemption—that is, of error but also of authority—was one of the reasons for Henry’s own Davidic iconography. Because the great prevalence of that iconography meant that reference to David was always easily interpreted as reference to Henry, Wyatt’s own ability to claim the poems as self-reference empowered the critical elements of his text, rather than truly effacing them. When Wyatt dramatizes the potential penitence of an adulterer, he maps a traceable critique of those who error and will not seek communal resolution through an acknowledgement of their sin. The lack of sight and self-knowledge that so dangerously characterize the David of Wyatt’s opening lines are further emphasized in David’s 57 While no full biography has, to my knowledge, been written for Elizabeth Darrell, she figures significantly in Wyatt’s history. In 1541, for example, Henry officially ordered Wyatt to return to his wife and end his relationship with his mistress (King’oo, ‘Rightful Penitence’, 156). He had at least one son with Darrell (DNB), and their relationship is one obvious factor in his long estrangement from his wife. See also Muir’s Life and Letters. 58 Henry’s public image had, by the 1530s, taken on some of the exaggerated dimensions it still retains, informed by the highly public nature of his several marriages (see Walker, Writing Under Tyranny, and Weir, The King and His Court). Southall offers 1534 to 1542 as the widest possible range of dates for Wyatt’s composition (‘Wyatt’s Psalms’, 496).
130
Gender and Position-Taking in Henrician Verse
confrontation with Nathan, who enters the text as a righteous advisor evocative of the best humanist and biblical traditions. Nathan emerges as the essential advisor to the King, whose own lost self-knowledge, emphasized by an insistent repetition of ‘blynde’,59 has convinced him that others are similarly blind to his sin. The language evokes a sense of convincing rhetoric, as well; as the King’s advisor, Nathan ‘settes afore hys face’ the outline of ‘thys case’, and ‘shewethe’ the consequences of the King’s rash actions.60 Further, the text preserves no sense that Nathan has been ‘sent’ by God, as in most biblical translations. Instead, Nathan has spied out the King’s folly through his own observation, and he comes to correct the King out of his own sense of moral obligation. ‘Instead of an agent of God who foresees the future, we have a man who investigates David’s secrets and before whom David throws himself to the ground when the treachery is brought into the open’.61 As a depiction of the appropriate relationship between a king and his advisors, Wyatt’s prologue suggests the necessity of corrective criticism. While Wyatt would have seen this suggestion as a necessary element of true service, his larger work clearly indicates that he had, at this point, sufficient experience in Henry’s court to no longer believe in the efficacy of such criticism, even when it could be safely offered.62 In his presentation of even the deeply flawed David as receptive to the kind of criticism Henry will no longer hear, Wyatt uses the translation as an outlet for that repressed expression. David’s response emphasizes his trust in his advisor and the appropriateness of the action Nathan has taken. Once Nathan has laid out the ‘case’ of his sin, David Hys heate, hys luste, his pleasure all in feare Consume and waste, and streyght hys crowne of gold Hys purple pauler, hys scepter he letteth fall And to the ground, throweth him self wyth all63
Eschewing his kingly trappings, David accepts his advisor’s judgment and withdraws himself to a cave to offer penitence through his psalms. Robert 59 Wyatt, ‘Prologe’, lines 31–32. 60 Wyatt, lines 34 and 38. 61 Halasz, ‘Wyatt’s David’, 328. 62 As Walker concludes, ‘by the early 1540s the natural supposition of a Wyatt […] was that the sort of directly political literature [he] was creating was not—and indeed should not be—for the King. Theirs was writing about the king, despite him, and in many ways against him’ (Writing Under Tyranny, 416, italics in original). 63 Wyatt, ‘Prologe’, lines 45–48.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
131
G. Twombly has examined the problem of David’s first attempts at penance; essentially, David has eschewed the performance of kingliness for the performance of the penitent.64 Twombly’s focus is on the insufficiency of performed penitence which does not truly accept the inability of the penitent to affect his own fate. Given this interpretation of correct penitence, a further corollary emerges to the idea of performed majesty and of correct performance of authority. As he throws aside the performance of king, David’s education in penitence may also be an education in rule. The double function of David’s penitential journey is emphasized by the complicated relationship Wyatt creates between his poetic self, his narrator, and his poem’s object. Halasz sees this relationship as speaking to Wyatt’s complicity within the corrupt court to which he responds: ‘as a poet, he addresses the king with a consciousness of loss. He can call the king to account, but he cannot do so from a position of innocence or spiritual purity’.65 Thus, when Wyatt’s David first begins his psalms of penitence, there is an element of unworthiness or failure in his first appeals, elements not inherent to the psalm. Twombly, specifically, sees the problem as one of spiritual fitness, arguing that the ‘first of Wyatt’s seven paraphrases opens with a David emotionally unprepared for prayer, uncomfortable with his own motives, and engaged […] in a kind of running argument with both God and himself’.66 The problems of David’s initial approach are apparent in his language, which does not so much balance claims of goodness, aspirations to mercy, and evocations of God’s justice as bounce furiously between extremes, presenting, in this first attempt, no logical or emotional ‘truth’: That the repentaunce, whych I haue and shall May at thy hande, seke mercy as the thynge Of onely comfort to wretched sinners all Whereby I dare with humble bemonynge Of the goodness of thee, this thynge requyre Chastyce me not, for my deseruinge Acoordynge to thy iuste conceaued yre O lorde I dreade, and that I did not dreade I me repente67 64 Twombly, ‘Wyatt’s Paraphrase’, 335–380. 65 Halasz, ‘Wyatt’s David’, 325. 66 Twombly, ‘Wyatt’s Paraphrase’, 352. 67 Wyatt, Certayne Psalmes, 6r–10v, ‘Domine ne infurore’, lines 4–12.
132
Gender and Position-Taking in Henrician Verse
Wyatt’s David imagines himself forgiven, in terms of mercy, but then frustrates the terms of that mercy by trying to judge his own ‘deseruinge’,68 first presenting himself as humble and supplicant, then shifting to a view of his sins as ones which can only deserve punishment. Confronted with the idea of a just God, David retreats into fear, frustrating the Reformist energies towards a doctrine of mercy that will be more successful later in the psalms. The relatability of the narrator’s compromised position and the invitation to consider the ongoing doctrinal complications emphasize the place of the text within Wyatt’s community and society, offering courtiers, specifically, and English citizens, more broadly, points of access for reading their concerns into the text. The shifting doctrinal perspectives evoke the shaky grounds of Wyatt’s England and a church which, subject to the variations of one man’s mind, underwent a series of confusing and sometimes contradictory shifts.69 However, while the latent critique suggests the dangers of this model, the narrative suggests that hope exists for the King and that capitalizing on that hope is largely predicated simply on recognizing it. The end of the first psalm presents one of the most troubling consequences of David’s limiting self-will, as Wyatt’s translation emphasizes the near-threat of David’s words to God at the end of the psalm. Apparently fearing that mercy will not be granted—a reflection of his failed understanding of the nature of divine grace—David attempts to strike a bargain, of sorts, with his God: Then yf I dye, and goo where as I feare To thynke ther on, howe shall thy great mercye Sounde in my mouthe, vnto thee worldes care For ther is none, that can the laude and loue For that thou wilt no loue, among them there70
David tries to cast ‘love’ here as reciprocal, as if the King’s worship of God is like God’s love to the King. In this case, if God fails to show his love by failing to grant his mercy, the King may withdraw his love by withdrawing his praise. David fails to understand a doctrine of justification, when he 68 Wyatt, line 13. 69 Tadmor’s Social Universe considers, partly, the consolidation of doctrinal positions through translation and regularization, offering several examples of such shifts; one relevant example would be the ‘forbidden degrees’ of marriage (76–77). 70 Wyatt, ‘Domine ne infurore’, lines 62–66.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
133
sees both kinds of love as potentially equivalent; he fails to understand a doctrine of absolute mercy when he thinks it might fail.71 Throughout the f irst few psalms and their prologues, Wyatt’s David struggles between multiple interpretations of grace, mercy, and justification. In the second prologue, David ‘specifies the absolute inwardness of redemptive merit, yet craves the formulation of it in outward terms’.72 This seems also true, though, of the narrator’s voice. Partly, the function can be understood as an acknowledgement of the limitations of metaphor, which must serve to describe the state of the soul and of religious experience. However, the second prologue seems invested in highlighting, rather than mitigating, that insufficiency: Whoseo hathe sene, the sycke in hys feuour After truce taken, wythe the heate or colde And that the fytte is paste, of hyse feuour Drawe fayntinge syghes, let hym I saye beholde Sorowefull Dauid, after hys languor […] Yt semed nowe, that of hys faulte the horrour Dyd make a ferde nomore hys hope of grace Thee threates whereof in horrible terrour Dyd holde his harte, as in despaire a space Tyll he had wyll to seke for hys succoure Hym selfe accusynge, beknowynge hys case Thynkynge so beste, hys lorde to appeace And yet not healed, he fealethe hys dysease73
A fever is both an insufficient synonym for the danger in which David understands himself to be and a potentially perfect metaphor for an early modern understanding of God’s grace as either already present or already absent in a person’s life; either the fever will pass or it will not, and the early modern patient has little to do with the operation. However, David remains unwilling to accept this position. The complications of Wyatt’s line ‘dyd make a ferde nomore hys hope of grace’74 highlight the mental acrobatics 71 Halasz, who sees the sequence as ultimately condemnatory, concludes that ‘David condemns himself […] by suggesting his necessary separation from the spiritual content of his prayer’ (‘Wyatt’s David’, 334). 72 Twombly, ‘Wyatt’s Paraphrase’, 356. 73 Wyatt, Certayne Psalmes, 11r–12r, ‘The Auctor’ I, lines 1–16. 74 Wyatt, ‘The Auctor’ I, line 10.
134
Gender and Position-Taking in Henrician Verse
of David’s arguments. That this is Wyatt’s composition, rather than part of the translated psalms, makes the grammatical difficulties of this section even more obviously essential to Wyatt’s meaning. It should not seem to David that the horror of his fault any longer threatens—or in any way impacts—his certainty of grace. Instead, though, the threat of his fault holds his heart in terror and despair, which drives him to renew his supplication. Even though it seems that David is restored through mercy to grace, he is unwilling to accept this state, and focuses instead on his attempts to appease his lord through, essentially, a careful and well-structured argument. David’s refusal to accept his own lack of control in the situation reflects a serious misunderstanding of the religion he is meant to head, and Wyatt’s readers would understand both the danger of such a misunderstanding and the particular problem of a church head who could not submit himself to mercy, rather than rely on justification.75 Throughout the narrative, the problem of David’s position as a king who is also the spiritual leader of his people is a peripheral but significant concern. In fact, the text’s sidelong approach to the issue emphasizes exactly how prescient the issue was to Wyatt’s contemporaries, and how aware Wyatt was of the dangers—the ‘profound and potentially unsettling public implications’76—of his work. Both the psalms and Wyatt’s inserted prologues obliquely address David’s status, and so reference his role as a religious and political leader. In one of David’s attempts attempt to self-expatiate, he explains that ‘I for byscause, I hydde it stylle wythin / Thinkinge by state, in fault to be preferred / Do fynde by hyding of my fault my harme’.77 David acknowledges that he has created greater danger for himself because he thought himself above others and perhaps above judgment, or at least subject to different rules of judgment. As Halasz observes of the entire sequence, ‘David’s sin is personal or private, but it unfolds itself through, and has consequences for, his princely or public power’.78 Indeed, as the lines above imply, David himself understands his sin as partially informed by his princely, public power; he suggests that he partially falls as a consequence of his preferment to that power. 75 Halasz, contrasting Dante’s David with Wyatt’s, summarizes that, ‘When Wyatt separates the princely David from the authority of his song, he joins Dante in the insistence that temporal concerns be subject to spiritual authority, but he cannot redeem the figure of David, who functions in his poem as an object lesson in the unrepentant transgression of the limits of princely power’ (‘Wyatt’s David’, 340). 76 Walker, Writing Under Tyranny, 355. 77 Wyatt, Certayne Psalmes, 13r–15r, ‘Beati quorum remisse sunt’, lines 21–23. 78 Halasz, ‘Wyatt’s David’, 330.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
135
The problems of David’s position, then, imply a potential danger for those who must follow him. Wyatt’s prologues also emphasize the helplessness of those followers, as when he notes that so close the caue was, and unkoweth That none but god, was recorde of hys payne Els hadde the wynde blowen, in all Israell eares Of theyr kynge, the wofull playnte and teares79
The lines make Israel a kind of absent witness, reinstating the public in the text while also, paradoxically, emphasizing their removal. The people have no place as witness and must simply hope that their King can reconcile himself to the doctrine he is meant to convey to them, offering an imitation, of sorts, of David’s own struggle with God, where he must learn to trust his own lack of agency. This relationship, of David as God’s servant just as the people must be his, continues from the preceding interlude, where the relationship is first figured in apparently appropriate language that then slides into dangerous ambiguity. As the seruaunte, in hys maysters face Fyndynge pardon, of hys passed offence Consyderynge his greate goodnes, and hys grace Gladde teares dystylles, as gladsome recompence Ryghte so Dauid, semed in thee place A marble Image of synguler reuerence Carued in the rocke, wythe eyes and hande on hyghe Made is by craft, to playn, to sobbe, to syghe80
The final line reopens the fissures of David’s penitent performance; if he ‘made is by craft, to playn, to sobbe, to syghe’, who is the maker?81 The doctrinally correct answer is ‘God’, but then the problem emerges of God’s complicity in David’s suffering. Of course, religious thought in the period was more than suff iciently complex to hold God’s absolute power and absolute mercy as apparently contradictory but inherently co-active ideas. However, Wyatt’s David does not, at this point, present an argument for 79 Wyatt, Certayne Psalmes, 20r–21v, ‘The Auctor’ III, lines 21–24. 80 Wyatt, Certayne Psalmes, 15v–17r, ‘The Auctor’ II, lines 9–16. 81 Wyatt, ‘The Auctor’ II, line 16.
136
Gender and Position-Taking in Henrician Verse
any comfortable acceptance of this theology; rather, David himself seems only to alternate between two simplified positions. David’s own conflicted approach to both positions also implicates his role as his own ‘maker’: both in the sense that he is responsible for his own sin and in the sense that he creates a performance of penitence for that sin. Finally, the poet himself becomes the maker, especially of these prologues. The narrator/poet makes David plain, sigh, and sob. Wyatt forms this king, who may or may not yet understand appropriate penitence; the implication may be that he wishes for a similar influence on his own king—that he could make him consider and confront this doctrinal position. That Wyatt may be the greater force here is emphasized when he undercuts the potential for David to be his own maker. As David falls once more into his habits of justification, he laments ‘For I my selfe, loo, thinge moste vnstable / Formed in offence, conceaued in lyke case / Am nought but synne from my natyuytie’.82 David’s protest that this defense ‘be not […] sayde for myne excuse’ only emphasizes that this does, in fact, sound like an attempt to excuse his behavior.83 Further, the excuse implicates God as the causer of man’s frailty in a less-than-comfortable extension of doctrine. The moment emphasizes the ways that ‘Wyatt destabilizes both the narrator and David in order to expose the temporal, spiritual, and poetic issues that the figuration of David allows’.84 Even nearing the end of the sequence, Wyatt’s David is a deeply conflicted figure who works through, but does not ultimately succeed in working out, the conflict of a man’s drive for justification and his need for greater mercy. Views of the conclusion of the poem differ sharply. Twombly sees the end of the poem as fundamentally redemptive; he argues that ‘Wyatt’s sequence, roughly, is a movement from helpless uncertainty in the heart and mind, through a frantic and paradox-ridden effort to dispel uncertainty, to a state where David can speak with earnestness, forthrightness, and without the irony that comes from hidden motive’.85 Halasz, though, counters that the conclusion offers no such clear vision of David’s final state. Instead, she sees David’s voice as implicated in a larger system of self-interest that is meant to mirror the problems of Wyatt’s England. Even after the process of the psalms, Halasz argues that ‘the pointing toward political allegory reveals that David’s example can be enacted in bad faith, his voice assumed for 82 Wyatt, Certayne Psalms, 21v–24v, ‘Miserere mei deus’, lines 30–32. 83 Wyatt, line 33. 84 Halasz, ‘Wyatt’s David’, 336. 85 Twombly, ‘Wyatt’s Paraphrase’, 359.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
137
interested purposes. What is true for the princely David is also true for the singer-self of the psalms’.86 Halasz argues, ultimately, that this psalmic sequence shows David as an unsuccessful penitent. In the final discourse, in her argument, Wyatt ‘crosses the boundary between his text and the psalmic text and shows David’s ostensibly sacred words to be shaped by another moment of David’s private discourse, not his discourse with God, but with himself and his own interests’.87 At this moment, Halasz argues that David’s self-interested appeals have betrayed his repentance as ‘mere appearance’.88 However, that David appeals to God in the interest of perhaps preserving his son can also be read as a less purely self-interested moment. First, this moment emphasizes the parallel to Henry, obsessed with the question of a son’s inheritance. Wyatt’s depiction reflects that such an obsession, while temporal and worldly, is not necessarily sinful. The counterweight of this scene against Surrey’s depiction of Aeneas as heir to Troy is especially suggestive of the poets’ distinct projects; Wyatt’s work, here, certainly focuses on the sympathetic desire of the king for his own heir, rather than suggesting any kind of substitution. I submit that Wyatt’s paraphrase stops short of showing David as subject again to his sin, and instead leaves an open ending which suggests the dangers that such temporal interests, however understandable, may pose to the immortal soul of the penitent. In this same moment, Twombly sees instead ‘a reawakening of a sense of kinship with the race and a resulting notion that David’s own problem is not unique and that the work of spiritual regeneration has therefore some broad historical principle behind it’.89 Twombly’s argument, when paired with Halasz’s, offers a more comprehensive view of Wyatt’s project. As Twombly articulates, there is a definite and traceable concern with the communal in Wyatt’s Psalms that strongly contrasts the energies of exceptionalism and even of condemnation of the commons in Surrey’s Aeneid. Wyatt aims to show the kinship of the king with the larger race—and in this aim, his work is fundamentally political, for he claims a spiritual equality with and so a right to criticize his social superior. A 1537 bilingual devotional guide on the psalms by Robert Redman ‘preserves two dissimilar approaches to the Penitential Psalms’,90 which King’oo 86 87 88 89 90
Halasz, ‘Wyatt’s David’, 342. Halasz, 323. Halasz, 335. Twombly, ‘Wyatt’s Paraphrase’, 377. King’oo, 107.
138
Gender and Position-Taking in Henrician Verse
describes, using Roland Greene’s theoretical terms, as the fictional—relating the Psalms as composed all in sequence, by David—and the ritual—relating the psalms by their function.91 King’oo then illustrates the ways in which Wyatt’s Psalms are f ictional rather than ritual: ‘Whereas in the ritualized paraphrases […] the Penitential Psalms require identification and self-effacement on the part of those who read or hear them, in Wyatt’s fiction they beg for interpretation and analysis instead’.92 Because they are presented as a narrative, the work of a community of readers is essential to the form Wyatt gives his psalms. Wyatt writes an element of that readerly criticism into his work through his addition of the narrative voice. The 1549 edition of his Psalms, published under a more actively Protestant king, emphasizes this element, ‘since it not only sets the narrator’s voice apart from David’s but also privileges that voice, giving it extra command’ in the headings titled ‘the Auctor’.93 This increased assumption of command comes with risks, but because Wyatt is no longer available as a culpable body, the voice can instead claim an authority for poetry. Once the poet himself is gone, ‘the prologues derive extra clout from being associated with the incomparable wisdom and courage of Wyatt himself’,94 who, as Surrey’s poetic labor has guaranteed, will serve as a figure of poetic genius and the privileges that it ought to be afforded. That project also, though, begins the long history that will divorce Wyatt’s work from the communal production that shaped his composition and that manifests in the insistent presence of the reader within his Psalms.
C. Parr and Hybridized Position-Taking Katherine Parr’s devotionals reflect those methods through which women who wrote might endorse, subvert, alter, or use the restrictive ideals of courtly performance; her biography reflects those methods through which extant patriarchal powers endorsed, assimilated, or punished different outlets of women’s education and expression. While Micheline White has clearly established the role of collaboration in Parr’s processes of textual creation, we can connect that collaborative strategy more insistently, here, to the processes used in courtly manuscript production and, again, to women’s 91 92 93 94
King’oo, 107. King’oo, 119, italics in original. King’oo, 121. King’oo, 121.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
139
essential participation in the creation of those processes. Because of her highly visible role and the publication of her texts, Parr’s work amplifies the performative, self-crafting elements of those courtly manuscripts. In some ways, Parr’s publications represent an alternative history to the one we have read through our field’s previous focuses on Skelton, Wyatt, and Surrey—and through the individualizing, authority-oriented focus of the last, particularly. Through her published devotionals, Parr crafts a particular, self-preserving image of self, but that self is both exceptional and communal, both self-authorizing and dependent on appeals to particular audiences. I want to first pull forward the considerations of Davidic imagery already engaged in the second chapter and in the preceding section. As White reads it, in Parr’s first publication, the Psalms or Prayers, ‘Parr’s translation of Fisher served […] to represent Henry to […] his people as a particular kind of Davidic monarch—repentant, desiring of wisdom and obedience, in need of divine military assistance, and thankful for God’s help’.95 A lynchpin of White’s argument is the connection between Parr’s choices and annotations Henry himself made in his Psalter—annotations which themselves, as White argues, are ‘courtly displays of his wartime identity’.96 Placed in conversation with Surrey’s apparently anti-monarchical focus on military prowess and Wyatt’s quieter critique through Davidic imagery, Parr’s translation emerges as a re-appropriation of these themes and images. This re-appropriation enlarges themes which Henry had successfully deployed for himself, earlier in his reign; in this later and public deployment, though, Henry works with Parr to authorize her representations and writing work, rather than presenting work directly authorized with his own name. As White establishes, ‘Henry and Parr must have engaged in serious political discussions as the materials were assembled, translated, and edited’.97 Such political discussions may manifest in the ways changes to Fisher’s work tend to legitimize Henry’s authority. Parr ‘alters the Latin description of Christ in ways that stress that Henry is ruled only by Christ, a subtle reminder that Henry is no longer subject to any sort of external ecclesiastical authority’;98 she also ‘dramatically recasts an entire petition so that Henry’s obedience to God’s will is framed in active rather than passive terms’.99 These changes, though, must also reflect Parr’s and Henry’s familiarity with the position-taking power of 95 96 97 98 99
White, ‘Royal Iconography’, 555. White, 555. White, ‘Literary Collaboration’, 30. White, ‘Royal Iconography’, 568. White, 568–569.
140
Gender and Position-Taking in Henrician Verse
translation, and that consideration of strategy is my primary concern in this section. Fisher’s psalms, as prayers created by connecting disparate bits of the bible, themselves already reflect the arguments in the final two chapters about transcription as composition, a technique which Parr herself practices extensively. Mueller has described Parr as borrowing from Fisher and Cranmer ‘the selecting, adapting, and free recombining of source materials that yields a transformative synthesis’.100 If Fisher and Cranmer are two sources for this practice, though, Parr would also have accessed these strategies through an awareness of the position-taking practices of courtly manuscript production—an awareness she would have gained both through her marraige and through her friendship with Margaret Douglas, a prominent member of her household.101 Parr’s primary literary practice is in combination and replacement and her devotionals are, for the most part, composed of bits of other works, put together to emphasize the themes and beliefs she most wants to foreground and altered, though not always consistently, to underscore the authorial and authorizing voice of the queen. To claim that a large part of Parr’s own image-making is invested in clarifying her position as a queen may seem self-evident, but the very instability of that role in the 1530s and 1540s suggests that a recursive process of definition is in play. Parr needs to define herself as a queen, and, for her own safety, needs to define what a queen is. As part of that definition, she turns very specifically to a core element: a queen is a king’s wife—and that position contains some special separation or access to divinity, even if only by proximity, which authorizes the Queen’s writing and actions. Thus, even as Parr’s position-taking work partakes of the strategies of courtly manuscript production, that work shifts under the pressure of individual, political power concerns and the necessarily distinct considerations of public, print publication. As one example, Parr may transfer manuscript practices of placement into her print publications, to a distinct effect: Psalms or Prayers concludes with two final prayers which are then re-printed in Parr’s Prayers or Medytacions, at least suggesting an invitation to the audience to read connections between the two texts. That connection both links Parr’s name on the Prayers or Medytacions to the ostensibly unattributed Psalms or Prayers and pulls the explicit royal authorization of Psalms or Prayers forward onto Parr’s second text.102 100 Mueller, Katherine Parr, 3. 101 Mueller, 13. 102 Mueller agrees that in ‘reprinting the two concluding prayers’ Parr ‘affirms her authorship’ of the later text (Katherine Parr, 18).
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
141
The significance of Parr’s position-taking, as a queen writing, is reinforced by the language of fidelity and marriage that she often uses when speaking to or about God. This choice recurs throughout her work, but takes on particular weight in Prayers or Medytacions because the text is the first printed with her name displayed. Early modern women were, of course, often encouraged to think of God and Christ in such marital terms. However, for a queen specifically to reference her lord is necessarily, simultaneously, for her to reference her husband and king, as well.103 The most striking examples of this language are those which link the marital figuration to carnal language: ‘Quicken my soule, and all the powers thereof, that it maie cleaue fast and be ioined to the in ioyful gladness of gostly rauishynges’.104 In expressing desire for union with God (and so peripherally with her husband the King), the queenly speaker expresses no shame or hesitance. The condemnation of similar desires elsewhere in the text makes it clear that such emotions are only allowable within certain accepted, circumscribed arenas. The ‘bodily necessitees’ and ‘voluptuous pleasures’105 of this world are to be condemned and persistently avoided. Eagerness at the idea of union is only permissible in the speaker’s anticipation of God ‘visit[ing her] in suche wise, as thou doest visite thy moste faithfull lovers’,106 or, as an extension would imply, within the faithful bonds of marriage—a rather speaking delineation, given the fate of her predecessor and the fact that Parr had rather recently been pursued by one of Henry’s courtiers. The image of the soul as the feminine partner in a union with Christ is not, of course, limited to women, but the use of the feminine form nonetheless gains particular weight and significance when used by a woman writing. As Janel Mueller points out in her comparison of Parr’s work and Stephen Gardiner’s texts, this use of a gendered voice can operate to ‘confound […] gender differences because this determinant operates meaningfully only in human social relations, not in the relations between humans and the divine’.107 Parr’s use more insistently emphasizes the equality of all human souls, for she intentionally and often reminds the reader that her use of feminine personal pronouns refers not only to the author’s soul, but to the author’s physical body, as well. Her use of the gendered pronoun encompasses 103 Mueller sees this same near-conflation; as she puts it, Parr ‘experiences God’s grace and the king’s graciousness to her as closely concurrent if not convergent sources of blessings’ (Katherine Parr, 3). 104 Parr, Prayers or Medytacions, 13v. 105 Parr, Prayers or Medytacions, 10v. 106 Parr, Prayers or Medytacions, 7v. 107 Mueller, ‘Complications of Intertextuality’, 152.
142
Gender and Position-Taking in Henrician Verse
both the physical and the spiritual, making it difficult to separate references between the two within Parr’s texts. Within the context of Parr’s writing, the ‘she’ of the soul is also the ‘she’ speaking and writing, and all souls are also referenced by this one personal pronoun. In presenting a queen as a model, this self-referentiality both encourages the common people to consider themselves capable of following her noble example and empowers the feminine speaker by making ‘her’, and so the other ‘hers’ of England, potentially equal, at least in spiritual terms, to their male counterparts. These reminders of the femininity of the author work in another particularly significant way in Parr’s texts: in her privileging of simplicity and direct connection to God over learning and tradition, even as she demonstrates her erudition and education. Broadly considered characteristic of Parr’s work, this tendency is particularly succinctly expressed by Edith Snook: Parr ties this portrait of herself as one of the unlearned to her advocacy for the position that the Bible is the sole source of doctrinal authority. She dismisses alternative traditions as ‘man’s doctrine’ and because of her claims to simple, feminine piety, she need not even contend with that tradition […] Although Parr is claiming that the unlettered need grace, she is also locating theological authority elsewhere than in learned men. Parr’s representation of herself as one of the simple is a rhetorical deployment of gender.108
Parr intentionally inhabits the feminine as a defensive position against the learned men who might disagree with her. She masterfully evokes the feminine when she can use the apparent weaknesses to her advantage, manipulating the ideas of the feminine as uninstructed or simple to match them to the emerging Protestant ideals of the soul. She effectively appropriates the old, negative vision of femininity for a new, positive Christianity. Speaking in this public voice, Parr moved into a highly specific position among the women of her time. In so doing, she not only articulated a feminine perspective on king, country, and religion, but also made visible, through her negotiation, the silencing authority of which she was always aware and with which she had to engage to publish her work. Mueller comments specifically on a striking marginal note made by a woman reader, one Anne Dyson, near one of Parr’s passages in Prayers: ‘ye faithfull dooe servisee to god with thyr soul’. As Mueller articulates, ‘it is poignant to find this appreciative jotting from a sixteenth-century reader, responding to emphasis laid by a 108 Snook, Women, Reading, and the Cultural Politics, 48.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
143
woman writer, her contemporary, on the expressive capacities in the souls’ silences—a condition widely identified today as a historically feminine one’.109 This note speaks quite clearly to the ways in which Parr’s work as an author legitimized the feminine experience for her contemporary women readers, even as that same work reached beyond lines of gender to embody the emerging ideals of a newly Protestant nation. Though her position may have as often limited as licensed her authorial work, Parr the educated, religious courtier moved beyond and within the ideal of the feminine voice to create works that reflected the larger projects in which she was involved. In so doing, Parr the author and queen also created speaking commentaries on the gender roles through, and with which, she had to work to conceptualize herself as woman, queen, and author. Parr’s texts act as position-taking citations and support for her public positions and as self-authorizing and self-preserving performances of her own—and her husband’s—authorities. White’s and Mueller’s extensive work with Psalms or Prayers establishes a clear foundation and framework for considering Parr’s composing practices and their relationship to her political and personal positions; we can use that framework to approach the fascinating hybridity of Prayers or Medytacions, joining Parr’s language to her form as coordinated pieces of her position-taking work in the text. Mueller carefully describes what we can and cannot know about the relationship between the print versions of Prayers or Medytacions and the manuscript copy of the text which Parr gifted to a woman courtier.110 In that manuscript copy, the text is organized in, as Mueller identifies them, versicles—a term that powerfully suggests the connections between the short, often Psalmic phrases specifically identified and the poetic influences on Parr’s composing strategies. Prayers or Medytacions exists as a manuscript text and as a print text; the manuscript form of the text emphasizes the possibility of reading the work as something between poetry and prose. If Parr’s emphasis on authorship and on her queenly voice align her text with the position-taking strategies of Surrey and of her royal husband, her practices also draw extensively on the larger, more communally-oriented manuscript work of other courtly writers. In negotiating this middle space, moreover, Parr may provide one influential model for the two young women writers with whose works I conclude this chapter—indicating ways of activating useful pieces of resistant positiontaking verse work within appeals to established authority and self-protective impulses. 109 Mueller, ‘Devotion as Difference’, 118. 110 Mueller, Katherine Parr, 369–421.
144
Gender and Position-Taking in Henrician Verse
D. Two Translations into Prose by Early Modern Englishwomen This final section addresses a category of work which moves more clearly afield of courtly verse production than Parr’s complex compositions, but which, like Parr’s texts, can still be shown to work in and respond to those systems. At least two quite young women at the Tudor court, near the end of Henry’s reign and the beginning of his son’s, chose to translate important works from verse for presentation to their parents. In both cases, these women chose to translate the verse into prose. I suggest that while these authors—Elizabeth Tudor and Lady Jane Lumley—wanted to access some of the cloaked criticism available to them through translation, they also sought to distance themselves from any implication of real—that is to say, stability-threatening—rebellion by avoiding the acknowledged strain of resistance that had adhered to verse translation in the Henrician court. In moving from the translated works of Wyatt and Surrey to those by Parr, Elizabeth, and Lumley, I am conscious of Patricia Phillippy’s critique: When women’s writing has been inserted into comprehensive literary histories, its inclusion has been both partial and strategic: a small number of ‘anomalous’ works disrupt the monolithic narrative of the masculine canon. More often, women’s writing has been treated as if enacted remotely, distinct from masculine practice and canonicity and incommensurate with conventional standards of merit or value.111
My work here, particularly in its engagement with women’s manuscript practices, indicates that one issue at hand is whether manuscript poetry production at the Henrician court should be considered a masculine practice, at all. Further, the analysis here of Wyatt and Surrey indicates their own measures of incommensurability, and, while Surrey’s deviations from larger manuscript practice function distinctly, I consider his work, in many ways, more anomalous than Parr’s, Elizabeth’s, or Lumley’s. Finally, I am concerned with these last two works precisely because they are not enacted remotely—because the young women authors enact a sophisticated and informed revision of the larger courtly manuscript practices. Here, then, we both ‘seek connections among women’s work or between women’s writing and relevant texts in the masculine canon’112 and trouble the masculinizing of that canon. 111 Phillippy, ‘Sparking Multiplicity’, 2. 112 Phillippy, 13.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
145
The connections between these texts and to other texts by Henrician women engage other key questions raised by Phillippy, who particularly interrogates the relationship of women’s writing to subversion or orthodoxy and to tradition or rejection of that tradition.113 These are especially complex intersections for Elizabeth and Lumley. Lumley and Elizabeth respond to the accumulated meaning of courtly manuscript verse—and so respond to a myriad of uses of position-taking work. The young women writers do not, that is, necessarily respond to the exact works of those manuscripts, but primarily to the larger environment, likely influenced by Parr’s translation work and modification of verse-composing strategies. Elaine V. Beilen long ago established a practice for ‘analyzing women writers as a group, and arguing that they form an early modern tradition of women’s writing’.114 More recently, and in one of the most concise and thorough-going overviews of the current state of scholarship on women’s writing, Margaret J.M. Ezell emphasizes the ways in which ‘[l]ooking away from authorship as an isolated individual act to view the cultural context in which texts are created and consumed helps to reveal women active as literary agents’.115 Within these contexts of both courtly production and, more specifically, women’s writing, we can best understand those ways in which Elizabeth and Lumley chose to implement or reject those strategies which would be likely to influence reception of their works. A number of parallels exist between the two works, in terms of both production and thematic content. Elizabeth’s work, a translation of Marguerite de Navarre’s Le Miroir de l’âme Pécheresse, was presented to her stepmother as a New Year’s present; Elizabeth dates the prefatory letter to Parr as the last day of 1544, when Elizabeth would have been eleven years old. Jane Lumley’s work was likely written in 1550, and Harold H. Child shows that the volume was paired with her husband’s translation of Erasmus’s Education of a Christian Prince in a present to Jane’s father.116 Born in 1537, Jane would have been about thirteen at the time of her work. Perhaps because both women were so young, earlier critics dismissed both works as fundamentally schoolroom exercises, even implying that Lumley might have instead worked from Erasmus’s Latin version, printed earlier in the century. However, Lumley clearly titles her work as ‘translated out of Greake into Englisshe’,117 and, as a 113 Phillippy, 18–21. 114 Beilen, Redeeming Eve, xvi. 115 Ezell, ‘Invisibility Politics’, 35. 116 Child, ‘Introduction’, Iphigeneia at Aulus, trans. Lumley, vii. 117 Lumley, fol. 63r. My numbering for the translation is taken from Child’s edition.
146
Gender and Position-Taking in Henrician Verse
gift demonstrating her scholarly acumen to her father, Lumley’s text is likely, as she presents it, a translation from the Greek. As she had only recently married, her father was well aware of his daughter’s abilities, particularly given his clear investment in her education.118 Perhaps because Elizabeth’s education is more widely documented, the authenticity of her work has not been called into question, but there is frequent discussion of some other ‘assigning’ the work to Elizabeth, framing her engagement with the text as purely mechanical. Elizabeth’s Miroir Departing from that reading of mechanical or technical engagement, I want instead to consider the implications of this text as one that Elizabeth or her mentors actively chose as an appropriate text for the young royal to engage.119 Perhaps those who worked with her or were aware of her work saw the entry as the kind of courtly rebuke that was expected of ideal courtiers—and a great prince was first to be an ideal courtier. Perhaps, rather, the work was meant to flatter Henry’s vanity—to imply that this young princess did not see her father’s hypocrisy. Or perhaps the motives can be combined. The text could be used to subtly remind Henry of the ideals to which he should aspire, shaming him into a better example to both preserve his final queen and to live up to his younger daughter’s expectations. Elizabeth’s actual translation work foregrounds appeals to motherly or brotherly love, evocations of filial duty to a father, and justifications for women’s power and voice, suggesting that, in this early work, she was more interested in self-preservation and promotion than in guiding her father’s larger political agenda. In the short prefatory letter, Elizabeth draws a parallel between the narrator, who turns to God for salvation of her imperfections, and her own appeals to her stepmother, following a discussion of the larger argument 118 Oxford DNB. 119 Anne Lake Prescott begins her analysis of Elizabeth’s translation from the premise that ‘whoever asked or encouraged her to translate the Miroir forgot or did not know that it is more than a piece of piety by a famous and friendly patron of moderate reform’ (‘Pearl’, 68). She argues that whomever this other agent was, they must not have been familiar with the work, since it would ‘have occurred to any adult who read beyond the first two hundred lines that this poem, with its impassioned evocation of God as a great king and judge who is kind to daughters and does not execute adulterous wives, was a most unsuitable means of displaying Elizabeth’s talents at Henry’s court’ (71). I am not discarding this possibility, but the argument here outlines an alternative reading, based in my understanding of Elizabeth’s translation work as occuring within a larger courtly context of position-taking.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
147
of the text’s narrative of salvation with a passage appealing to Parr’s mercy for the text: althoughe i knowe that as for my parte, wich i haue wrought in it: the (as well spirituall, as manuall) there is nothinge done as it shulde be, nor else worthy to come in youre grace handes, but rather all vnperfytte and vncorecte: yet, do i truste also that oubeit it is like a worke wich is but newe begonne, and shapen: that the fyle of youre excellent witte, and godly lerninge, in the reding of it […] shall rubbe out, polishe; and mende (or els cause to mende) the wordes (or rather the order of my writing), the wich i knowe in many places to be rude, and nothinge done as it shuld be120
Mirroring the soul’s appeals to God in the text and as outlined in her preface, Elizabeth’s appeal specifically mentions the shortcomings of her spiritual as well as her manual work; she appeals to her stepmother to forgive and guide both. The parallel is interesting because by addressing Parr, specifically, Elizabeth elides the more traditional connections between God and king that figure with some prominence in the translation which follows and which, as discussed, frequently recur in Parr’s own work. Instead, she replaces the masculine figure with the maternal, as indeed she replaces Marguerite’s ‘pere’ with mother in a later line in the text.121 This maternal replacement also places Elizabeth’s work more firmly within a specifically feminized chain of literary exchange. Elizabeth is not just writing as a princess to a queen, but also as one woman writer to another. Parr’s own work offers similar practices in position-taking and self-preservative cloaking to both the manuscript tradition and Elizabeth’s translation work. When Elizabeth chooses to work from the potential of verse translation, evidently preferred for more strident position-taking, into prose translation, she both affirms her stepmother’s authority and emphasizes her own relationship to that power. Moreover, Elizabeth chooses work placed in a network of royal women’s power. Unlike Wyatt’s biblical translation or Surrey’s Roman epic, the weight of Elizabeth’s Miroir relies not on the power of the established canon, but on the relative positions of author, translator, and recipient. The work is assigned cultural importance because of the royal women involved in its creation. 120 Elizabeth I, Le Miroir, 42. 121 For further discussions of this replaced word, see Prescott, ‘Pearl’; Snyder, ‘Guilty Sisters’; and Mueller and Scodel, ‘Introduction’, Elizabeth I: Translations.
148
Gender and Position-Taking in Henrician Verse
Elizabeth’s language in the letter to Parr does mirror some of Marguerite de Navarre’s traditionally self-effacing language in her own original preface. However, as Susan Snyder has pointed out, this self-effacement slides into assertions which suggest women’s equality with men, in at least spiritual, if not sometimes more general, terms.122 Elizabeth’s translation preserves this effect, beginning with an assertion of the limitations of work by a woman that quickly moves into an assertion of the worthlessness of all men and worth of all believers: ‘beholde rather the matter, and excuse the speche, consydering it is the worke of a woman: wiche hath in her neyther science, or knowledge, but a desire that eche one might se, what the gifte of god doth when it pleaseth hym to iustifie, the harte of a man. For what thinge is a man, (as for hys owne strenght) before that he hath receyued the gifte of fayth’.123 In this preamble, as Snyder sees in the original, the specific ‘“Femme”[…] slides into [a generic] “homme”, and Marguerite’s gendered humility transmutes into a veiled assertion of equality with men’.124 Elizabeth preserves this sense, which gains perhaps further weight from her appeal to Parr as a figure of intellectual and spiritual power in her own preface. John N. King considers the specific influence of Parr in the creation, at the late Henrician court, of ‘an ambitious program that fused Bible reading with private theological study’ and which ‘specifically addressed the requirements of a female readership’.125 This influence suggests Parr’s larger association (potentially exaggerated for her stepdaughter) with a more equalizing evaluation of women’s role in religious life and literature. The beginnings of both the original and translated texts suggest a space for women’s expression as justified through the same forces as men’s and so draw parallels that confer authority on women’s writing. This theme is reinforced by Snyder’s suggestion that, for both Marguerite and Elizabeth, a thematic concern is the role of a sister who is intellectually equal, or even superior, to a younger brother and celebrated heir. Speaking of Marguerite, Snyder writes that ‘for all her adoration [she] could not help being aware from their early years that her abilities were as good or better than her brother’s. Intellectually and even diplomatically, she outshone him’.126 The characterization is satisfyingly appropriate to Elizabeth, as well. Both relationships are further complicated by the women’s necessary dependence 122 Snyder, ‘Guilty Sisters’, 443–458. 123 Elizabeth I, Le Miroir, 44. 124 Snyder, ‘Guilty Sisters’, 450. 125 King, ‘Patronage and Piety’, 43. 126 Snyder, ‘Guilty Sisters’, 449.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
149
on these figures; ‘the one whose potent presence makes her inferior and worthless is also the one who confers status on her, the one through whom she reclaims worth’.127 For Marguerite, many of these tensions may have been lessened by her close relationship which her brother, who ensured that status and worth were consistently conferred on his sister and who had left the country under the regency of his mother at least twice.128 Elizabeth may have known enough of this relationship to be inspired by this ‘natural’ model of closeness in France, as exemplified by Louise, Francois, and Marguerite.129 Elizabeth may attempt to highlight the sisterly models of the Miroir as a consideration of a version on that model including her beloved stepmother and young brother, while excluding her Catholic older sister and her father, whose decline, by 1544, would have been apparent. By translating this Reformist text, with its structural reliance on familial relationships, Elizabeth could appeal to her stepmother at present and set up a precedent for established sisterly loyalty and support when her brother came into his own. Perhaps as part of this system, Elizabeth uses some of her most evocative language in her translation of a passage which figures the traditional image of the sinner in a forest or wilderness of sin. Because of Elizabeth’s careful elision of a clear subject, though, the figuration becomes equally powerful as an image of a believer or a virtuous person set amongst the vices of the court: For i am to farre entred emongest them and that worse is, i haue not power, to obtayne the true knowledge of one […] If i thinke to loke for better, a braunche cometh and doth close myne eyes: and in my mouth doth fall when i wolde speake the frutte wich is so bytter to swalowe down. If my spirite be styrred for to karken: than a great multitude of leaffes doth entre in myne eares and my nose is all stoped with flowres.130
Although the lines are ostensibly about the experience of the soul trapped in guilt, the imagery is suggestive of the experiences of a woman trapped in a court turned against her—whether against her in terms of preservation of her religious virtues or in terms of political honest and clarity. The speaker 127 Snyder, 452. 128 Snyder, 447–448. 129 Snyder outlines these key parallels and patterns, but argues that Elizabeth ‘probably was not aware of these parallels between translator and author’, instead suggesting that Elizabeth would see these similarities primarily in the language of the work itself (454). Given the Tudor valuation of Marguerite, though, as outlined by Prescott (‘Pearl’, and acknowledged by Snyder, as well), Elizabeth might have indeed known about some of these biographical similarities. 130 Elizabeth I, Le Miroir, 46.
150
Gender and Position-Taking in Henrician Verse
does not have the power ‘to obtayne […] true knowledge of one’;131 that is, she cannot penetrate the nature of her sin or of the group surrounding her. Her senses are confused by excess and bombardment. In this case, Elizabeth’s choice to move from verse into prose also emphasizes the narrative sense of the entrapment and makes the figurative language function more dramatically and less poetically. The text goes on to join this imagery to imagery of the carceral. The movement from jungle to jail evokes the oppressive combination of confusion and rigidity in role performance that often characterized poetic depictions of court conditions. Within this garden of sin or vice, the speaker imagines ‘my poore soule, a slaue, and prisonnere doth lye, withoute clartie, or light hauinge both her fete bound by her concupiscence, and also both her armes through yuell vse’.132 Her language suggests the problems of being bound to a place full of dangers—both in the larger spiritual sense of the world of flesh and in her more immediate sociopolitical context. Elizabeth’s language also, then, figures the servant relationship to God in ways which seem designed to plead her case before a more immediate master, in apparent appeals for protection from some of these dangers. In these appeals, Elizabeth preserves a greater tendency towards Marguerite’s original copia than in the translation generally.133 She compiles multiple offenses in her metaphor of service to God, lamenting her own inadequacy with ‘Alas, what a mayster withoute to haue deserued any goodnes of hym but rather serued hym sloughtfully and withoute ceasse offended hym, euery daye, yet is he not slake in helping me’.134 The appeal to these masculine figures is again emphasized in her later line, ‘O what goodnes and swittenes Is there any father to the daugther, or els brother to the syster, wiche wolde euer do as he hath done’.135 While her language ostensibly underlines the inadequacy of such relationships, she also appeals to a desire to mitigate that inadequacy, implicitly asking that brother and father mirror divine love in their treatment of her. Given, as Prescott points out, Elizabeth’s tendency to eliminate Marguerite’s repetitions in most of the text, her choice to maintain them in ‘Father, father: alas, what ought i to thinke: shall my spirite be so bolde to take vpon hym to call the, father, ye, and also, our father’ becomes striking for its difference from the rest of her rather pared-down 131 Elizabeth I, 46. 132 Elizabeth I, 46–48. 133 More characteristically, as Prescott points out, Elizabeth cuts down on repetitions, as though ‘[e]ven Marguerite’s taste for repeated words […] struck Elizabeth as wasteful or loud’ (‘Pearl’, 68). 134 Elizabeth I, Le Miroir, 50. 135 Elizabeth I, 54.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
151
text.136 She follows the lines by specifically referencing the Pater Noster as a precedent for this address. These lines are taken from Marguerite with little alteration, but there is nonetheless a poignancy in Elizabeth’s appeal to a text to establish her ability to name her father that, if it strikes the reader five hundred years later, was likely not lost on the unusually politic eleven-year-old writing to her fourth stepmother. The same problem of legitimacy and naming is further emphasized in one of the few areas where Elizabeth alters an entire phrase of her translation, in the lines ‘I am not worthy (i tell it afore euery man) to call myselfe be called thy childe, but (o bountiefull father) do no worse, vnto me, but as to one of thy householde seruauntes’.137 The deletion of a choice of phrase, rather than just of a misplaced letter, is unusual in the edition. Elizabeth has specifically edited out her agency, choosing the passive voice construction and eliminating her ability to name herself this Father’s child. Nonetheless, Elizabeth’s translation also frequently emphasizes the relative security of this connection in contrast to worldly relationships. Of God, she claims ‘there is true paternitie in the’138—a striking formulation for then still-bastardized Elizabeth—and finally comes to the claim that God ‘hast broken the kinrede of myne olde father, callinge me, daugther of adoption’.139 Elizabeth suggests her religion as a source of secure identity, but she also suggests that, if her father, Head of the Church, would mirror his Father, he must offer her more ‘true paternitie’.140 While the filial language of some of the translation is more tentatively evocative of Elizabeth’s troubled relationship with her royal father, other moments in the text offer criticism that, were the work not a translation, could only be seen as overtly referring to Henry VIII. Marguerite’s original came as near to criticism of the king’s power, but Henry’s increasing use of execution as punishment adds a greater element of irony to the line ‘If i haue deserued death he (as a kinge) shall geue me grace, and pardon, and delyuer me frome prison, and hanging’.141 That Henry continued to enact increasingly strict laws, from which he often decidedly did not give grace or pardon, also affects the later line, ‘Thou gyuest vs a lawe, and punishement if we do not fulfulle it: and thyselfe wolde not be found to it, forbiding vs the thinge wich thyselfe didyst’.142 This second moment is somewhat abated, placed, as it 136 Elizabeth I, 58; Prescott, ‘Pearl’, 61–76. 137 Elizabeth I, Le Miroir, 66–68. 138 Elizabeth I, 58. 139 Elizabeth I, 62. 140 Elizabeth I, 58. 141 Elizabeth I, 68. 142 Elizabeth I, 74.
152
Gender and Position-Taking in Henrician Verse
is, in the voice of Miriam, who the contemporary reader understands to be ‘wrong’ in her questioning of both her brother and of God. However, the text itself is not as clear in its condemnation. Though Miriam still suffers her traditional punishment and is still restored to her community through the work of her brother, her questions are neither answered nor invalidated. Her doubt is only wrong because of who she doubts, and the charge—that her brother gives laws without following them—is dangerously applicable to Henry, as well. In her translation of Marguerite’s many lines on the insufficiency of human relationships, Elizabeth cannily maintains the implied compliments to those whose relationships are found sufficient, in their mirroring of divine love. Particularly informative is the formulation of her translation of Marguerite’s claims to a shared inheritance as a natural result of a shared relationship: ‘Nowe than that we are brother, and syster togyther, i care but lytell for all other men. Thy laundes are my owne inheritaunce, lett vs than kepe (if it pleaseth the) but one husholde’.143 Snyder again highlights how Marguerite’s original uses this apparently subject relationship to actually empower and entitle the woman speaker, making her subordination to her brother a claim to share his lands.144 While ‘if it pleaseth the’ does offer a shield of deference to the claim, the language is nonetheless bold, and even bolder in the mouth of a disinherited daughter.145 Once more, Elizabeth’s depiction of the strength of divine love, as figured through metaphors of familial relationships, seems also designed to inspire strength in her actual, temporal relationships. However, even as Elizabeth valorizes these ideals, she emphasizes that reality is rarely so kind. In the original text, the claim might have seemed overblown: If any mother hath taken any care for her sonne; If any brother hath hyd the fautte of hys sister. I neuer saw it (or elles it was kepte wonders secrette) that any husbande wolde forgiue his wife, after that she had offended hym and did returne vnto him. There be inoughe of them, wiche for to auenge their wronge, did cause the iudges to condemne hym them to dye146
Surely Marguerite saw instances of kindness, and her own family was a close one that likely did cover one another’s faults and which certainly took care 143 Elizabeth I, 76. 144 Snyder, ‘Guilty Sisters’, 452–453. 145 Elizabeth I, Le Miroir, 76. 146 Elizabeth I, 76.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
153
of one another. Elizabeth’s life, in contrast, might indeed have shown her that this grim litany largely reflected the state of the world.147 Following this indictment of man, the translation spends some halfdozen pages on the adulterous wife, almost tediously both justifying that her husband should put her aside, or have her more severely punished, and celebrating the clemency of a husband who does neither. The passage concludes with the loving God-husband recalling his unfaithful wife with forgiveness, saying ‘But thou, wich hast made separacion of my beade: and did put thy false louers in my place and commyted fornicacion with them: yet, for all thys, thou mayest come vnto me againe’.148 In this case, of course, Henry cannot live up to the model provided by divine love; he has already failed this particular test of clemency, and Elizabeth’s careful work to offer multiple justifications for such punishment leading up to this line suggests that she was aware of the dangerous ground on which she trod. Nonetheless, she emphasizes the model of clemency as perfect. Throughout her work, Elizabeth’s translation also emphasizes the doctrine of grace and mercy that Wyatt’s earlier Psalms had presented as David’s struggle. Whether her text was designed also to encourage Henry’s Reformist energies, or only to appeal to her stepmother’s more decidedly Protestant inclinations, Elizabeth’s early work would later prove useful propaganda for her reputation as a most Protestant princess. Designed to strengthen her position through its appeals to specific family members and through its careful presentation of the potential dangers and benefits of familial relationships, the work would instead ultimately serve to strengthen her position as a queen who no longer had any such strong relationships on which to rely.149 Jane Lumley’s Iphigeneia In analyzing Lumley’s text, this chapter moves beyond the chronological parameters of Henry’s court, gesturing towards the endurance and evolution 147 As Prescott acknowledges, ‘because so much of what Marguerite says concerns family love, any errors are statistically likely to touch on that matter. Nevertheless, Elizabeth’s at times startling deviations are worth noting and […] cannot be without significance’ (‘Pearl’, 68–69). 148 Elizabeth I, Le Miroir, 86. 149 Elizabeth’s work here also offers an opportunity to consider the consequences of the Henrician courtly system for later verse production. Stevenson and Davidson mention, for example, the ‘highly coded and controlled exchange of verses and emblems, compliments and warnings between Queen Elizabeth and her cousin Mary Stuart’ (Women Poets, xxx). This constrained translation work by a young Elizabeth, thoroughly informed by techniques of courtly coding, suggests some influences for that exchange.
154
Gender and Position-Taking in Henrician Verse
of the techniques practiced in the first part of the sixteenth century. Lumley’s achievement in this translation, given her age, is necessarily shaped by the influences under which she has been raised, and those influences are perhaps stronger, or more purely distilled, in such a young author than they are in older, more experienced writers who have access to a larger number of cultural systems. If Lumley’s text is written in 1550, then the date of her work is sufficiently near the end of Henry’s reign to trace similar influences and literary practice. Further, her father’s position as Lord Chamberlain near the end of Henry’s reign suggests that Lumley could, potentially, have been aware of Elizabeth’s work translating foreign verse into English prose, and taken the princess as a model of women’s power, specifically as strengthened by familial relationships; Lumley would almost certainly have been aware of Parr’s publications and reputation as a translator and writer.150 Child notes that all of the texts save one in the manuscript which preserves Iphigeneia are recorded in ‘Lady Lumley’s autograph’;151 the included facsimile from the volume shows a relatively unadorned and wonderfully neat hand. This facsimile also shows the ‘ornamental flourishes [which] complete the half-filled lines at the end of speeches and also frequently occur after catchwords’, as noted by Child.152 Given the larger manuscript practice of the time, these flourishes suggest that Lumley did not want to allow for misinterpretation or for later additions to the work. Lumley’s work to avoid such changes suggests her investment in her project as a preservation of her own voice, rather than as a larger cultural project on which others might collaborate.153 150 Considering a later manifestation of Lumley’s courtly connections, Davidson offers an especially striking example of the overlap of women’s material and manuscript cultures in ‘an illuminated manuscript on vellum which [Anne] Bacon had made and sent to Jane, Lady Lumley’ which preserved the inscriptions the Bacons had included in their Long Gallery (‘Spatial Texts’, 194). He characterizes this gift as ‘one seriously intellectual member of court circles reaching out to another’ (196) in which ‘instead of a house made to be read like a manuscript, we have a manuscript made to be read as a paper house’ (196). The document provides an illustration of Lumley’s place within courtly writing communities—and, potentially, a bridge between these earlier position-taking strategies and later evolutions. 151 Child, ‘Introduction’, v. 152 Child, x. 153 Day offers an alternative reading of collaborative practice in Lumley’s work, contextualizing the familial interplay of education, resources, and advisement as a form of collaboration. As Day puts it, the family manuscripts ‘promenade humanist discourse, cultural practices of gift exchange, and familial relations as select collaborative partners’ (‘Literary Gifts’, 140). I agree completely with this importantly more comprehensive view of collaborative manuscript practice, but intend to delineate between the more immediate opportunities for collaborative work on the page offered by, as discussed in the following chapter, certain courtly manuscripts and
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
155
This concern with self-identity is mirrored throughout the translation in the figure of Iphigeneia. In ‘The Argument of the Tragadie’ which begins this entry in the manuscript, Iphigeneia is named before her father.154 Indeed, hers is the second proper name in the text, preceded only by Calchas, the prophet. Iphigeneia’s identity is quite quickly linked to her place as Agamemnon’s daughter and as a virgin.155 Throughout the text, though, this identity is confirmed as Iphigeneia’s choice, just as the decision to go to the sacrifice will ultimately be her choice. Further, Iphigeneia rejects the additional identity which she could take on as Achilles’s wife, accepting his protection only insofar as he will enable her purpose. This kind of selective self-identification is a necessarily resistant act for an early modern woman, but Lumley contains the potential radicalism of her position-taking by reinforcing mores of chastity and spiritual education as central elements of women’s self-formation. One major preoccupation of the text seems to be the connection between Iphigeneia’s imminent sacrifice and her pretend marriage. When Agamemnon first reveals his plot to Senex, he specifically links his misdeed to the fact that ‘nowe I haue determined the deathe of my daughter, under the color of mariage, and none knoweth of this’.156 Even after Agamemnon has recanted his plan, the servant rebukes him for the horror of his deceit, emphasizing that ‘thou haste determined to sacrafice thy owne childe, under the colour of mariage’.157 As Deborah Uman outlines, Agamemnon’s ‘pretense that [Iphigeneia] is to be married rather than killed creates a none-too-subtle analogy between marriage and sacrifice, an analogy that illustrates the very public role of domestic arrangements and […] reveals the futility of Agamemnon’s impulse to restrict his daughter to the private realm’.158 Daughters, as Uman’s argument outlines, will be sacrificed in some form or another to their fathers’ ambitions, and the bounds of what it means to sacrifice a life are blurred by Lumley’s insistent focus on the metaphorical connection. A series of repetitions after Achilles discovers the plot emphasizes this connection between marriage and sacrificial death, but also complicates any connotation that marriage should be feared. Indeed, the responsibility Lumley’s practice which, as I see it, attempts to create some control around textual alteration and marginalia. 154 Lumley, Iphigeneia, fol. 63v. 155 Lumley, fol. 63v–64r. 156 Lumley, fol. 69r, italics mine. 157 Lumley, fol. 69v. 158 Uman, Women as Translators, 77–78.
156
Gender and Position-Taking in Henrician Verse
even of a pretend husband is emphasized both by Clytemnestra’s pleas and by Achilles’s concerns. In the space of twenty-eight lines, the concern is brought up in remarkably similar formula three times: Clytemnestra reproaches Achilles that ‘the reproche shalbe yours, seinge my daughter beinge sente for under the color of your name’;159 Achilles responds that he will help because ‘if she beinge sent for in my name shulde be slaine, then truly it wolde turne to no small dishonor to me’;160 and he ends the same speech by reinforcing that ‘it shoulde sounde to no litell reproche to me, if that throughe my occation your daughter shulde be slaine’.161 While this kind of repetition may partially account for the dismissal of the play as a schoolroom exercise or unskilled work, repetition was a key strategy of many contributors to early modern English manuscripts.162 The use of nearly identical phrases emphasizes Lumley’s theme: that Achilles’s role even as faux husband makes him partially responsible for Iphigeneia, and that this role has the potential to confer honor or dishonor on his name. Achilles’s real desire for Iphigeneia when he learns of her virtue further mitigates the repetitions as related to any fear of marriage qua marriage. After Iphigeneia first expresses her resignation to the sacrifice, Achilles responds, ‘Trulie I wolde counte my selfe happi if I mighte obteine the O Iphigeneya to be my wife, and I thinke the O grece to be uerie fortunate bicause thou haste norisshed soche a one: for you haue spoken uerie well’.163 Although Iphigeneia still chooses to go forward to the sacrifice, Lumley’s language and Euripides’s characterization do not present this husband figure as threatening or domineering. Further, his speech not only honors Iphigeneia’s virtue, it also explicitly references her worth as increasing the renown of her country. Her honor adds to the honor of her motherland, and both are increased specifically through her rhetorical skill and her virtue. With her repetitions, Lumley successfully emphasizes the complications, already a focus of Euripides’s original text, of the father/daughter relationship in patriarchal cultures where women are given in marriage. Only through the father does the daughter have potential access to learning, as demonstrated by Iphigeneia’s rhetorical skill; security, as demonstrated 159 Lumley, Iphigeneia, fol. 84r–84v. 160 Lumley, fol. 84v. 161 Lumley, fol. 85r. 162 As Clymer articulates, ‘repetition has often been considered indicative of practice quaintly primitive or embarrassingly visceral, something to be transcended as early modern becomes modern. Yet, to think so is to overlook how instances of repetition consistently structured important modes of thought and behaviour’ (Ritual, Routine, and Regime, 4). 163 Lumley, Iphigeneia, fol. 92v.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
157
in her confident and affectionate conversations with both parents; and, by extension of perhaps both of these, the secure future offered by a good marriage, as demonstrated by the cruel charade of betrothal to Achilles, who emphasizes his willingness to take on such an accomplished and virtuous bride. Lumley offers a dense network of associations, then, when she offers this translation, in combined presentation with her new husband’s work, as a gift to her father. However, while Agamemnon may fail to ultimately uphold his obligations as a parent to protect his daughter, Lumley’s text, with its emphasis on Iphigeneia’s choice and its insistent references to marriage, also foregrounds that through his earlier, successful parenting of a learned daughter, he has upheld his obligations as a subject, to protect his country. Iphigeneia herself is insistent that her father not be blamed, repeatedly exhorting her mother not to mourn and insisting that ‘he is compelled to do it for the welthe and honor of grece’.164 Her rejection of Achilles and following directions to her mother foreground the larger concern with renown and with proper duty: Herken O mother I praye you unto my wordes. for I perceiue you are angrie withe your husband, whiche you may not do. for you can not obtaine your purpose by that meanes: And you ought rather to haue thanked Achilles, bicause he so gentelly hathe promised you his helpe, which maye happen to bringe him into a greate mischefe. I wolde counsell you therfore to suffer this troble paciently, for I muste nedes die, and will suffier it willingelye. Consider I praie you mother for what a lawfull cause I shalbe slaine […] I shall not onlie remedie all thes thinges withe my deathe: but also get a glorious renowne to the grecians for euer.165
Iphigeneia’s rebukes to her mother encourage a reading of her willing selfsacrifice beyond submission; clearly, her decision is not based entirely on acquiescence to the presumed wisdom of her elders. These complex interactions of filial and patriarchal traditions intersect with those Alexandra Day sees in her analysis of Lumley’s translations of Isocrates’s orations. In these, Day articulates, Lumley may have ‘performed the actions of a son’, but she also ‘revise[s] th[e] pattern [of father to son transmission of 164 Lumley, fol. 94r. Even without knowledge of the usual ending to the myth, though, Clytemnestra’s response here is ambiguous, if not condemnatory: ‘If he hath done this willinglye then trulye he hathe committed a dede farre unworthie of suche a noble man as he is’ (fol. 94r). 165 Lumley, fol. 91v-92r. The speech does also contain the line ‘one noble man is better than a thousande women’ (fol. 92v), but Iphigeneia’s worth as savior of thousands of men serves to undercut the misogynistic claim.
158
Gender and Position-Taking in Henrician Verse
scholarly wisdom], for Lumley is a daughter writing back to her father with learned authority’.166 Performing authority in this text, Iphigeneia reveals an awareness of her death as political strategy—one that may belong to her father, but in which she is an actor, nonetheless. Other characters voice one further obvious result of Iphigeneia’s decision; as the Chorus declaims, ‘In dede by this meanes you shall get your selfe a perpetuall renowne for euer’.167 Iphigeneia’s education in appropriate duty has also been one in honor, and her actions show a desire for self-valorization alongside her avowed duty to her country—one that the Chorus verbalizes, even if Iphigeneia does not. Iphigeneia’s choice is of fame and valor over marriage, ultimately, as she rejects Achilles’s offers to attempt her protection. Thus, within her translation, Lumley successfully presents a woman’s self-identification with what could be considered ‘masculine’ values, while protecting her own voice and identity with the mask of translation and by avoiding the suspicion of resistance that likely accompanied verse translation. Day characterizes Lumley’s work as ‘a culturally significant example of gender performativity that constitutes a disruption of the masculinist conception of western scholarship not only because Lumley is a woman writer, but because of what she chooses to write, what she writes it on, and who she writes it to’.168 Perhaps taking knowledge of Elizabeth’s and Parr’s works, or the work of royal women broadly, as her model, Lumley negotiates a space for position-taking work that draws on the systems of translation without accessing the riskier elements of those systems. *** Considering both the distinct and the shared strategies of women and men in translation alongside the larger system outlined in this section allows us to sketch in some of the details of the ways position-taking verse worked for Henrician courtiers. In the translations examined in this chapter, the translations by men seem to seek grounds for a kind of equality with, and certainly a right to criticize, those in power, while the translations by women instead seek to describe a relationship to one in power marked by particular obligations on both sides. Simultaneously, though, these women make subtle arguments both for women’s spiritual equality and for women’s linguistic accomplishments. The next chapters will show that 166 Day, ‘Literary Gifts’, 139. 167 Lumley, Iphigeneia, fol. 95r. 168 ‘Literary Gifts’, 140.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
159
women at court frequently did have an interest in more explicitly resistant position-taking projects, and the works from the Devonshire Manuscript offer potential evidence that women were, at minimum, conf ident in manipulating the position-taking potential of translation in transcribed manuscript forms. The prose translations considered here, though, are more limited in any terms of resistance or radicalism. This is likely partly because of the translators’ young ages; these young women may have been more interested in self-identif ication or, in Elizabeth’s case, selfpreservation than in a larger power project, and indeed establishment of independent identity for a young woman was itself a potentially radical act of empowerment. The very fact that these young women chose to translate into prose rather than verse, though, suggests that their educations had made them aware of the resistant potential for verse translations, which they then avoided as counterproductive to their more personal projects because more obviously identifiable as a form of position-taking opposed to dominant powers. That sense of counter-productivity can function in multiple directions. Surrey’s strained relationship with the conventions of position-taking codes, which begins to emerge in his resistance to communal submersion, can be linked to Henry’s own abandonment of verse production and to Parr’s work, which I have framed as hybridizing the demands of the communal manuscript and a more individually authorizing print text. Coded work is work for those who are not (or at least not fully) in charge—those who are more comfortable with subordinate positions are correspondingly stronger users of coded position-taking strategies. The most privileged men in the aristocratic system are, then, selective in their participation in coded courtly verse, and that selection may influence the historical and historiographical arcs by which translation moves from highly resistant positioning in the Henrician court to a framing as feminized work later in the period. The access points of translation offer us an avenue for engaging the larger framework of significance, working with the exceptions as well as the established relationships to re-consider the contingent contexts through which Henrician courtiers read and composed verse.
Works Cited Beilen, Elaine V., ed. Early Tudor Women Writers. Surrey: Ashgate, 2009. Beilen, Elaine V. Redeeming Eve: Women Writers of the Renaissance. Princeton UP: Princeton, NJ, 1987.
160
Gender and Position-Taking in Henrician Verse
Carlson, David R. ‘The Henrician Courtier Writing in Manuscript and Print: Wyatt, Surrey, Bryan, and Others’. In A Companion to Tudor Literature, edited by Kent Cartwright, 151–177. Malden, MA: Wiley-Blackwell, 2010. Childs, Jessie. Henry VIII’s Last Victim: The Life and Times of Henry Howard, Earl of Surrey. New York: St. Martin’s, 2007. Clarke, Danielle. ‘Translation’. In The Cambridge Companion to Early Modern Women’s Writing, edited by Laura Lunger Knoppers, 167–180. Cambridge: Cambridge UP, 2009. Clymer, Lorna. ‘Introduction’. In Ritual, Routine, and Regime: Repetition in Early Modern British and European Cultures, edited by Lorna Clymer. Toronto: U of Toronto P, 2006. Davidson, Peter. ‘Spatial Texts: Women as Devisers of Environments and Iconographies’. In Phillippy, History, 186–202. Day, Alexandra. ‘Literary Gifts: Performance and Collaboration in the Arundel/ Lumley Family Manuscripts’. In Gender, Authorship, and Early Modern Women’s Collaboration, edited by Patricia Pender, 125–148. Palgrave Macmillan, 2017. Elizabeth I, trans. ‘Translation of Marguerite de Navarre’s Le Miroir de l’âme Pécheresse’. In Elizabeth I: Translations, 1544–1589, edited by Janel Mueller and Joshua Scodel, 40–128. Chicago: U of Chicago P, 2009. Ezell, Margaret J. M. ‘Invisibility Politics: Aphra Behn, Esther Inglish and the Fortunes of Women’s Works’. In Phillippy, History, 27–45. Greenblatt, Stephen. Renaissance Self-Fashioning. Chicago: U of Chicago P, 1980. Halasz, Alexandra. ‘Wyatt’s David’. Texas Studies in Literature and Language 30, no. 3 (1988): 320–344. Hall, Kim F. Things of Darkness: Economies of Race and Gender in Early Modern England. Ithaca: Cornell UP, 1995. Hamlin, Hannibal. Psalm Culture and Early Modern English Literature. Cambridge: Cambridge UP, 2004. Herman, Peter C. Royal Poetrie: Monarchic Verse and the Political Imaginary of Early Modern England. Ithaca: Cornell UP, 2010. Howard, Henry, Earl of Surrey, trans. The Aeneid. By Virgil. Edited by Florence H. Ridley. Berkeley: U of California P, 1963. Irish, Bradley J. Emotion in the Tudor Court: Literature, History, and Early Modern Feeling. Evanston: Northwestern UP, 2018. Kesselring, K. J. ‘No Greater Provocation? Adultery and the Mitigation of Murder in English Law’. Law and History Review 34, no. 1 (2016): 199–225. King, John N. ‘Patronage and Piety: The Influence of Catherine Parr’. In Silent but for the Word: Tudor Women as Patrons, Translators, and Writers of Religious Words, edited by Margaret P. Hannay, 43–60. Kent, Ohio: Kent State UP, 1985.
Men’s and Women’s Approaches to Tr ansl ation and Authorit y
161
King’oo, Claire Costley. Miserere Mei: The Penitential Psalms in Late Medieval and Early Modern England. Notre Dame, Indiana: U of Notre Dame P, 2012. King’oo, Claire Costley. ‘Rightful Penitence and the Publication of Wyatt’s Certayne Psalmes’. In Psalms in the Early Modern World, edited by Linda Phyllis Austern, Kari Boyd McBride, and David L. Orvis, 155–174. Burlington, VT: Routledge, 2011. Lumley, Jane, trans. Iphigeneia at Aulus. By Euripides. Edited by Harold H. Child. England: Chiswick Press, 1909. Mueller, Janel. ‘Complications of Intertextuality: John Fisher, Katherine Parr, and “The Book of the Crucifix”’. In Early Tudor Women Writers, edited by Elaine V. Beilen, 75–108. Surrey: Ashgate, 2009. Mueller, Janel. ‘Devotion as Difference: Intertextuality in Queen Katherine Parr’s Prayers or Meditations (1545)’. Huntington Library Quarterly 53, no. 3 (1990): 171–197. Mueller, Janel, ed. Katherine Parr: Complete Works and Correspondence. Chicago: U of Chicago P, 2011. Muir, Kenneth. Life and Letters of Sir Thomas Wyatt. Liverpool: Liverpool UP, 1963. The Oxford Dictionary of National Biography Online. Oxford UP, 2015. Web. https:// www.oxforddnb.com/. Accessed 31 May 2021. The Oxford English Dictionary Online. Oxford UP, 2019. Web. https://www.oed. com/. Accessed 31 May 2021. Parr, Katherine. Prayers or Medytacions, Wherein the Mynd is Stirred, Paciently to Suffre All Afflictions […] London: Berthelet, 6 November 1545. Parr, Katherine. Psalms or Prayers Taken Out of Holye Scripture. London: Berthelet, 1544. Phillippy, Patricia, ed. A History of Early Modern Women’s Writing. Cambridge: Cambridge UP, 2018. Phillippy, Patricia. ‘Introduction: Sparking Multiplicity’. In Phillippy, History, 1–24. Prescott, Anne Lake. ‘The 2011 Josephine Waters Bennett Lecture: From the Sheephook to the Scepter: The Ambiguities of David’s Rise to the Throne’. Renaissance Quarterly 65, no. 1 (2012): 1–30. Prescott, Anne Lake. ‘The Pearl of the Valois and Elizabeth I: Marguerite de Navarre’s Miroir and Tudor England’. In Silent but for the Word: Tudor Women as Patrons, Translators, and Writers of Religious Works, edited by Margaret Patterson Hannay, 61–76. Kent: Kent State UP, 1985. Richardson, David A. ‘Humanist Intent in Surrey’s Aeneid’. English Literary Renaissance 6 (1976): 204–219. Songes and Sonettes, Written by the Right Honorable Lorde Henry Howard Late Earle of Surrey, and Other. Edited by Richard Tottel. London: Tottel, 1557. Snook, Edith. Women, Reading, and the Cultural Politics of Early Modern England. Aldershot, UK: Ashgate, 2005.
162
Gender and Position-Taking in Henrician Verse
Snyder, Susan. ‘Guilty Sisters: Marguerite de Navarre, Elizabeth of England, and the Miroir de l’âme Pécheresse’. Renaissance Quarterly 50, no. 2 (1997): 443–458. Southall, Raymond. ‘The Date of Wyatt’s Psalms’. English Studies 71, no.6 (1990): 496–500. Stevenson, Jane and Peter Davidson. Early Modern Women Poets (1520–1700): An Anthology. Oxford: Oxford UP, 2001. Tadmor, Naomi. The Social Universe of the English Bible: Scripture, Society, and Culture in Early Modern England. Cambridge: Cambridge UP, 2010. Tudeau-Clayton, Margaret. ‘Supplementing the Aeneid in Early Modern England: Translation, Imitation, Commentary’. International Journal of the Classical Tradition 4, no. 4 (1998): 507–525. Tudeau-Clayton, Margaret. ‘What is My Nation?: Language, Verse, and Politics in Tudor Translations of Virgil’s Aeneid’. In The Oxford Handbook of Tudor Literature, 1485–1603, edited by Mike Pincombe and Cathy Shrank, 389–403. Oxford: Oxford UP, 2009. Twombly, Robert G. ‘Wyatt’s Paraphrase of the Penitential Psalms of David’. Texas Studies in Literature and Language 12, no. 3 (1970): 345–380. Uman, Deborah. Women as Translators in Early Modern England. Newark, Delaware: U of Delaware P, 2012. Virgil. Eclogues. Georgics. Aeneid: Books 1–6. Translated by H. Rushton Fairclough. Loeb Classical Library 63. Cambridge, MA: Harvard UP, 1916. Walker, Greg. Writing under Tyranny: English Literature and the Henrician Reformation. Oxford: Oxford UP, 2005. Weir, Allison. Henry VIII: The King and His Court. New York: Ballantine, 2001. White, Micheline. ‘Katherine Parr, Henry VIII, and Royal Literary Collaboration’. In Gender, Authorship, and Early Modern Women’s Collaboration, edited by Patricia Pender, 23–46. Palgrave: 2017. White, Micheline. ‘The Psalms, War, and Royal Iconography: Katherine Parr’s Psalms or Prayers (1544) and Henry VIII as David’. Renaissance Studies 29, no. 4 (2015): 554–575. Wyatt, Thomas, Sir. Certayne Psalmes Chosen out of the Psalter of David / Commonlye Called thee. VII Penytentiall Psalmes, Drawen into Englyshe Meter by Sir Thomas Wyat, wherunto is Added a Prolage of Auctore before Euery Psalme, Very Pleasaunt & Profttable to the Godly Reader. London: Thomas Rayuald and John Harryington, 1549. Zaharia, Oana-Alis. ‘“De interpretatione recta…”: Early Modern Theories of Translation’. American, British, and Canadian Studies 23, no. 1 (2012): 5–24.
5.
Transcription as Translation: Writing the Language of Manuscript Poetry Abstract: This chapter considers the strategies of transcription in circulated manuscripts, where play on the page is the core concern. Transcription often involved studied revision on a scale ranging from particular pronouns to entire stanzas. Like the practice of translation, transcription was understood as a method through which courtiers could reimagine a text, making works more immediately resonant with their world. The imaginative engagement with the text is of a different sort than that in translation alone, though, because the text can be more explicitly rearranged or broken apart to join with other texts. This chapter examines the products of the Devonshire manuscript as a collaborative project where transcription unites with multiple authorship. Keywords: Devonshire Manuscript; Tudor verse transcription; Margaret Douglas; Mary Shelton; Mary Fitzroy; early modern women’s writing
As established in the previous section, courtiers’ practice of adapting translation to express frustration in the Henrician court has received a great deal of critical attention since Stephen Greenblatt first outlined the practices of ‘courtly making’;1 more recently, attention turned to the ways in which transcription might serve similar purposes. Largely independent of this question, research has also begun to explore the ways that manuscript production could create and strengthen communities and the contributions of women to that production. Here, I want to demonstrate how we can unite these threads to reinterpret both how manuscript production was used, particularly by women, and, as a result, how we might re-read translations by the more thoroughly researched courtly makers like Wyatt and Surrey. Alongside comparison to translated work by courtly women like 1 Greenblatt, Renaissance Self-Fashioning.
Quoss-Moore, R.M., Gender and Position-Taking in Henrician Verse: Tradition, Translation, and Transcription. Amsterdam: Amsterdam University Press, 2023 doi 10.5117/9789463723534_ch05
164
Gender and Position-Taking in Henrician Verse
Elizabeth and Lumley, we can use new understandings of manuscript work to better contextualize those thoroughly studied translations as working within the larger position-taking system outlined in this project, both informed by and informing other writers’ strategies of verse production and courtly critique. Speaking of the Devonshire Manuscript specifically, Arthur F. Marotti articulates the collection’s function as ‘the medium of socioliterary intercourse within a restricted social group’2—an exemplar, of sorts, of his larger arguments that manuscript verse ‘was embedded in specific social situations’ to which writers responded ‘in terms of shared sociocultural assumptions’.3 Manuscripts offered a useful response space for a range of reasons, not least that, as H.R. Woudhuysen outlines, they allow ‘authors and scribes to reach precisely the audience they wanted to address’.4 That audience element is densely recursive, because ‘a reader in a manuscript culture, with a fluid text subject to change, is responsible for participating in literary production as well as consumption’.5 While I attempt, here, to engage those elements of coding and reception whose outlines seem most traceable, these efforts are frustrated (even as they are also enabled) by the nature of manuscript verse production: ‘the narrower the audience, the more specifically targeted it is, and the more personalized both the means of production and mode of distribution […] the less need be said about it’.6 As we move into the last section of the work, my arguments around precise interpretations become somewhat fuzzier, not least because these authors are both being more opaque in their writing and writing in more insularly interpreted networks. Meaning becomes less stabilized because the shared context is more stable and specific; writers want to access greater flexibility for those outside of the creating circle to maintain a protective insularity amongst the manuscript’s contributors. Peter Beal summarizes some of the more positive elements of this opacity: ‘manuscript circulation provided an especially immediate and congenial culture in which […] risqué, topical, and selectively communal material could most effectively flourish: when the interaction between author, text, and reader of such things was at its liveliest, most coordinated, and most appropriate’.7 Particularly in courtly poetry, where ‘poems were an extension of artful, polite behavior and, at the 2 Marotti, English Renaissance Lyric, 39. 3 Marotti, 7. 4 Woudhuysen, Circulation of Manuscripts, 12. 5 Ezell, Social Authorship, 40. 6 Beal, In Praise of Scribes, 18. 7 Beal, 19–20.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
165
same time, ways of formulating actual or wished-for social transactions’,8 the personal and the private become essentially inseparable,9 partly as an intentional function of the practice and partly because we are outsiders to this community and its ‘appropriative and social reading habits’.10 Not least, we must balance the need for contextual information against an acknowledgement that, as Marcy North argues, ‘[a]nonymity […] brings the scholar much closer to early conceptions of authorship than naming does, because it can represent collaboration, compilation, and the communal production of literature’.11 This sets up a particularly complex set of issues for consideration around attribution within socially created manuscripts, where personal context certainly informs interpretation and where anonymity and collaboration are intentionally deployed. In summation, transcription in manuscripts offered a communal form of production, a distinct kind of deniability, and specific structural context relationships with other writers and other works. Woudhuysen observes that ‘women in particular found the use of the manuscript to their advantage’,12 and these elements seem key to that efficacy. Women’s contributions must be understood not as just another access point or contributing element for analysis, but rather as a defining force in the practices of transcription and sharing of verse in Henrician manuscripts. As Margaret J.M. Ezell writes, ‘the manuscript text operates as a medium of social exchange, often between the sexes, neither private nor public […], and a site at which women could and did comment on public issues concerning social and political matters’.13 Partly, this commentary manifests in the distinctly communal nature of manuscript poetry of this type, where, as Victoria E. Burke and Jonathan Gibson note, ‘[t]ranscribers were rarely merely passive copyists’14 and where, as Patricia Phillippy argues, a more sensitive consideration of ‘[t]raditionally passive acts of reception—reading, but also […] translation, transcription, and editing—qualify the image of an autonomous author as the sole creator of her work’15 . Women’s production so defined the genre that even manuscripts which do not contain women’s hands likely can still be understood as conforming or reacting to the larger cultural phenomenon, where women’s 8 Marotti, English Renaissance Lyric, 9. 9 Heale, ‘Female Voices’, 15; Stevenson and Davidson, Women Poets, xxxviii. 10 Solomon, ‘Representations’, 671. 11 North, Anonymous, 4. 12 Woudhuysen, Circulation of Manuscripts, 13. 13 Ezell, Social Authorship, 40. 14 Burke and Gibson, ‘Introduction’, 6. 15 Phillippy, ‘Sparking Multiplicity’, 15.
166
Gender and Position-Taking in Henrician Verse
role was central. The analysis of transcription practices here will focus on the Devonshire: a text that does contain women’s hands among the twenty some autographs, likely owned by a woman, and composed primarily, but not entirely, in the 1530s.16 The Devonshire is usually understood to contain much more transcription than original composition, but neither original author nor transcriber is often clearly designated. These features make the manuscript an ideal sample for understanding why manuscript production, one of the primary modes of poetic composition at Henry’s court, might deemphasize originality and sole authorship and instead emphasize social composition and styles of adaptation. Transcription did not necessarily, or even usually, mean the strict and exact re-inscribing of a section of text. In fact, the practice often involved studied revision on a scale ranging from particular pronouns to entire stanzas, often offering an entirely different effect from that of the original piece. The effect of transcription also depends largely on context, and the Devonshire serves to demonstrate courtiers’ proficiency at placement. Like the practice of translation, transcription was understood as a method through which courtiers could reimagine a text, making works more immediately resonant with their world. This underscores the sense in which such verse represented position-taking; whether the energies expressed were essentially conservative or progressive, the preserved systems of coding are activated in an inherently political process of participant negotiations. The strategies of transcription, like translation, also allow writers a certain freedom in their critiques, particularly of the King, because a lack of ownership is inherent to both forms. Compared to translation, transcription allowed for both greater flexibility and a different kind of deniability. The imaginative engagement with the text is of a different sort, because the text can be more explicitly rearranged or broken apart to join with other texts and because the text was communally produced.17 This chapter examines the practices of manuscript contributors, particularly women, and looks at the products of the manuscript as a collaborative project where transcription 16 The most authoritative and comprehensive sources on the Devonshire, cited throughout this work, are the Social Edition and Heale’s edition. 17 Discussing selections and placement in less communally involved manuscripts, Burke identifies that ‘[s]tructure can take several forms in the manuscript itself, thematic structure in terms of its contents, and formal and rhetorical structure enabling us to consider aesthetic categories. When we consider these manuscripts as comprised of individual items which can be read dialogically with other items, miscellanies can be appreciated as dynamic literary artefacts’ (‘Materiality and Form’, 237). Communal production further energizes the dialogic potential of the manuscript.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
167
unites with multiple authorship. That project works to preserve and promote position-taking energies, as authors transcribe in a complex system of reciprocal validation, enlarging and reworking the themes of their own and others’ work even as those themes gain power from the extant traditions and exchanges. The Devonshire’s effect is best examined as that of a whole document, or with the understanding that every poem is entered in the context of those already existing and in awareness of those that will later be entered. Bradley J. Irish notes that ‘[t]he current binding can be tentatively dated to the early to mid-sixteenth century’, so the manuscript’s physical form today is likely very nearly the same as, if not identical to, the original book as it appeared before any of the pages were filled.18 Irish then summarizes the process of the manuscript’s creation thusly: ‘[m]embers of the queen’s circle seem to have entered poems on an ad hoc basis, with little intention of compiling a finite or bounded collection’.19 I want to consider the relationships here through a different angle, though, considering that while, as Irish points out, a blank book doesn’t precisely determine the volume or type of contents that will be filled in over time, it does still provide a type of boundary—and considering that its use by a community of writers provides a kind of format that affects composition and encourages structurally informed reading and writing practices. Because the manuscript was in its current form, it was already, in certain ways, a finite, bounded collection—not in the sense that the shape of the words within was predetermined, but in the sense that the elements’ spatial relationships and the limits of the text were already defined. When a transcriber moved to include a poem, that poem could be placed in a particular context or placed in a space that made it a work to be commented on. Powell comments on both the sense that the manuscript constitutes ‘a variety of uses and social environments […] centred at different points in the manuscript’ indicated by distinct sections and groupings20 and the strongly dialogic interactions in most of those spaces,21 a feature identified in women’s manuscript work more broadly by, among others, Burke.22 Marotti specifically observes that ‘[b]lank pages or blank spaces at the bottoms of pages invited compilers as well as others into whose hands manuscripts fell to insert their own poems’.23 Given the manuscript practice 18 Irish, ‘Gender and Politics’, 87. 19 Irish, 87. 20 Powell, ‘Marginalia’, 2. 21 Powell, 3. 22 Burke, ‘Materiality and Form’. 23 Marotti, English Renaissance Lyric, 173.
168
Gender and Position-Taking in Henrician Verse
as demonstrated in the text, writers who chose less inhabited pages must certainly have expected that others would crowd around their works or actively comment on them. We can return to and expand, as well, Irish’s reference to ‘the queen’s circle’; the manuscript may have circulated during the reigns of Anne Boleyn, Jane Seymour, Anne of Cleves, Katherine Howard, Katherine Parr, and, in the case of the latest entry in the manuscript, Elizabeth I. Considering these potential contexts emphasizes the interplay of both the internal context of the placement of poems in the manuscript and the external contexts of court events. Not only does each internal context offer particular interpretive possibilities, but the rapid succession of queens emphasizes the increasingly anxious atmosphere of a court where five women served as queen within the space of eight years. Any given woman who wrote in the manuscript could have been part of a queen’s circle one year and not the next. The writers of the manuscript respond to this atmosphere by expressing their fear, their frustration, and their resentment in their collaborative poetic project, preserving their voices of resistance against Henry from within a culture where outlets for such expression were increasingly circumscribed. The nature of manuscript collections at the time emphasizes the importance of context and the reason a manuscript collection would be an attractive outlet for frustrated courtiers to share their thoughts. Micheline White has contributed some of the most important recent work in understanding women’s literary networks as networks—that is, as communities where knowledge, ideologies, and strategies are intentionally shared to create markers of affiliation24—while Susan Frye shows that manuscripts later in the sixteenth century were used to demonstrate connection to a particular community.25 Jane Donawerth then outlines the important dimension of valuation added by circulation through and between such communities.26 The systems converged in the ways that verse could be used to ‘establish a larger political support community and even to influence political decisions’.27 Manuscripts offered another dimension of support because their value included ‘the marks of caring, of personal concern, and of previous owners attached to them’.28 Manuscript production could create position-taking and -interpreting communities on an intimate level—bound 24 White, ‘Women Writers’. 25 Frye, Pens and Needles. 26 Donawerth, ‘Women’s Poetry’. 27 Donawerth, 18. 28 Donawerth, 8.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
169
together through reciprocity and through their modes of, and outlets for, expression. These bounds help explain the relative insignificance of original authorship indicated by the lack of attribution and the practice of free borrowing inherent to Tudor poetic production. Certainly, the Devonshire, like other manuscripts of the time, frequently incorporates signatures, initials, and personal emblems, but the relationships of these signs to signs of authorship are rarely straightforward.29 This expectation of assigned authorship, moreover, ‘is dependent upon the notion that assigning authorship is what matters’.30 Instead, the manuscript indicates that, as David R. Carlson observes, ‘the considerable bulk of the courtier-poetry was circulated and collected at first, among the courtiers themselves, without attribution, as if personal authorship of most individual poems was of no or only occasional consequence among the people who first wrote and read them’.31 Indeed, I would concur that Carlson’s own analysis demonstrates the ‘as if’ in this phrase could be replaced with ‘because’. In his analysis of Wyatt’s style, Jeff Dolven neatly argues that at court, such alignment between ‘authorship and […] signature style would have spoiled [the] traffic’ in manuscript poetry.32 The poem in a manuscript is a means of showing one’s ability to adapt a phrase to express something about oneself. Dolven uses sartorial style as an analogy; courtiers don one another’s compositions and adapt them as people don clothing made by someone else, not out of a lack of originality but to showcase the original way in which they select, alter, combine, and wear the clothing.33 In the cases where the writers of the Devonshire do take care to record some kind of ‘ownership’, the identities are as likely to be that of the transcriber, the object, or the dedicatee as of an ‘author’ in our modern sense. That slippery sense of authorship is a central tenet of Tudor poetry that can be easily overlooked when the focus narrows too far on ideas of personal identity expressed through the composition of a poem rather than through the 29 As Woudhuysen summarizes, compilers of miscellanies who put names or initials to the poems they collected were not acting under oath: ascriptions might be added, deleted, or changed for a variety of reasons. Some compilers place their own names or initials beneath poems as witnesses that they were responsible for the transcription. The same names or initials might serve to remind the compiler who had supplied him with the poem he had copied […] That the name attached to a poem indicates precisely who composed it is only one of a number of possibilities. (Circulation of Manuscripts, 160) 30 Goldberg, ‘The Female Pen’, 21. 31 Carlson, ‘Henrician Courtier’, 162. 32 Dolven, ‘Reading Wyatt’, 77. 33 Dolven, 71–77.
170
Gender and Position-Taking in Henrician Verse
expression. The way in which a writer manipulated or presented a particular poem could say quite as much as an ‘original’ composition. One contributing factor to this presentation was the way in which ‘the intersection of the visual and the verbal created power signifiers’.34 Each time a contributor to the Devonshire added some kind of attributive or identifying mark, the meaning of the entire manuscript was altered. Some of this interpretive power is lost to us, because we will never have full access to the original audience’s knowledge of relationships and writing styles. As Solomon acknowledges, these groups would have been deeply familiar with one another and ‘sensitive to subtle idiosyncrasies of expression in matters of calligraphic style, pressure of pen strokes, punctuation, and use of space’.35 Given the difficulty in modern scholarship of even assigning a specific hand to a potential writer, a full restoration of such considerations is beyond our reach. Other elements of this power, though, carried over into print production and preserved for us examples of the ways such signifying power could be interpreted. Goldberg offers an example of just this kind of practice when he examines Tottel’s use of Surrey’s name as the most prominent for the miscellany. Tottel’s original title for the anthology was Songes and Sonettes Written By the Ryght Honorable Lord Henry Howard, late Earle of Surrey, Thomas Wyatt the Elder and others. The title functions ‘as an authorizing name […] a style that is announced and to which the poems are subsumed—a style, which, as everyone knows, is Tottel’s imposition in terms of editorial changes, rewritings, the provision of titles for poems, and the like’.36 Tottel gains authority through his use of Surrey’s name, but he also signals that he will be using a certain style to conform the poems to the vision of his own collection. Our own language of reference for the Songes and Sonettes points to our view of the authorizing and editorial role of Tottel, who freely revised as he edited. Dolven makes the point that ‘[t]he self-fashioning power of these poems was by no means limited to their first authors’.37 The power of the manuscript is that it both signals the conforming power of style and announces the differences of self-fashioning, with each new hand, new version, or new identifying mark. The end effect of this process is nicely expressed by Carlson: ‘a poetry for which no-one can be held responsible […] corporate submersion rather 34 Frye, Pens and Needles, 55–56. 35 Solomon, ‘Representations’, 672. 36 Goldberg, ‘Female Pen’, 22. 37 Dolven, ‘Reading Wyatt’, 78.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
171
than individuation’.38 The corporate submersion is not just the standard approach to writing; such submersion is a technique which strengthens the possibilities for protected position-taking by denying responsibility. If a poem cannot be attached to a particular author, or if a writer can disavow an effect as merely a product of the process of copying, then positions which act in opposition to, for example, monarchical power can be protected. The authors have, in effect, plausible deniability. As North argues, ‘authors turned name suppression into a tool for the making of names and for the preservation of social boundaries. Combined with the female voice [especially], anonymity created the illusion of authenticity only to make authentication impossible’.39 While such denial will also remove some of the power of such energies—denying an authorizing name or an individually identifiable stance—the ability to express resistance even within constraint is elemental to this kind of verse position-taking.
A. Context and ‘Correctness’ in Manuscript Transcription The f irst poem of the manuscript provides an elegant prologue to the position-taking effects of the collection as a whole, particularly as a potential commentary on the Douglas-Howard affair or, at a minimum, on the larger surveillance culture that led to that affair’s very public end. This first poem reflects the ways that the collection has caught readers’ and scholars’ attention as a source for Wyatt’s poetry, as a witness to some of the particularly dramatic moments at Henry’s court, and, most recently, as a source for understanding women’s role in early Tudor book culture.40 The work seems to be identified as Wyatt’s—and has since been often attributed as such—and focuses on the narrative of a forbidden love affair discovered and punished by a culture of surveillance. Although attributed 38 Carlson, ‘Henrician Courtier’, 168. 39 North, Anonymous, 4. 40 For further context on the Devonshire and on women’s manuscript miscellanies, see Clarke, ‘Women’s Manuscript Miscellanies’; Shirley, ‘Reading Gender’; and Heale, ‘Love Lyric’ and ‘Female Voices’. Powell offers particularly useful context on considering the manuscript in the specific context of Wyatt’s oeuvre, concluding that: the only way to understand the manuscript and the social community it represents is precisely through this attempt to ‘get inside the circle’ [North’s phrase for efforts to identify the authors of the Devonshire], including the effort to restore anonymity (or alternative authorship) to a large body of anonymous poems once considered to be ‘Wyatt’s’, and to assign firmer authorship to Wyatt and other poets where the evidence exists. (‘Marginalia’, 11)
172
Gender and Position-Taking in Henrician Verse
to Wyatt, the poem is not in his known hand, which appears nowhere in the manuscript; rather, the poem is entered in the hand generally identified as ‘Hand 1’, emphasizing the lack of ‘responsibility’ and ‘authorship’. Here, I will use intentionally the term ‘writer’, meaning simply the person who wrote a particular poem in the manuscript, thus avoiding both the complication of ‘transcriber’ where some pieces may be original compositions and the obvious problems of ‘author’. 41 This first poem provides an immediate example of the work done in manuscript that translates less easily to print culture, when the effort is made to transfer it at all. On the page, the apparent refrain ‘therefore take hede’ which ‘ends’ each tercet runs alongside, to the left, in a repeating column joined to each stanza by a series of brackets. The separation of the refrain from the poem emphasizes the message by creating a greater space for the work, adding width to the work and using all of the space on the page, unlike many entries which take up only a small fraction of the available writing area. On one hand, this may seem to counter the writer’s claim that private space is an illusion (saying that those who ‘thowgt ryght sure none had theym sene’ are misguided). 42 However, the poem is then both entered into the public manuscript, surrounded by works that offer commentary and response, and internally divided—the pieces of the poem are separate on the page. This separation also emphasizes the repetition of the refrain. That repetition may begin to incorporate the sense of absurdity that often accompanies repetition; this sense is emphasized when the repetition is interrupted by the alteration of the second to last instance of the refrain: ‘therefore take hede’ becomes ‘therefor take’. 43 The writer’s failure to complete the refrain manipulates the repetition to suggest an unwillingness to continue; a break is created that both uses and works against the refrain. The effect may mirror the strain of prescribed courtly behavioral patterns and of a surveillance society: sloppiness enters in which destabilizes the meaning of the repetition by emphasizing that it is repetition; that is, a pattern of prescribed behavior may make the behavior more obviously disingenuous, in so far as the behavior is inherently performance and as that performance is emphasized by the nature of replication. If everyone behaves in a particular manner towards the King or towards 41 Where authorship is agreed upon, it is noted, but many poems do not allow a clear distinction between the work of writing, transcribing, or creating, and this process of workmanship and alteration is largely my focus. 42 BL MS Add. 17492, 2r, ‘Take hede be tyme’, line 11. For purposes of line numbering, I consider the refrain to be the final line of each stanza, and I include the struck-through line 9. 43 Line 17.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
173
some other social superior, the performative nature of that behavior will be most emphasized when there is a trip in the performance. When someone makes a mistake or a misstep, the sameness of the other performances is emphasized. Although such performance was itself full of meaning, any break has the potential to be read as an outburst suggesting impatience with the accepted, prescriptive models. This potential for disruption makes the writer’s decision not to ‘fix’ the line potentially important. Writers frequently correct themselves throughout the manuscript; writers are also not shy about correcting one another’s work. Of course, not every work in the manuscript is subjected to such editing, but it is also worth noting that this poem, on the first ‘text’ page of the manuscript, likely would have been subject to particularly frequent viewing and so often considered for editing or alteration. While an argument could also be made that, as the first page of content, it may have frequently been skipped over—in much the manner a modern reader might skip an oft-read preface or a title page—this seems unlikely. The page that now serves as the very first page, or sort of flyleaf, for the Devonshire is covered in doodles and unrelated signatures, suggesting that the first few pages were often revisited by different writers. Certainly, a person executing some flourish or sign on the first ‘blank’ page might then skip forward, but the evidence suggests that the many writers of the manuscript viewed this poem repeatedly and chose not to ‘correct’ or alter the original form, perhaps recognizing the line as more thought-provoking in its original state. Another set of lines illustrates ways that corrected mistakes can be used to enrich an interpretation of manuscript practice. Between the second and third stanzas, the line ‘for in lyke case there sselv of dyveris skools’ has been struck through. 44 This seems, on the one hand, a fairly simple and honest mistake, likely to indicate that the writer was copying from another written version of the poem.45 Apparently, the writer has started with the first six words of the third stanza, whose first line reads ‘ffor in lyke case ther selves 44 Line 9. 45 Research with manuscripts also often, necessarily, opens the possibility that such revisions are the composing work of the original author—changing a word or starting with one line and then choosing to use the phrases to different effect, variously. As the general consensus includes this poem in Wyatt’s canon, and this is not Wyatt’s hand, I have set aside such considerations here. However, a core point of my argument is that poems that we give single authorship arise from much more complex composition practices. Although I feel copy work on 2r is the simplest answer in this case (as the writer seems likeliest to be accidentally jumping, in reading and copying, between the first part of one line and the second part of another), such a simplification could still always be mistaken.
174
Gender and Position-Taking in Henrician Verse
ha hathe bene’46 and then, in moving between reading and writing, skipped to the fifth word of the fourth stanza, where the first line is ‘all thowgth theye be of dyvers skoolles’.47 Assuming that the writer is working from copy does not eradicate the interest of the deleted line, and in fact draws the eye to some interesting discrepancies. In the deleted line, ‘for’ takes the single ‘f’, ‘there’ includes a terminal ‘e’, and ‘sselv’ acquires a double ‘s’; ‘dyveris skools’ is similarly of markedly different spelling than in the recopied line one stanza later.48 Interpretively, only ‘there’ seems to open up much difference, but the fact of the difference suggests the fluidity of the copyist’s work. The writer, this fluidity indicates, was not slavishly copying the version from which she worked letter for letter, which suggests that the writer originally saw this line as making sense, grammatically and with the sense of the poem.49 Considering the line as a potentially ‘correct’ line does add a particular sense of threat not elsewhere included: duplicity or multiplicity within the person or people who threaten the poem’s object. They are of various schools. Particularly, this duplicity or multiplicity can be understood, within the line, to refer to a time when those people have also sought a forbidden or hidden love affair; that is, when they were ‘in lyke case’, they were, within ‘there sselv[,] of dyveris skools’.50 The ‘correct’ lines do not offer such an alternative; the people who threaten the object may be of ‘dyvers skoolles’, but there is no suggestion that the division is internal; rather, several different factions are united together against the object.51 The poem gains force from both meanings, but both meanings are only possible because the writer at least momentarily saw the ‘mistaken’ line as a logical possibility. The temporal context of the poem then informs this mistaken line, written in a court in the throes of at least two significant scandals based in such ‘hidden’ love affairs—both the Douglas-Howard marriage and the accusations against Anne Boleyn. In the mid-1530s, the hypocrisy of a ruler who condemned hidden affairs would have been a topic whose critical force was apparent to all courtly readers. The writer’s choices offer politically resistant interpretations not only in these lines whose difference marks them as ‘wrong’, but also in considering 46 ‘Take hede be tyme’, line 10. 47 Line 14. 48 Line 9. 49 My use of ‘she’ as a pronoun for Hand 1 is explored later in this chapter. A different scenario, worth noting, is that of a copyist working as a peer read another version; this case, though, still indicates the potentials of alternate versions of the verse. 50 ‘Take hede be tyme’, line 9. 51 Line 14.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
175
some of the more difficult spellings presented by the poem in this transcription. Though the entire manuscript, of course, exists without an external standardized spelling, this does not at all mean that the authors are not concerned with spelling. Helen Baron outlines this as a particular concern of Mary Fitzroy, and indeed uses this concern to link her extant letters, with their many line edits, to her work in the Devonshire.52 Many different hands reflect this concern, where words are marked out and replaced with a preferred spelling; poets were well aware of the multiple potential meanings of a word. These changes in the choice of word forms and spellings, then, may at least partly be understood as an attempt to create or contain particular meanings; Jonathan Goldberg explores this possibility in Mary Shelton’s transcriptions, particularly, pointing out that her spellings seem often designed to elude easy interpretation.53 The analysis extends to the other hands, though it should not be separated from Baron’s points about Fitzroy’s concerns for correct performance of status. The two projects are, in fact, linked: a person of importance and intelligence knows the ‘right’ spelling, but she also knows that the right spelling is the one most appropriate to her project—that is, the spelling most likely to evoke alternative connotations useful to her and to eliminate those possibilities which might undermine the larger sense of her transcription or composition. Women poets may have been particularly attuned to these difficulties, given the impossible terms of ideal woman’s courtiership, as famously expressed in Castiglione.54 Women’s spelling becomes a manifestation of that ‘difficult mean’, both meaning more than it says and resisting readers’ efforts to pin down meaning. There are several examples of this kind of work in the poem; for example, the somewhat difficult ‘Sewrlye there les ye ^can te not^ nott blynde’.55 The sense of this line almost demands ‘eyes’ for ‘les’, although ‘less’ can also be made to work with some effort, but the line becomes more interesting for the interruption caused as the reader puzzles through the meaning. Potentially, the reader can understand ‘les’ for ‘lies’, as a missing ‘y’ makes as much sense as ‘les’ as a spelling for ‘eyes’.56 Further, ‘sewryle’ can be read ‘surely’ or ‘severally’—that is, the line can be read as simply ‘surely their eyes you can’t blind’ or ‘severally their lies you can’t not blind’.57 There are, 52 Baron, ‘Fitzroy’s Hand’, 318–335. 53 Goldberg, ‘The Female Pen’, 17–38. 54 Castiglione, Courtier, 212–213. 55 ‘Take hede be tyme’, line 7. 56 Line 7. 57 Line 7. Heale decides on ‘surely their eyes ye cannot blind’, in her edition (Devonshire, 52, ‘Take heed betime’, line 7).
176
Gender and Position-Taking in Henrician Verse
of course, a number of possible combinations involved here, as well; the confusion calls special attention to the line, particularly the troubling ‘les’. The word becomes indecipherable and undefinable; even as the line itself insists this subject is beyond the object’s control, it is also beyond the reader’s. The reader, then, assumes the object position, unable to hide and unable to control or even fully understand the thing that sees, understands, and at least effects, if does not control, the object/reader. In this sense, the object is a parallel to the Henrician courtier, and the message becomes a warning. In a court ruled by a paranoid king, even personal choices, like love affairs, can become treason; the poem represents the impulse to hide one’s actions, in this context, as natural and perhaps universal, but also as essentially futile. Simultaneously, though, the poem offers its own success in eliding full interpretation. Control may not be possible—but perhaps full knowledge is equally inaccessible for others on the spectrum of power. The theme of the poem—the impossibility of secret love in a society of surveillance—also touches on several motifs of the manuscript selections generally. First, the secrecy of love does not seem to imply inconstancy or falsehood within the affair. Although the lovers wish to be ‘unspyde’,58 the secrecy is the only indication that something inappropriate may be happening, and even this sense depends on the audience’s contemporary anxieties. Second, this secrecy is understood to be temporary: through the manuscript, temporality is emphasized, as is change, both in the scenario of the poem and by the nature of composition and contribution over time. Even when lovers are constant, the world is not. The instability of such attempts at secrecy is further underscored by the fact that the poem, about a wish for secrecy, itself publicizes the lovers’ relationship, albeit anonymously. Third, love is threatened by an external force which, fourth, comes from surveilling others. Additionally, the poem is a warning about the impossibility of escaping that threat, and, if the deleted line is understood as nonetheless a part of the composition as it exists, about the hypocrisy and duplicity of the individual who threatens. Finally, the poem works to encourage reader identification with the object, both by the typical use of direct address to a putative listener and by the process of obscuring meaning to involve the reader in the process of interpretation and frustration of understanding. From within the surveillance culture of the unstable Tudor court, manuscript production offered an opportunity to frustrate that surveillance, as such poetry was ostensibly not entirely or originally the author’s. 58 ‘Take hede be tyme’, line 20.
177
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
Only a few pages later, another entry highlights some of the opportunities for manuscript production as a space to register dissatisfaction with the terms of service in the Tudor system of surveillance. Carlson has noted, in early modern manuscripts generally, ‘the phenomenon of the repetition of the same item [i.e. poem] more than once in the same manuscript anthology, regularly rather than exceptionally’.59 A repetition of a poem on 3r occurs much later in the volume, on 75v. The poems are both versions of a work accepted as Wyatt’s translation of Aquiliano. The poem centers on a complaint of insufficient remittance for service offered and of the inability of the victim to act effectively against this insuff iciency. The version on 3r omits a line, while throughout the version on 75v the transcriber inserts periods (not necessarily used to indicate sentence endings). The periods stabilize the meaning of the poem in some sense, emphasizing the sarcastic tone of the speaker. But the effect also somehow feels more distant, as if 75v’s speaker speaks from closure while 3r’s does not. This sense is perhaps emphasized by the relative placements—of 3r on its own page, while 75v is only one of several poems inscribed in a much smaller hand. That is, the placement of the event on the page suggests that it can be contextualized and so minimized for writer on 75v, but looms larger for 3r.60 Although I cannot replicate that effect, I do include the two poems here for comparison: Example from 3r My harte I gave the not to do it paine 1 But to preserve was yt was to the taken I served the not to be forsaken but that I should be rewardyd againe I was content they slave to remain 5 but not to be paid vnder suche fassyon nowe sins in the ys no maner of reason
Example from 75v My herte I gave the not to do yt paine but to preserve / yt was to the takin I seruid the not to be forsakin but that I shulde be rewardid againe I was content thy seruante to remaine but not to be paide vndre suche fasshion now sins that in the is none other Raison
59 Carlson, ‘Henrician Courtier’, 168. 60 BL MS Add. 17492, 3r, ‘My harte I gave’ and 75v, ‘My herte I gave’.
1
5
178
do displease the not tho I do reffreyn vnsacyate off my wo and my desyer ffar well I say partyng ffrom the ffyre ffor he that beleves leryng in hand ploues in the water and sows in the sand61
Gender and Position-Taking in Henrician Verse
10
Displease the not if that I do restraine vnsatiat of my woo . and thy desire assurid bye crafte. texcuse thye 10 faute ffarwell I saie parting from the fire for he that beleuith bering in hande plowithe in water and sowith in sande /62
The most drastic change in the two versions is the deletion, in 3r, of the tenth line usually included in versions of the poem.63 As a result of that deletion, if the lines are strung together, the ‘sentence’, as it were, in 3r reads ‘nowe sins in the ys no maner of reason / do displease the not tho I do reffreyn / vnsacyate off my wo and my desyer’;64 alternatively, the ‘sentence’ can be understood to end at ‘reffreyn’, starting a new phrase at ‘vnsacyate’.65 The ‘sentence’ of 75v, though, is markedly different, as indicated by the shift from ‘my’ to ‘thy’,66 and made even more so by this writer’s decision to insert punctuation: ‘now sins that in the is none other Raison / Displease the not if that I do restraine / vnsatiat of my woo . and thy desire / assurid bye crafte. texcuse thye faute’.67 The added line separates the ‘ffarwell’ from the speaker’s choice to refrain or restrain,68 as the verb is now instead joined to the first part of line nine, while the second half of the line becomes the beginning of the phrase which ends in 75v’s line ten. The potential meanings of the two poems open up in quite different directions. For 75v, the speaker may be 61 BL MS Add. 17492, 3r, ‘My harte I gave’. 62 BL MS Add. 17492, 75v, ‘My herte I gave’. My footnotes for this analysis will maintain the page references, given the nearly identical first lines of the two versions. 63 Both poems actually eliminate another line from the Egerton MS which I consider, for current purposes, the ‘standard’ version of the poem. This version restores the Petrarchan rhyme scheme, as line eleven (that is, the line that would be placed between lines 10 and 11 of the version on 75v) reads ‘But syns it please thee to fain a default’ (in Collected Poems, 13). 64 3r, ‘My harte I gave’, lines 7–9. 65 Heale compromises with a semicolon at the end of line 9 in her editing of both versions—that is, following ‘refrain’ in 3r and following ‘restrain’ in 75v (Devonshire, 54, 201). 66 75v, ‘My herte I gave’, line 9. 67 Lines 7–10. 68 Line 11.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
179
understood to say that they will likely restrain from excusing the beloved’s fault; that they will no longer be pursuing the beloved and that the beloved’s desire will excuse the beloved’s fault; or that they will stop the pursuit, disappointing both the speaker’s and the beloved’s aims or desires. 3r, in contrast, says much more simply that the lover plans to stop, and that the beloved addressed should not/shall not be displeased by this cessation. The implication, since both ‘wo’ and ‘desyer’ are attributed only to the speaker,69 is that the addressed person could only be upset because there is ‘no maner of reason’ in them.70 The pronoun change in the two poems, though subtle, is quite important, since it further emphasizes the likelihood that 3r either intentionally altered the poem or was working from a different draft which prioritized subtly different concerns. In 3r, the speaker has not had either ‘wo’ or desire fulfilled, or, potentially, the object has not been satisfied by the speaker’s woe or desire; in 75v, both the speaker and the addressed are deprived of something, since one ‘woo’s and one ‘desire’s.71 The difference in spelling also opens up greater potential meaning for Hand 1, who may claim to be deprived of woe as well as woo—the two near homonyms are linked by the tone of the speaker’s complaint. If she addresses the object as ‘vnsacyate’, she points out the illogic of desiring her woo or woe; if she herself is ‘vnsacyate’, then she combines her suit with her sadness, suggesting the two are interchangeable.72 3r’s choices make the poem more clearly resistant and resentful, emphasizing the potential dangers of this un-reciprocal relationship by making the speaker more clearly the single victim—and a victim who has no recourse. Such small changes in both spelling and in word choice reverberate through the tones of the two versions. In the first line of the poem, Hand 1 transcribes ‘harte’ where Hand 8 transcribes ‘herte’.73 Hand 1 thus connects to expected tropes unifying love and the hunt through the figure of the hart/heart; Hand 8 instead calls to mind the theme of the hurt heart. The verb choices ‘reffreyn’ and ‘restraine’ adopt and encourage subtly different moods.74 To refrain is a sort of negative action; a not-doing that is somewhat passive. To restrain implies, rather, to hold back actively. Rather than passively letting an opportunity pass by, the speaker in 75v is instead working to hold something back or working against something. One further alteration 69 70 71 72 73 74
3r, ‘My harte I gave’, line 9. Line 7. 75v, ‘My herte I gave’, line 9. 3r, ‘My harte I gave’, line 9. 3r, ‘My harte I gave’, line 1, and 75v, ‘My herte I gave’, line 1. 3r, ‘My harte I gave’, line 8, and 75v, ‘My herte I gave’, line 8.
180
Gender and Position-Taking in Henrician Verse
between the two versions is in their respective fifth lines, where Hand 1 writes ‘I was content they slave to remain’.75 Hand 8 preserves almost the same line, with some spelling changes or corrections, but also changes ‘slave’ to ‘seruante’, restructuring the implied terms of service.76 Like the difference in verb choices, the difference suggests a greater submission or a greater passivity for the speaker in 3r, and a more active speaker, more of an agent, in 75v. The last line may be particularly rich for identifying evidence, given the work Terttu Nevalainen has produced on gendered linguistic shifts in Tudor-Stuart England. One of Nevalainen’s studies focuses on the use of (e)s or (e)th endings, which, she finds, ‘became associated with register differences in the evolving standard language, -(e)th with formal and literate styles and -(e)s with informal and oral’.77 Further, the two endings began to be used along gendered lines, as well as regional: In the first four decades of the sixteenth century, the regional patterns of diffusion remain almost the same: the North continues to promote -(e)s, and the Court and East Anglia resist it. London’s weight, however, drops below 0.5. Although women’s overall frequency of the variable is low, a difference also begins to emerge between the sexes: women favoring -(e) s, men disfavoring it.78
The difference is particularly important, here; however little we know about the writers of the Devonshire in some cases, we do know that the writers would all have been at court, where the (e)th ending is found to be significantly preferred. Hand 1, transcribing the selection on 3r, uses ‘(e)s’ for the endings of verbs in the final two lines; Hand 8, transcribing the selection on 75v, prefers the ‘th’ conclusions. Although naturally insufficient to conclusively gender, much less identify, either hand, the evidence is still an important element of the characterization of each writer. Hand 1 is not consistent in the (e)s/(e)th use, sometimes using both in one poem 75 3r, ‘My harte I gave’, line 5. 76 75v, ‘My herte I gave’, line 5. The use of ‘slave’ as opposed to ‘servant’ may also be echoed in Surrey’s language choices in ‘Love, that liueth’, discussed in Chapter Three. However, if Hand 1 is likely readable as a woman, the interplay of gender roles and poetic choice comes to the fore again: Surrey’s choices seem more informed by discomfort, anger, and resentment, while the writer here reads as, if not less, then still differently resistant to the power of the beloved. 77 Nevalainen, ‘Gender Differences’, 46–47. Nevalainen’s research focuses on analysis of signed letters from writers of known genders; her own works offer the best overview of her methods. 78 Nevalainen, 49.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
181
as on 4r, but the use in 3r is also definitely not anomalous for this writer. Hand 8 is quite regular in the writer’s preference for (e)th use. As reported by Nevalainen, the total instances of (e)s use in letters written by persons at court for 1500–1539 is 735, representing a relative frequency of only 1 percent.79 The relative frequency of use by all men in the sample from the same period is 4 percent, while for women it is 29 percent.80 Given that the data will skew towards a higher proportion of courtly women in the sample of all women than courtly men in the sample of all men, the correlation between (e)s use and feminine gender seems fairly strong in courtly writing of the period.81 While a closer study which differentiated by both gender and class for this particular pattern would help strengthen this connection, the available data offer a reasonable suggestion that Hand 1 is that of a woman.82 There are some further distinguishing marks between the two hands which are relevant for the poetic analysis. Hand 1 enters six poems, all on the first six pages of the manuscript, but interrupted by other hands, while Hand 8 enters, astonishingly, over sixty poems, in unbroken sequence from a first entry on 69r to its last on 87v. The evidence of the two hands suggests a greater importance for the poem for Hand 1, simply in terms of placement and proportion compared to total production in the Devonshire. Hand 8 is the neater hand, and its entries are carefully organized on the page in a more or less regular script, maintaining size and ruling with relative consistency. 79 Nevalainen, 54. 80 Nevalainen, 54. 81 Forty-nine percent of the men whose letters were used for the data belonged to the gentry, nobility, or royalty, in comparison to 85 percent of the women (Nevalainen, 46). For the period from 1520–1539, there are perhaps twenty fewer women than men sampled; these women also contribute fewer words per sample than the men (Nevalainen, 46–47). This emphasizes the likelihood of a writer who demonstrates particularly gendered linguistic markers being of the correlated gender, in this instance. 82 Hand 1 also follows a linguistic shift strongly preferred by women on 2r, in the first transcription discussed, where the writer uses ‘yow’ rather than ‘ye’ for a subject, twice, though ‘ye’ is also used twice. Nevalainen finds of ‘you’ in the subject place that the shift ‘was led by women throughout the country’ (45). She found ‘a radical increase in the general frequency of you between 1500 and 1590 from below 20 percent to nearly 100 percent. The 70 percent mark is reached by the female writers in the first half of the sixteenth century, but men’s use of the form only reaches the frequency of 50 percent by the 1560s’ (Nevalainen, 45). However, the use in 2r is less relevant because there is no other copy of the poem in the manuscript to compare and because it is less conclusive. Nevalainen’s results for this shift from 1520–1539 show a 29.2 percent incidence of use among men and a 72.4 percent incidence of use among women. At an internal incidence of use of 50 percent, Hand 1’s use in 2r falls fairly neatly between the two camps.
182
Gender and Position-Taking in Henrician Verse
Hand 1 is fairly neat, as well, though markedly less regular and less carefully ruled; this is particularly true of Hand 1’s transcription of the final four lines of this particular poem, where, in the middle of line nine, the hand nearly doubles in size, while the ink suddenly lightens markedly. Hand 1’s shift in the poem, of course, also provides a potentially alternative explanation for deletion of line ten, in 75v’s version, from the 3r version; perhaps, given the shift into somewhat sloppy handwriting, the transcription is also sloppy, and the writer has simply missed a line. However, the deletion impacts the tone of the poem sufficiently that exploration of alternatives is still worthwhile; further, the hand does not make any other pronounced changes in the final lines, or suddenly skew into corrections or deletions. Given the remaining space on the page, the writer had ample room for such corrections; the choice not to make them may therefore be important. The complete lack of gendering for the speaker or the addressed individual further opens up applications for this poem, and the relative flexibility of the terms used emphasize the potential for love poetry to be used to quite different effect in complaints of the period. While ‘woo’ and ‘desire’ do have some connotations linking them to courtly love traditions, the poem’s general emphasis on service and reciprocal indebtedness makes the terms quite applicable to courtly systems of patronage and service.83 Only the first line make the following exchange explicitly refer to love, and even in that line, the ‘heart’ that is given can easily be understood in terms of fealty and allegiance. The terms of courtly love were borrowed for courtly service, and the terms of courtly service helped define the hierarchy and traditions of courtly love. Hand 1’s version, if the hand does belong to a woman, may be further limited even in her resentments. Because women’s frustrations may be seen as less acceptable or valid by the most powerful men at court, and because her complaints are not offered the same kind of validation in the larger system, her resentment of that system—and of the 83 To the outlines of the distinctive features of women’s writing outlined by Nevalainen we can add her and Helena Raumolin-Brunberg’s information about linguistic change across classes, supplemented by information about class movement in Henry’s court. Both Susanna Hornebout and Levina Teerlinc, for example, were artisan women who gained status through marriage as a direct result of their service to Henry (Frye, Pens and Needles, 78). The two cases together tell us that more typically ‘courtly’ women born to aristocracy shared space with artisan women promoted on the basis of their talents and through marriage and patronage. Circles of courtly women, then, had access to socially aspirant linguistic shifts, to connect the specifics of this milieu to Nevalainen and Raumolin-Brunberg’s work. This overlap of forms used by women and forms used by socially aspirant persons may also be tied to the position-taking energies of the manuscript; as Nevalainen and Raumolin-Brunberg point out, ‘there may be covert prestige in vernacular forms’ (Historical Sociolinguistics, 134).
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
183
lack of acknowledgement from those she renders service—may be inversely, proportionately greater.
B. Responsive Reading and Composition Other poems more clearly inscribed by women in the text indicate similar blurring between love complaints and complaints of service. Heale emphasizes that the ‘formulae of amorous courting offered a storehouse of phrases and a range of subject positions more usually associated with the male speaker, but not necessarily gender specific’.84 Fairly standard complaints of the masculine speaker may gain more radical energy when the complaint is made by someone who is ‘supposed’ to be in an inferior position of service. An example of the change effected by women’s writing can be found in ‘My ywtheffol days ar past’ as written on 68r and 68v by Mary Shelton. The page begins with a series of stylized openings, where a hand distinct from that inscribing the poem begins three times to write ‘Madame’, and after the third attempt continues: ‘Madame margeret et madame de Richemont Ie [v]odroy bien quil fult’.85 Heale argues that, joined with the heading of the two women’s names, the French ‘jottings imply a sense of regret and loss that must have been shared by both the women and their friends’.86 That the hand which enters these names has not even been suggested to be a variant of Shelton’s emphasizes the communal nature of the text; the poem is one of lamentation and loss, and the unknown hand connects Shelton’s laments, ostensibly about service, to the suffering of her friends and fellow ladies-in-waiting. Oddly, very few scholars have given serious consideration to the possibility that Shelton wrote this piece entirely herself, despite the fact that this manuscript represents the earliest known version of a poem which remains unattributed. An initial at the end of the poem has often been read as ‘W’, but may as easily be interpreted as ‘M’ for ‘Mary’, though such initials are as likely to identify a transcriber or even love object as an author. In general, of course, a central tenet of my larger analysis is that authorship is less important than the sort of ownership or appropriation of style which is declared through manuscript revision. However, the case for the early modern writers of the Devonshire is not necessarily the case for the modern 84 Heale, ‘Female Voices’, 14. 85 BL MS Add. 17492, 68r. 86 Heale, ‘Introduction’, in Devonshire, 17.
184
Gender and Position-Taking in Henrician Verse
scholarship. The possibility that it was not Shelton who appropriated the poem by changing ‘she’ to ‘thee’, but rather a later writer or, indeed, Tottel who appropriated the poem by inserting a masculinized voice has important implications for understanding the poem, the manuscript, and Tottel’s revision.87 If Shelton is instead offering an original composition within this communal project, as she appears to do at other points, her voice takes on a larger dimension through the context of such an original. Rather than a poem composed by a single speaker in lamentation, the voice becomes one of a disenfranchised community who finds, in poetry, a rare outlet to express their frustrations, anxieties, and, particularly in this poem, misery. Because it is impossible to say whether Shelton here copies from another complete version or whether that version is her own or someone else’s, the analysis of the process and product must shift. The focus cannot be on alteration or deletion, but only on technique. Uncharacteristically, but not uniquely, Shelton uses lines to underscore stanza breaks throughout the poem. While this decision may add a sense of anchoring or clarity, that sense is immediately, simultaneously, undermined by Shelton’s very difficult spelling. Goldberg has argued that ‘what makes the texts certainly by Mary Shelton in the manuscript particularly elusive and difficult to transcribe is that their spellings allow quite alternative readings’.88 In the context of this larger analysis of the manuscript, this effect is highlighted as a strategy. What we see in Shelton’s writing is not an amateur, either in writing or spelling, but rather a master. Her technique obscures her writing, but in an intentional and often quite effective manner, emphasizing instability and difficulty as themes in her work. This perspective must be restored to women’s writing, to avoid minimizing the ways that such manipulations were not only practiced, but perfected, in manuscript writing that needed to record resistance without inviting destruction from the patriarchal culture in which it needed to survive. Throughout, the poem uses a beginning-of-line repetition similar to that of an earlier transcribed poem from Wyatt, also in her hand. Margaret Douglas is fond of refrains, as reflected in her works examined in the following chapter; Shelton seems fond of this beginning-of-line style. In this particular poem, the repetition of ‘my’ and ‘I se’ asserts a very definite, central identity, countering the self-destructive words of the speaker.89 However, the very placement amongst the manuscript production of the 87 This possibility is engaged in the Social Edition, but not in Heale’s edition. 88 Goldberg, ‘Female Pen’, 27. 89 Shelton, BL MS Add. 17492, 68r–68v, ‘My ywtheffol days’.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
185
Devonshire also implies that this ‘I’ speaks for more than just the single poet, particularly in the context of all these borrowed, manipulated, and re-formed verses.90 Instead, the entry becomes one that invites further use by those who identify with its message of disempowerment. The poem, in this earliest available form, does not include the gendered pronoun use of later versions, and only the conventions of Petrarchan and courtly love poetry mark the work as complaint about a lover. Because those conventions were also in the service of complaints about courtly life, and sometimes even of service to God, Tottel’s version may have created a misinterpretation of a poem not connected to the courtly love tradition in any constricting sense. Rather than being a love lyric that is secretly about court, Shelton’s version here can easily be read as fairly explicitly about court, with only the metaphorical language allowing for a smokescreen of the potential alternative reading. The plural ‘they’ particularly emphasizes a threatening other that is more or greater than a single, scornful lover. Shelton’s ‘the wold me gladly kel’ has a great deal more force than the hyperbole assumed when the subject ‘the’ is a lover; the cruelty of her ‘they’ or ‘thee’ may be real, with real political power behind it.91 This sense is emphasized by the terms of suit implied earlier in the stanza, where the objects ‘wry’ at the speaker’s ‘mon’,92 and ‘wold b gan’ when the speaker is ‘comby’.93 Alleviation of the speaker’s plight is within the object’s gift, as is often the case for the more clearly Petrarchan lover. In this case, though, the page’s header, ‘Madame margeret et madame de Richemont’, slants the poem’s meaning sharply into more political terms.94 The header refers to Margaret Douglas and Mary Fitzroy, the latter the widow of Henry’s son, Henry Fitzroy. As outlined in the introduction, Henry created the law that made Douglas’s tragic first marriage treason only after he discovered the contracted marriage. The plight was entirely the King’s creation, and 90 I would contrast this effect, in the manuscript poetry, with what Solomon calls ‘the vicarious performance of the “lyric I” in print’. In the print context, as she summarizes, ‘the focus shifts from a communal audience to an individual reader, so that the lyric, at the moment of “going public”, ironically becomes most private’ (‘Representing’, 680). 91 Shelton, ‘My ywtheffol days’, line 32. While ‘they’ is not necessarily plural and ‘thee’ is also a possible meaning, the word here does seem to offer some kind of contrast to the singular ‘you’ in lines 31 and 33, making ‘they’ the reading I see as most likely and ‘thee’ a secondary possibility. 92 Shelton, lines 26–27. 93 Shelton, lines 28–29. That is, the person or persons addressed wry (likely in the contemporary sense of cover up, hide, or turn away) at the speaker’s moan and would be gone when she comes by. 94 BL MS Add. 17492, 68r. The editors of the Social Edition particularly observe that the distinction between different forms of names in the header ‘reveal a great deal about gender identity, Renaissance practice, and courtly reality’.
186
Gender and Position-Taking in Henrician Verse
so could certainly have been solved by him. Similarly, Fitzroy found herself in an untenable position when trying to claim the properties due her after her husband’s death. Her case could easily have been settled by Henry’s intervention, particularly since her husband had been his acknowledged son, yet her suit was not settled until the summer of 1538, a full two years after Henry Fitzroy’s death.95 The heading on 68r strengthens the connection between the complaint of the poem and the complaints of each woman during the decade of the most work in the manuscript. The writer lodges a protest by leaving off Douglas’s surname, pointing to her complicated status as both widow and, by the terms that made her original marriage illegal, never married. She becomes simply ‘Margaret’, referred to neither by her original surname nor by the surname of the husband that has been denied to her. This omission then emphasizes Fitzroy’s title and her connection to her deceased husband, thus also emphasizing the legal grounds for her claim to the disputed property. The single contribution Fitzroy herself most certainly makes to the manuscript, on 55r–v, seems also to relate to her difficulties at the English court and in her marriage. While Fitzroy’s hand has only rarely—and to my mind, unconvincingly, given Baron’s extensive work with her hand in letters—been disputed,96 the authority of her poem as a version of her brother’s, Henry Howard, Earl of Surrey’s, ‘O happy dames’ has been undermined by her frequent corrections and difficult spellings. Baron uses her comparison of the hand and style of the poem to Fitzroy’s letters to point out that the earlier ‘suggestion that Mary Fitzroy’s frequent deletions in D show an attempt at a literratim transcription of her original, including its (supposed) redundant flourishes resembling the terminal es abbreviation […] is undermined […] by […] Mary Fitzroy’s frequent habit of erroneously adding es to words’.97 In addition to these unnecessary endings, usually uncorrected, Fitzroy ‘deletes many words and then rewrites them merely with a different spelling’.98 While some have seen these marks in the poem particularly as characteristic of laborious transcription, Baron’s analysis of Fitzroy’s letters reveals the limitations of these claims: 95 Baron, ‘Fitzroy’s Hand’, 318. Beverly Murphy indicates that the difficulty was entirely based in Henry’s resistance to settling Mary’s agreed upon jointure and connects this difficult directly to the Howards’ larger fall from favor, both because of the execution of Anne Boleyn and because of the Douglas–Howard marriage (Bastard Prince). 96 For an overview of the history of misattributions of Fitzroy’s hand, see Baron (‘Fitzroy’s Hand’, 324). 97 Baron, 320. 98 Baron, 320.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
187
If, therefore, these characteristics are interpreted as a quasi-scholarly effort to transcribe accurately letter by letter a difficult original, the deletions and rewriting in her correspondence must also be interpreted as attempts to copy someone’s difficult original […] it is far more likely that she was habitually uncertain and anxious about her writing in general and spelling in particular.99
Baron convincingly argues, then, that Fitzroy’s writing, whether original or copied, is likely to show marks of concern with getting the ‘right’ spelling. I want to add to Baron’s analysis a different perspective on what Fitzroy would have considered ‘right’ and on why this concern is so manifest in her writing. Baron concludes of her analysis of Fitzroy’s copy of her brother’s work that ‘the unfortunate consequence of using D and not T as a copy-text is that Mary Fitzroy’s unusual spellings render the poem more obscure, despite her many superior readings’.100 Taken within the larger context and project of the manuscript, though, the obscuring effect of Fitzroy’s spelling choices emerges as a valued strategy of composition. In his analysis of the poem, which he ascribes to Shelton rather than Fitzroy, Goldberg makes the useful point that ‘these crossings out and substitutions might […] indicate her rethinking whether or not the word she just wrote is the word she wants, and then deciding that it is—or, perhaps, deciding that the respelling says better what she wants’.101 Although the ‘she’ shifts, the point still holds, and applies to Fitzroy’s letters as well as it does here. Trying to contain or carefully orient meaning, indeed, might be particularly likely to result in the sort of anxiety that Baron sees as surrounding Fitzroy’s writing in the manuscript and in her letters. Fitzroy’s desire to carefully control her meaning results from her social context: a woman in a contested position in a surveilling culture. She is, and must stay, highly alert to the importance of appropriate performance carefully conveyed, but she is also sensitive to the opportunities for resistance and for expression of frustration which may be open to her if she proceeds with sufficient caution. Her writing choices reveal her literacy in the conventions of manuscript circulation and verse position-taking. These literacies inform the spellings which Fitzroy does choose, which are often, as Baron notes, atypical. One such choice is ‘woffulle’ for woeful in third line. In eschewing the ‘oo’ spelling of ‘woe’ which is so common throughout the manuscript, particularly in Shelton’s transcriptions, Fitzroy 99 Baron, 320. 100 Baron, 323. 101 Goldberg, ‘Female Pen’, 27.
188
Gender and Position-Taking in Henrician Verse
erases the connotative link to ‘woo’.102 Rather, the double ‘ff’ and double ‘ll’ give the emphasis to the second syllable, ‘ffulle’, thus underscoring the enormity of the emotion which overwhelms the speaker. Her choices in ‘mowernenge’ (mourning) and ‘mowren’ (mourn) in lines seven and twenty-one both evoke contemporary spellings of ‘morn’, which included ‘morowen’ or ‘morwen’.103 Time is thus emphasized through her spelling choices, emphasizing the temporal nature of the manuscript as subject to changes caused by altering and commenting hands. Thematically, Fitzroy signals her awareness of the larger project in which her transcription is now participating. The second use emphasizes the lonely waking of the speaker, as the conclusion of the stanza: ‘wherewith I wake with hes his retoreneretourne / whoosse […] abssente fflame [do]otht make me boren / bwt whan I ffynde the lak[e] lorde howe I mowren’.104 Fitzroy’s ‘boren’ for ‘burn’ has the further connotative effect of figuring the flame of the beloved as a giver of life, reinforcing the gendered roles of the speaker and object. The poem’s premise is based in those roles—men go to war, while women lament at home. Fitzoy’s choice of spelling, within the gendered context of the poem, may both reinforce the agency of the masculine speaker as the instigator of a reproductive act and emphasize the powerlessness of the feminine speaker, giving even her potential biological reproductive power over to the absent object of the poem. However, in the assumption of a feminine speaker, Fitzroy finds an outlet for frustration at this lack of agency. She also takes on a poem by her brother and appropriates the verse for her own use, thus demonstrating a flexibility in the bemoaned lack of agency. The prominence of women’s writing in the Devonshire allows Fitzroy the chance to re-appropriate her famous brother’s appropriation of the feminine voice, suggesting that even the most conventional performances of gendered roles may still contain hidden room for position-taking and for women’s control. Placing Fitzroy’s verse alongside Shelton’s offers an extended understanding of her spelling work, as both women’s poems demonstrate how specific choices in word formation could activate or eschew particular meanings. A similar thread is emphasized in comparing poems that occur multiple times in the Devonshire, as in the 3r and 75v versions of ‘My harte/herte I gave’. While some changes in different versions of a verse may be ‘errors’, others seem more clearly to be choices—and even errors create a new, readable 102 Fitzroy, BL MS Add. 17492, 55r–v, ‘O happy dames’, line 3. 103 Fitzroy, lines 7 and 21. 104 Fitzroy, lines 19–21.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
189
form of a text. In reconsidering manuscript poetry and its influences on early modern print and on later canon formation, our engagements here reveal the rewards of interrogating multiple versions and revisions. Such interrogations allow us to consider what each did for its readers and why each was considered valuable as a communally-created, context-embedded text for the original manuscript circulators, even if we are cut off from a full appreciation of their contexts and so of the full potential interpretations.
C. Forms and Dialogues Thus far I have focused my analysis primarily on the shape and texture manuscript writing creates within individual poems, adding to my analysis some of the recent work which may allow us to trace gendered markers in otherwise unidentified hands; I want now to turn fuller attention to the technique of grouping as used in manuscript creation, particularly as it reinforces the use of manuscripts to preserve communal grievances and modes of resistance made possible through communal writing. In a point that encompasses essentially all of the major strategies of the manuscript transcriber, Marotti establishes that ‘In the system of manuscript transmission, it was normal for lyrics to elicit revisions, corrections, supplements, and answers, for they were part of an ongoing social discourse. In this environment texts were inherently malleable, escaping authorial control to enter a social world in which recipients both consciously and unconsciously altered what they received’.105 These verse groups contain lyrics that are not quite supplement and not quite answer, but that nonetheless emphasize the alteration and revision of meaning inherent in the practice of placement throughout these social compositions. As Burke articulates, ‘[t]he search for coherence in miscellaneity may lead us to particular texts that can be read in multiple ways, in dialogue with the texts around them’.106 The analysis of contextual importance will be extended in the final chapter, where I examine the epistolary verse in Devonshire, but here we can begin analyzing what poetry does in its immediate physical context. The group here under analysis includes two apparent ‘fragments’, on 22v, as well as the ‘complete’ versions of the poems from which those fragments are pulled, which cover 23r through the top of 24v. Interrupting these dual versions of poems, as the bottom entry on 22v, is a short, four-line composition in 105 Marotti, English Renaissance Lyric, 135. 106 Burke, ‘Materiality and Form’, 220.
190
Gender and Position-Taking in Henrician Verse
Shelton’s hand and concluded with her signature, generally treated as an original composition. The selection, here treated as a deliberate group, then concludes with one additional, longer poem on 24v and 25r, followed by a substantial blank on 25r, with 25v left completely blank. The section begins with the first three lines of ‘The knot which fyrst my hart dyd strayn’ and the first two words of the fourth line, entered by Hand 3, who also completes the poem on the later pages.107 The selection is annotated above and below by Hand 10, including ‘fynys quod Iohn’ which follows the clearly unf inished partial quatrain.108 Hand 10 then transcribes the second poem on 22v: the first seven lines of a version of ‘He Robyn gentyll robyn’, a particularly elastic poem which frequently recurs in the period.109 The version of the opening lines transcribed here by Hand 10 is distinct from the version entered two pages later by Hand 3, where the first line is ‘Hey Robyn Ioly Robyn tell me’.110 The ending note ‘fynys quod Iohn’ after the first, though, and the concluding line after the second, seem attempts to codify these interruptions, making the cut-offs appear intentional rather than accidental or disrupted. Even if the original copyist was, in fact, simply interrupted, the annotating Hand 10 then marks the poem as though this abbreviated version could be treated as a finished quotation.111 This is a particularly important annotation in a manuscript that pays little attention to delineation as a general rule; possibly, the annotation can be taken as an attempt to further separate the hands on the page, distinctly marking where one writer leaves off and another takes up. In any case, the potential attribution to ‘Iohn’ does not necessarily challenge the accepted attribution to Wyatt; Hand 10 may be identifying, among other possibilities, the hand of the poem’s copyist, rather than the original author of the poem. This possibility is complicated, though not negated, by the ‘W’ with which Hand 10 marks the conclusion of the seven included lines of ‘He Robyn’.112 If the ‘W’ at the end of the second poem is understood as a potential attribution, then the ‘Iohn’ entered by 107 BL MS. Add. 17492, 22v, ‘The knot which fyrst’. 108 BL MS Add. 17492, 22v. For a different, though primarily not contradictory, reading of these entries, see Powell, ‘Marginalia’, 9–10. Powell offers distinct hand attributions in this section (10). 109 BL MS Add. 17492, 22v, ‘He Robyn gentyll’. 110 BL MS Add. 17492, 24r, ‘Hey Robyn Ioly’. 111 Notes like these occur regularly throughout the manuscript. My point here is not that this mark is unusual, but that its use here, after a fragment that is easily identified as a fragment (given the fuller version of the poem only a page later) suggests that the manuscript’s original users saw the fragment as useful on its own terms. 112 BL MS Add. 17492, 22v.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
191
the same hand on the preceding poem may also mean to function as an identification of the author. However, Wyatt was not the only ‘W’ at court, and the initial may still refer to the copyist—or, it may refer to Wyatt as a subject referent.113 The consistent and definite feature of these marks on the page, though, is that they function to add a sense of finality to segments that would otherwise be more clearly ‘unfinished’. These short sections affect the reader’s understanding of, and expectations for, the four lines Shelton enters at the bottom of 22v. Primarily, the knowledge that these entries are excerpted from longer works may also create an expectation that a longer version of ‘A wel’ exists. The format follows the other two entries: a few lines of verse followed by an identifying note, in this case a signature generally agreed to be the writer’s and the original author’s.114 However, no longer version of ‘A wel’ exists anywhere in the manuscript, much less in this short grouping. The reader is engaged in the process of anticipating a longer version which is then denied, in a rhetorical move echoing the tropes of the short poem. Moreover, this denial influences the reception of the short verses earlier on the page. If Shelton’s verse is presented as independent and original, the reader may be encouraged to reinterpret the other verses independently, as their own versions of poetry. This version of ‘He Robyn’ offers some unique interpretations if viewed as its own poem, as the lady’s response is cut off from the last line, and possibly reinterpreted as the identifying mark that concludes the poem: ‘and yet she wyll saye W’.115 The woman no longer becomes indistinctly resistant; instead, she rejects the speaker in favor of another, specific lover. The object may then become an agent, offering resistance to the usual Petrarchan narrative in which even the beloved’s rejection is spoken of from the man’s perspective. Instead, the verse ends with her voice, and with her own utterance of choice. Just as the short poem by Shelton encourages independent consideration of the ‘incomplete’ versions that share that space, the longer versions of ‘Hey Robyn’ and ‘The knot which fyrst my hart dyd strayn’, coming in the same order as the two short selections from the poems, reinforce the expectations 113 Powell does argue that ‘the ascription is enough to raise doubt about Wyatt’s authorship of this poem’ and that, more generally, ‘this point in the manuscript is […] characterized by a sort of dialogic familiarity of entry that undermines our attempts to draw distinct conclusions about authorship’ (‘Marginalia’, 10). 114 Shelton, BL MS Add. 17492, 22v, ‘A wel’. Powell similarly addresses the possibility that Shelton perhaps leaves her signature ‘as evidence of her authorship’ (‘Marginalia’, 10); the editors of the Social Edition also acknowledge that this ‘could be an original creation’. 115 22v, ‘‘He Robyn gentyll’, line 7. Heale reads an ‘N’ in the final line (Devonshire, 92), but the editors of the Social Edition also read a ‘W’.
192
Gender and Position-Taking in Henrician Verse
that the next selection, which begins on 24v, will be the longer version of the third poem on 22r. Instead, an entirely new poem is entered, which continues onto 25r, though the fragments on both pages seem capable of standing alone.116 This poem which disrupts our expectation of a longer version of ‘A wel’ is itself a particularly interesting commentary on loss and power: ‘It was my choyse It Was my chaunce’ which is, furthermore, another excerpt. The thematic function of this longer poem, as related to Shelton’s entry, may then be considered in terms of the relationship between the short and long entries of ‘He Robyn’ and ‘The knot which fyrst my hart dyd strayn’. The shorter version of ‘The knot which fyrst my hart dyd strayn’ runs as follows: The knot which fyrst my hart dyd strayn Whan that your sarwant I becam doth bynd me styll for to Remain all wais117
The longer version adds a further 38 lines to this, beginning with the conclusion of the first stanza, after an altered spelling of ‘ways’: all was yor owne as now I am and if you fynd that I do fayne with Iust Iugement my self I dam ene To haue Dysdain118
There are relatively few changed spellings in the first four lines of the two poems; the exceptions are ‘dyd’/’did’ in line one and perhaps ‘sarwant’/’saruant’ in line two, though the strokes of ‘u’ and ‘w’ can be difficult to distinguish conclusively in the hand. The only alteration that immediately opens up interpretative difference is ‘wais’/’was’. In the second spelling, the potential verb use opens up: ‘all was your own’. The spelling in the first poem limits the meaning and so shuts down the insisted continuance of the line implied by the verb use: ‘doth bind me still for to remain / always’. The limited interpretation of ‘wais’ enforces the sense of the excerpt as complete; the reader knows the quatrain is not meant to stand alone based only on familiarity 116 BL MS Add. 17492, 24v–25r, ‘It was my choyse’. 117 22v, ‘The knot which fyrst’. 118 BL MS Add. 17492, 23r–23v, ‘‘The knot which fyrst’, lines 4–7.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
193
with the poem in other contexts or through gaining that knowledge upon reaching the longer version two pages later. In comparison to the first version, the longer work reinforces the theme of service, but also makes the terms of that service more explicit—the writer reinforces their fealty, but also takes on any judgment of that fealty as their own. The speaker determines the terms of the punishment, even if that punishment is to be for a failure of service to another. ‘He Robyn’ shows more difference between the two versions, no doubt partly because they are inscribed by two different hands. The shorter version, in its entirety, reads He Robyn gentyll robyn tell me howe thy lady dothe and thou shalte knowe of myn My ladye is vnkynde perdye allas why is she soo She loves another Beter then I and yet she wyll saye W119
The longer version alters the first stanza to two lines split between ‘me’ and ‘how’, with an odd, off-set placement of ‘of myn’ slightly above the rule of the second line.120 The poem goes on to add another twenty lines, organized in five quatrains, starting with an addition to line seven of the 22r version (line six of the version on 24r): where 22r cuts off and inserts the ambiguous ‘W’ mark, 24r concludes with the woman telling the lover ‘noo’. These shorter selections have at minimum a spatial relationship to Shelton’s four lines: A wel I hawe at other lost not as my nowen I do protest bot wan I hawe got that I hawe mest I shal regoys among the rest 121
The implicit pairing of this selection with the longer ‘It was my choyse It Was my chaunce’ offers an emphasis on agency in this shorter verse, 119 22v, ‘He Robyn gentyll’. 120 24r, ‘Hey Robyn Ioly’, line 2. 121 Shelton, 22r, ‘A wel’.
194
Gender and Position-Taking in Henrician Verse
potentially consistent with the resistant woman of ‘He Robyn’. The insistence on adding some sense of conclusion to the excerpts, moreover, encourages an understanding of all three short lyrics—that is, the shorter versions of ‘The knot which fyrst my hart dyd strayn’ and of ‘He Robyn gentyll robyn’, along with Shelton’s ‘A wel’—as independent but related. On its own, the theme of the first poem is devoted love or service, without end; the longer version then unusually emphasizes the speaker’s choice and control within this service. The second stands less firmly alone without the concluding initial taken as part of the poem, but it still sets up the premise of a dialogue in which a lover complains of the unkindness of a lady whose response to the speaker’s charges is cut off. The last poem, composed and signed by a woman, speaks of general loss and of potential restoration—but in language marked by suggestions of response. These three short poems, taken together, may be interpreted as a dialogue which interrupts the lovers’ complaints with the voice of the lady, who protests the misinterpretation of her actions and interest and who emphasizes the importance of her choice in the matter. The placement of ‘A wel I hawe at other lost’ encourages a dialogue with both preceding poems. To the lover’s traditional complaint of unrewarded but inescapable service, Shelton’s entry responds that she has the same experience: the same complaints, the same woe, and the same potential for success. The speaker of ‘A wel I hawe at other lost’ both insists on her place within a larger, communal experience and interrupts the potentially othering voice of the two traditional love lyrics—or, at least, of ‘He Robyn’—that might otherwise assign the woman an oppositional place. Although offering a less distinct sense of grouping, a set of poems on 65r and 65v show similar work in combining excerpt and potentially complete verse and present an opportunity of comparing Shelton and Margaret Douglas’s work together on a smaller scale, if in some sense the whole manuscript can be understood as a project of their larger collaboration. The poem Douglas transcribes on 65r is a three-stanza version of a poem she also writes on 58v and 59r in a four-stanza form, ‘My hart ys set nat to remowe’, on the general theme of constancy and devotion. As with the long and short versions of the poems discussed above, the repetition suggests that the placement and form of each is deliberate. Shelton then writes a short poem on 65r, ‘I ame not she be prowess off syt’, on the theme of failure to pretend emotions one does not have or hide those one is feeling. Douglas again takes over on 65v with an excerpt from a poem attributed to Wyatt, ‘Myght I as well within my song be lay’, which speaks particularly to bitter
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
195
sorrow.122 On the same page, Shelton then writes a poem on the utility of pretending happiness in the midst of sorrow, ‘To cowntarffete a mery mode’, a sharp counter to her own copy from the previous page and to Douglas’s lament just above. The group together forms a sort of narrative emphasizing the ultimately tragic fallout from the events to which Douglas may be responding in 65r. The poem clearly had a particular appeal to Douglas, given her multiple versions. In it, Douglas returns to the themes that mark the entries of the Douglas-Howard epistolary exchange. She avows, ‘I lowe ffaythffully’,123 and that, in return, her lover ‘welnot slake hes lowe / nor never chang hes ffantecy’.124 The verse also speaks to separation, though, affirming that ‘who ffeleth greve so yt hym hes’;125 the final stanza shares that the speaker has been ‘banysht hym ffro / hys speket hes syght and compa[n]y’.126 Regardless of the time of composition, Douglas’s contemporaries would have likely read the poem, transcribed by her, as a reference to the Douglas-Howard affair. It follows that these poems on grief, sorrow, and performance would have seemed to them to comment on these events, as well. Further, the poems together function as a kind of conversation: Douglas asserts her loyalty, Shelton pontificates on the necessity of showing one’s emotions, Douglas asserts her own desperate sadness, and Shelton changes her tune, arguing that perhaps, in some circumstances, hiding one’s emotions is better. Finally, Shelton writes a self-deprecating couplet after her poem, reading ‘ryme dogrel how many / myle to meghelmes’.127 Shelton’s argument for hiding one’s true emotions, alongside her disavowal of the worth of her own writing, emphasizes the warning tone that undergirds her work in these selections. After comparing a smiling face over a sad mood to a hood in the rain, Shelton’s poem on 65v continues ‘well the war wet that bar hed shod stod’ and concludes ‘betar a path than a halle owte’.128 If their contemporaries would have likely understood Douglas’s laments as relating to her personal tragedies—particularly her first husband’s 122 Three lines of ‘Myght I as well within my songe’ are entered on the next page, as well (66r). The lines are quite badly smudged on the 66r copy. The placement—and perhaps attempted obscuring—of these three lines offer several intriguing interpretive possibilities. I am, particularly, drawn to the possibility that the lines’ erasure emphasizes and prioritizes the placement of Douglas’s version in dialogue with Shelton. 123 Douglas, MS BL Add. 17492, 65r, ‘My hart ys set nat’, line 2. 124 Douglas, lines 3–4. 125 Douglas, line 7. 126 Douglas, lines 9–10. 127 Shelton, BL MS Add. 17492, 65v. 128 Shelton, MS BL Add. 17492, 65v, ‘To cowntarffete’, lines 4 and 7.
196
Gender and Position-Taking in Henrician Verse
death—then Shelton seems to advise that, in the current atmosphere, counterfeited compliance is the best outward show. In acknowledging such show, the poem elides the possibility of changing the inner emotion. Douglas chooses to end her excerpt from ‘Myght I as well within my song’ with the threat of ‘cryes remors and grow[ges]’.129 In the context of her friend’s anger, quite possibly against Henry, Shelton’s short poem suggests a show of compliance, rather than a return to loyalty or a true submission of will. Her final, dismissive note on her own poem, then, may be meant to create distance or to undermine the potential energy of the poetry. That distance may be especially necessary because the exchange between Douglas and Shelton offers such a subversive reading of Douglas’s public acquiescence to her kingly uncle. The back-and-forth of Shelton and Douglas, as one affirms absolute loyalty to a lover while one suggests performative restraint, neatly echoes the exchanges of the DouglasHoward epistolary verses, discussed in the next chapter. On the one hand, as shown in the analysis of those verses, loyalty and faithful love to each other were partly key to Douglas’s and Howard’s best hopes of defense. Nonetheless, to avow faithful and undying love in spite of the King’s displeasure was to express resistance. Moreover, Shelton’s f irst entry offers a stark vision of the consistency of the (woman) writer, insisting that she is not one who ‘kan make a y[og]y off al my woo’130 or ‘cloke my greffe’.131 When Shelton closes the exchange with the very different insistence that success actually depends on exactly such cloaking, that difference draws attention to the distinction between internal grief and necessary performances denying that grief, very much like the balance Douglas—and, indeed, any aggrieved courtier—would have needed to strike to survive the King’s wrath. That emphasis on performative disavowal of one’s emotions encourages a reading of Shelton’s final critical remark on her own entries as itself an action of protective cloaking; in exchange, the act of cloaking itself emphasizes the energies of the exchange between the two women. Examining the strategies used by the courtiers who produced the Devonshire allows a greater insight into poetic composition of the period as a mode of communal position-taking. While new research may allow us greater insight into identification of the hands used in the work, the original creators largely focused on the flexibility of ownership and authorship. 129 Douglas, MS BL Add. 17492, 65v, ‘Myght I as well’, line 4. 130 Shelton, MS BL Add. 17492, 65r, ‘I ame not she’, line 2. 131 Shelton, line 8.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
197
This flexibility extended to their ideas of what constituted poetry, as well as how spelling or punctuation could be used in their writing. Through obscuration and multiplication of meaning and authorship, poets like Shelton and Douglas, along with the other writers in the Devonshire, created and protected a community defined by coded, verse position-taking. As a reaction to the strictures of the late Henrician court, writers engaged in a kind of poetic composition designed to preserve their collective voices. The work of manuscript writers reflects how central women were to these compositional practices and indicates the particular energies added by their own diff icult performances at court. The careful negotiation of resistance and restraint exemplified in the works in Devonshire indicates that these practices, central to poetic production in the Henrician court, were largely formed by the perspectives and talents of courtly women. Understanding these compositional strategies as more typical of and foundational to courtly practice than individualized authorship enables the re-readings of Wyatt, Surrey, and Skelton engaged earlier in this work; such re-readings, then, help us recalibrate our field through a lens that better incorporates a more accurately gendered and power-inflected Henrician canon.
Works Cited Baron, Helen. ‘Mary (Howard) Fitzroy’s Hand in the Devonshire Manuscript’. The Review of English Studies 45, no. 179 (1994): 318–335. Beal, Peter. In Praise of Scribes: Manuscripts and their Makers in Seventeenth-Century England. Oxford: Clarendon, 1998. Burke, Victoria E. ‘Materiality and Form in the Seventeenth-Century Miscellanies of Anne Southwell, Elizabeth Hastings, and Jane Truesdale’. In English Manuscript Studies 1100–1700, Vol. 16: Manuscript Miscellanies c. 1450–1700, edited by Peter Beal and A.S.G. Edwards, 219–241. London: British Library, 2011. Burke, Victoria E. and Jonathan Gibson, eds. Early Modern Women’s Manuscript Writing. Burlington, VT: Ashgate, 2004. Burke, Victoria E. and Jonathan Gibson. ‘Introduction’. In Burke and Gibson, Manuscript Writing, 1–8. Carlson, David R. ‘The Henrician Courtier Writing in Manuscript and Print: Wyatt, Surrey, Bryan, and Others’. In A Companion to Tudor Literature, edited by Kent Cartwright, 151–177. Malden, MA: Wiley-Blackwell, 2010. Castiglione, Baldesar. The Courtier. Translated by George Bull. New York: Penguin, 2003.
198
Gender and Position-Taking in Henrician Verse
Clarke, Elizabeth. ‘Women’s Manuscript Miscellanies in Early Modern England’. In Teaching Tudor and Stuart Women Writers, edited by Susanne Woods and Margaret P. Hannay, 52–60. New York: MLA, 2000. The Devonshire Manuscript. 1525?–1567? MS BL Add. 17492. British Lib., London. Dolven, Jeff. ‘Reading Wyatt for the Style’. Modern Philology 105, no. 1 (2007): 65–86. Donawerth, Jane. ‘Women’s Poetry and the Tudor-Stuart System of Gift Exchange’. In Women, Writing, and the Reproduction of Culture in Tudor and Stuart Britain, edited by Mary Burke, Jane Donawerth, Linda L. Dove, & Karen Nelson, 3–18. Syracuse, NY: Syracuse UP, 2000. Ezell, Margaret J.M. Social Authorship and the Advent of Print. Baltimore: Johns Hopkins UP, 1999. Frye, Susan. Pens and Needles: Women’s Textualities in Early Modern England. Philadelphia: U of Penn P, 2010. Goldberg, Jonathan. ‘The Female Pen: Writing as a Woman’. In Language Machines: Technologies of Literary and Cultural Production, edited by Jeffrey Masten, Peter Stallybrass, and Nancy J. Vickers, 17–38. New York: Routledge, 1997. Greenblatt, Stephen. Renaissance Self-Fashioning. Chicago: U of Chicago P, 1980. Heale, Elizabeth. ‘“Desiring Women Writing”: Female Voices and Courtly “Balets” in some Early Tudor Manuscript Albums’. In Burke and Gibson, Manuscript Writing, 8–31. Heale, Elizabeth, ed. The Devonshire Manuscript: A Women’s Book of Courtly Poetry. Toronto: Center for Reformation and Renaissance Studies, 2012. Heale, Elizabeth. ‘Women and the Courtly Love Lyric: The Devonshire MS (BL Additional 17492)’. The Modern Language Review 90, no. 2 (1995): 296–313. Irish, Bradley J. ‘Gender and Politics in the Henrician Court: The Douglas-Howard Lyrics in the Devonshire Manuscript (BL Add 17492)’. Renaissance Quarterly 64 (2011): 79–114. Marotti, Arthur F. Manuscript, Print, and the English Renaissance Lyric. Ithaca: Cornell UP, 1995. Murphy, Beverly. Bastard Prince: Henry VIII’s Lost Son. Gloucestershire: The History Press, 2004. Nevalainen, Terttu. ‘Gender Differences in the Evolution of Standard English: Evidence from the Corpus of Early English Correspondence’. Journal of English Linguistics 28, no. 1 (2000): 38–59. Nevalainen, Terttu and Helena Raumolin-Brunberg. Historical Sociolinguistics: Language Change in Tudor and Stuart England. New York: Pearson, 2003. North, Marcy L. The Anonymous Renaissance: Cultures of Discretion in Tudor-Stuart England. Chicago: U of Chicago P, 2003. Phillippy, Patricia. ‘Introduction: Sparking Multiplicity’. In A History of Early Modern Women’s Writing. Ed. Patricia Phillippy. Cambridge: Cambridge UP. 1–24.
Tr anscription as Tr ansl ation: Writing the L anguage of Manuscript Poe try
199
Powell, Jason. ‘Marginalia, Authorship, and Editing in the Manuscripts of Thomas Wyatt’s Verse’. In English Manuscript Studies 1100–1700, Vol. 15: Tudor Manuscripts 1485–1603, edited by Peter Beal and A.S.G. Edwards, 1–40. London: British Library, 2009. Shirley, Christopher. ‘The Devonshire Manuscript: Reading Gender in the Henrician Court’. English Literary Renaissance 45, no. 1 (2015): 32–59. Solomon, Deborah. ‘Representations of Lyric Intimacy in Manuscript and Print Versions of Wyatt’s “They flee from me”’. Modern Philology 111, no. 4 (2014): 668–682. Stevenson, Jane and Peter Davidson. Early Modern Women Poets (1520–1700): An Anthology. Oxford: Oxford UP, 2001. A Social Edition of the Devonshire Manuscript MS (BL Add. MS 17492). Wikibooks, The Free Textbook Project. https://en.wikibooks.org/w/index.php?title=The_Devonshire_ Manuscript& oldid= 3469218. Accessed 1 July 2022. White, Micheline. ‘Women Writers and Literary-Religious Circles in the Elizabethan West Country: Anne Dowriche, Anne Lock Prowse, Anne Lock Moyle, Ursula Fulford, and Elizabeth Rous’. Modern Philology 103, no. 2 (2005): 187–214. Wyatt, Thomas, Sir. ‘My hert I gave the not to do it payn’. In Collected Poems of Sir Thomas Wyatt, edited by Patricia Thomson and Kenneth Muir, 13. Liverpool: Liverpool UP, 1969.
6. Resistance and Unity in the DouglasHoward Exchange Abstract: This chapter considers strategies of transcription and dialogue within manuscript creation, focusing on the exchange between Margaret Douglas and Thomas Howard in the Devonshire alongside other sections of the manuscript preserving exchanges wherein writers respond to and riff on one another’s works. The widespread use of such strategies emphasizes the literacies involved in the use of poetry at court; those particular literacies must be established for position-taking verse to circulate successfully via context-dependent codes. The established context of and comparison to translation and tradition allows for greater insight into exactly what transcription offered its authors, while also structuring a framework for considering how and why these three strategies intersected in courtiers’ versified position-taking. Keywords: Henrician poetry; Tudor verse transcription; Margaret Douglas; Mary Shelton; Devonshire Manuscript; early modern women’s writing
The previous chapter establishes some of the practices used, particularly by women writers, to express and preserve position-taking verse in the Devonshire Manuscript; earlier sections have established the broader system of which this position-taking was a part. Now, we return to the exchange and the incident with which this work opens: the Douglas-Howard exchange. This incident provides a distillation of the paranoia and anxiety caused by the shifts that occurred under Henry VIII’s reign: the increasing centralization of power and the increasing surveillance associated with the drive to protect that power. In response, both Douglas and Howard, or at least the voices apparently assigned to them in the Devonshire, use elements of the same performances imposed on them by this surveillance culture to express their resentment and resistance. Using the same gender
Quoss-Moore, R.M., Gender and Position-Taking in Henrician Verse: Tradition, Translation, and Transcription. Amsterdam: Amsterdam University Press, 2023 doi 10.5117/9789463723534_ch06
202
Gender and Position-Taking in Henrician Verse
and hierarchical roles that Henry himself had partially defined, Douglas and Howard define the legitimacy of their relationship and express the necessity of its primacy and privilege even in the face of official opposition. Using the rules of courtly love, Douglas and Howard both lay claim to an older and earlier system of fidelity. As David R. Carlson summarizes, Henrician court poetry ‘concerns itself with what it meant at the moment to be a courtier’ and with, more specifically, ‘the loss of the fundamental aristocratic prerogative’.1 Using the terms of service aligned both with that tradition and with Henry’s court, Douglas and Howard paradoxically argue for their rights within the system that, in their portrayal, has betrayed and incarcerated them. Three different entries, at minimum, seem intended to memorialize the events of the Douglas-Howard exchange and its aftermath in the Devonshire. The first of these, according to scholarly consensus, records an epistolarypoetic exchange between Douglas and Howard during their imprisonment.2 This exchange is recorded from 26r to 30r. Much later in the text, on 88r, Douglas records a single poem, ‘Now that ye be assemblled heer’, which seems to dramatize her own vows of fealty either during or immediately after her and Howard’s imprisonments; because the poem figures the speaker’s death as a result of the loss of her lover, the poem’s dramatic situation may suggest a moment after Howard’s death in the Tower.3 Finally, the same hand which records the epistolary-poetic exchange between Howard and Douglas records the last entries in the manuscript, organizationally, which begin after a few blank pages following Douglas’s composition. Following Bradley J. Irish, Elizabeth Heale, and Paul G. Remley, among others, I believe that these final entries can also be seen as intentionally related to the Douglas-Howard affair. This very neat hand is mainly in these two sets of entries, 4 and, in both entries, draws from the 1532 edition of Thynne mentioned in the introduction.5 1 Carlson, ‘Henrician Courtier’, 152–153. 2 The agreement is essentially universal. See Harrier, ‘Printed Source’; Heale, ed., Devonshire; Irish, ‘Gender and Politics’; Remley, ‘Mary Shelton’; and Southall, Courtly Maker. 3 Heale instead suggests that the poem ‘may have been originally composed in the Tower of London’, though not directly into the Devonshire (Devonshire, 246). 4 Heale sees this hand also in a short stanza on 59r and in at least two instances of ‘the half-line “I am yowrs an\” thought by Southall and Foxwell to be a response by Anne Boleyn to a supposed poem by Wyatt, but more likely to be an echo of the conventional expression, “I am yours and will be sure”’ (‘Love Lyric’, 310). 5 The Social Edition offers extensive notes on the use of the Thynne; one overview can be found in the commentary for the couplet which begins ‘And now my pen’ on 29v.
Resistance and Unit y in the Dougl as-Howard Exchange
203
The poems are entered by the hand TH2, which, as the initials suggest, has at times been thought to be one of the hands of Thomas Howard.6 More recent scholarship has avoided this particular attribution, though some critics have left the possibility at least partially open. Irish, for example, avoids explicitly endorsing this assignment, saying that ‘because no example of Howard’s hand has been discovered, we cannot know […] if he inscribed the poems personally while imprisoned in the Tower’.7 Much of Irish’s argument about the poems proceeds from the idea that indeed Howard did record these works; he acknowledges, though, that ‘it is possible that the poems were collected after Howard’s death, and entered by an unknown scribal hand, as tribute and memorial to Lady Margaret’s former lover’.8 Other recent scholarship has more clearly challenged the attribution to Howard, most notably in Remley’s work, though Molly Murray also questions the assignment. Murray’s suggestion is simply that some other courtier entered the poems after Howard’s death; Remley more specifically proposes Mary Shelton as the transcriber of the poems assigned to TH2.9 Though she avoids a full endorsement of any single candidate, Heale proposes a more comprehensive vision of the contributions made by this hand—including work on several other leaves, following Raymond Southall’s similarly broad reading of the hand.10 In fact, Southall’s was one of the earliest extensive analyses of the manuscript to assign this work to Mary Shelton, assigning this exchange, among dozens of others, to the poet. Harrier, however, challenged this analysis in his subsequent research, and his ‘diminution of Shelton’s contribution continues to exert an influence on critical opinion’.11 The hand is notably distinct from the hand most definitely assigned to Shelton; however, as Remley convincingly argues, ‘Shelton’s signatures show incontrovertibly that her handwriting was not restricted to a “scrawl” and that, in fact, she was capable of practicing any one of several distinct scripts on different occasions’.12 While it is true that this hand is remarkable partly for its neatness, a characteristic not shared by those works viewed as more certainly 6 The Social Edition defines this hand as one of the ‘two hands […] identified as that of Thomas Howard’; Heale offers some support for this attribution in her introduction (13–14), but does not intend a conclusive argument. 7 Irish, ‘Gender and Politics’, 89. 8 Irish, 89. 9 Murray, ‘Prisoner’; Remley, ‘Mary Shelton’. 10 Heale, Devonshire; Southall, Courtly Maker. 11 Remley, ‘Mary Shelton’, 42. 12 Remley, 49.
204
Gender and Position-Taking in Henrician Verse
Shelton’s, ‘the appearance of neatness in this section of the manuscript may reflect the […] fact that the leaves containing the Douglas-Howard items are ruled (albeit somewhat haphazardly) in pencil and may thus be distinguished from the preceding leaves, which lack ruling or any other sort of enhanced layout’.13 That is, the copying hand can be seen taking certain measures to ensure neatness in this section; reasonably, a writer who also possesses one hand which is elsewhere looser could be taking active measures here to write in a neater script. Additionally, ‘very strong evidence to indicate that Shelton was in fact the copyist who preserved the Douglas-Howard exchange is found in a hastily written sequence of characters that appear at the foot of […] 26v’.14 Remley argues that ‘the characters reveal a sequence of strokes and ligatures that bear close comparison with those seen in Mary Shelton’s three certain signatures’.15 The hand of this signature more closely resembles Shelton’s looser hand; notably, her signature on some of those other entries resembles the more careful hand of these poems. Remley calls the reading of this signature as ‘Mary Sh–lt–’ ‘almost certain’.16 Remley’s further conclusions about the nature of the manuscript necessitate additional consideration. He first addresses the courtly nature of the manuscript, arguing that in view of the furor caused by the impolitic betrothal of Douglas and Howard, it is perhaps surprising that their personal exchange circulated at all in a contemporary document such as the Devonshire manuscript, particularly if received critical judgments are upheld that would characterize the volume as a publicly maintained ‘courtly anthology’ or a frequently exhibited family guest book.17
As an explanation for this apparent paradox, Remley suggests that, instead, ‘the inclusion of the Douglas-Howard exchange is hardly surprising if the volume is viewed as a private document in which some of Henry’s disillusioned subjects were able to give voice to their dissent’.18 To extend the point of what it means to give a voice to dissent, though, the manuscript needs to be read as both a space for that dissent and as a (relatively) public document. The nature of position-taking poetry makes public resistance 13 14 15 16 17 18
Remley, 54. Remley, 54. Remley, 54. Remley, 54. Remley, 53. Remley, 54.
Resistance and Unit y in the Dougl as-Howard Exchange
205
possible, partly by limiting its scope through verse coding, yes, but also, as outlined in the last chapter, by limiting accountability and so providing some protected freedom of expression for its authors. The very neatness of the lines potentially suggests a shift in Shelton’s strategies that nonetheless indicate the utility of applying the broad strategies of transcription to these exchanges. Where in other poems Shelton uses irregular spelling and difficult handwriting to cloud and complicate her meaning, these exchanges are carefully, neatly recorded. First, this suggests greater importance in context; Shelton records the exchange with more care to signal that it should be treated as precious, in some sense. Second, this neatness may be meant to signal that there is less authorial action on the part of the transcriber than in other entries in the manuscript; if her contemporaries associated the particular messiness of her more usual hand with her original or manipulated compositions, the shift in hand may also signal a shift in style of transcription. Third, then, to return to Remley’s point, I would suggest that this neatness is meant to create greater audience accessibility. The previous chapter established the ways that Shelton used difficult formations and spellings as a position-taking strategy; if she chooses to avoid such strategies, she is aware of her choice to make these poems more readable and more easily interpretable. This choice does not decrease the political energies of the exchange, but instead increases the element of resistance by more openly reflecting a narrative of disagreement with Henry’s policies and a sympathy for his victims. Murray points out that the transcription of this exchange ‘offers material evidence not only of Thomas and Margaret’s communication within the prison, but also of the cooperation and collaboration of courtiers outside its walls’.19 In extending Murray’s point about collaboration, we can consider how the communal nature of the manuscript serves, also, to mask its writers—in a collaborative document, individual agency becomes obscured. Murray continues to the point that Given the complicity among friends and relatives in the surreptitious progress of the affair at court, it seems entirely likely that Shelton, or one of her associates, would have helped Thomas and Margaret exchange verse letters in the Tower, and moreover have ensured that these letters earned a permanent place that would preserve their shared literary efforts for posterity.20 19 Murray, ‘Prisoner’, 27. 20 Murray, 27.
206
Gender and Position-Taking in Henrician Verse
Remley advances a similar argument, claiming that ‘Mary embraced the opportunity to preserve the controversial Douglas-Howard exchange for posterity—or for some other purpose which, at this distance, must remain unfathomable’. 21 Returning to Murray’s points about communication, cooperation, and collaboration allows a reading where posterity is not, perhaps, entirely the point. Rather, the most important thing becomes the energy these poems have in their moment. These poems are recorded to be shared immediately, not only with future readers.22 The recorded exchange between Douglas and Howard, and the verses which seem to comment on the affair, masquerade as relatively harmless comment after the fact, toothless because the more immediate threat has already been neutralized through Howard’s death. At the same time, these verses gain force because of the very real consequences which followed from the events memorialized. Remley observes of the court context that ‘Mary Shelton and her companions […] were subject to a peculiar blurring of the distinction between the traditional ideals of the Chaucerian and Petrarchan love-poetry and the social and political exigencies of life at court’. 23 These poems show Douglas, Howard, and Shelton controlling and appropriating these blurred terms, subjects of, rather than to, the narratives they construct.
A. The Epistolary Exchange The first poem establishes the themes on which the poets base the argument formed by the exchange—and so, by extension, the message of the transcriber who chooses to preserve and re-contextualize their words. The first stanza establishes that ‘force’ has separated the two lovers,24 whose situation is then outlined in greater detail in the subsequent stanza: Alas that euer pryson stronge sholde such too louers seperate 21 Remley, ‘Mary Shelton’, 54. 22 Pincombe and Shrank identify this immediacy as a characteristic of most writing of the period, pointing to the strengths of Tudor literature as, also, its challenges: ‘it is frequently of the moment, reacting to particular events or sets of concerns; it often treats of its subject obliquely; it is generally written by the intelligentsia; and it is not shy about displaying its intellectual origins and aspirations’ (‘Travails of Tudor Literature’, 3–4). 23 Remley, ‘Mary Shelton’, 62. 24 BL MS Add. 17492, 26r, ‘Now may I morne’, line 2.
Resistance and Unit y in the Dougl as-Howard Exchange
207
yet thowgh ower bodys suffereth wronge ower harts shalbe off one estate25
The dramatic situation of the poem is quite literal to the lovers’ situation; the poet here references the difference between the two lovers in the speaker’s assertion that their hearts will be of the same estate—that is, though they themselves are not. The trope of unity—and of that unity as overcoming social difference—is repeated throughout the poems. The poet seems to answer, also, the charge that Howard has made this match for material gain; the poet assures the lover that ‘I wyll not swerue I yow Insure / for gold nor yet for worldly fere’.26 The poem also invests itself, in the final two stanzas, in a discussion essentially on the power of thought and the importance of positive thought.27 The speaker exhorts the beloved to think of their lover and to maintain a positive outlook; the speaker, in return, will do the same: I pray yow be off ryght good chere and thynke on me that louys yow best and I wyll promyse yow agayne to thynke off yow I wyll not lett 28
Irish, proceeding from the assumption that Howard is both speaker and writer, argues of this remonstrance that ‘while some of his poetic energy is directed externally, Lord Thomas seems far more comfortable turning inward, adopting in many poems a kind of stoic consolation anchored in two related notions: the rightness of his cause and the conviction that he and his beloved will be eternally reunited’.29 Although some of the later poems take a more condemnatory approach to the others who have imprisoned and separated the lovers, the speakers most consistently suggest this kind of stoicism and reflection as the surest route to preservation. The logic extends to the correlation between courtly love poetry and the role of the courtier, as well: having taken a risk that has incurred royal wrath, the courtly maker 25 Lines 5–8. 26 Lines 9–10. 27 The theme may, then, influence the poems written by Shelton and Douglas, discussed in the previous chapter, which represent entries in an argument on the value of performed gaiety or equilibrium in the face of tragedy. 28 ‘Now may I morne’, lines 15–18. 29 Irish, ‘Gender and Politics’, 93.
208
Gender and Position-Taking in Henrician Verse
now retreats into a more cautious position, but one which performs their own self-support and self-validation, dramatizing the rightness of the speaker’s actions while claiming not to offer explicit resistance or defense. Another possibility is raised by the lack of clearly gendered language in the poem. Heale points out this poem shares a line with a later entry set in the verse epistle section, on 29r.30 Though she contests this may be simply an indication that Howard is reusing a phrase or thought, she also questions whether the poem could ‘have been contained in, or be one of the “gentyll letters” sent to Howard from Douglas’.31 In either case, she surmises ‘what is clear is that the lyric expression of amorous fidelity is not considered gender specific’.32 The verse makes it clear that the language of passionate fidelity could be attributed to female speakers as well as to male speakers, making it at least probable that on occasion women able to write and playing an active part in the collection and enjoyment of courtly balets, might feel free to appropriate that language and the subject positions it implies for their own compositions.33
If the speaker is instead a woman—and a woman who would contextually read as Margaret Douglas—the speaker’s position becomes both more limited and more resistant. The woman speaker may, in fact, be limited to thoughts as defense; she is less likely to be able to directly respond to the charges against her. However, Douglas’s and Howard’s relative positions in the court also complicate these usual alignments; Douglas’s position has protected her from an explicit charge of treason, and she has been able to petition the King directly. That protection and favor, though, are simultaneously contingent on a kind of self-silencing, wherein Douglas may think that she has been unjustly charged, that her contracted marriage is valid, and that her position is right, but she must say that she submits to the King’s will and begs his pardon for her error. While a courtier who is a man must also practice this kind of self-abnegation, the very fact that such a position is often dramatized as a kind of feminization by men emphasizes the additional constraints implied by the use of a woman speaker. The sequence then continues with a more plaintive and dramatic entry, and one which, in a reference to ‘her […] whom I loue best’ seems to more 30 Heale, ‘Female Voices’, 16. 31 Heale, 16. 32 Heale, 16. 33 Heale, 17.
Resistance and Unit y in the Dougl as-Howard Exchange
209
definitely situate Howard as its poet/speaker.34 While the previous entry exhorts cheer and strength through fond remembrance, the poem here begins with a clear statement of the lover’s despairing state, whose ‘hart ys persed sodaynly’35 ‘[w]yth sorowful syghes and wondes smart’.36 The first lines of the following stanza continues the complaint, as ‘the bytter tears doth me co[n]strayne’,37 in spite of the speaker’s aims to ‘eschew / to wyte off them that dothe dysdayne / faythfull louers that be so trew’.38 The speaker’s feelings force the writing, thus offering a justification in the face of the admonishments of the first poem. While the first poem in the sequence establishes that to take comfort in thoughts and to perform good cheer is the best strategy, this second poem answers that this performance is impossible in the current conditions. Moreover, as the poem continues, the resistance to those that ‘dysdayne’ the ‘faythfull louers’ becomes sharper.39 The speaker continues with his plaint that ‘The one off us from the other they do absent / wych unto us ys a dedly wond’. 40 While the resistant tone is obvious, the wording is also strikingly careful. The speaker does not complain that these ‘they’ actively wound the lovers intentionally. Rather, ‘they’ are held responsible directly only for the separation; the separation is itself a wound, but the ‘wych unto us ys’ offers a distance between the actors and the effect. 41 Irish has already observed that the identity of this ‘they’ seems self-evident, acknowledging that ‘the presence of what one must assume to be King Henry and his ministers casts a long shadow over the sequence’. 42 In this fairly restrained entry, though, the ‘they’ may be construed as more mistaken than actively malevolent. The speaker’s appeals for the rightness of the lovers’ claims take on two further aspects: both a religious argument and an appeal to the larger public are added to the lovers’ defense. Howard moves clearly into a specifically religious argument for his rights as the poem progresses. The lovers are in the right, ‘seyng we loue in thys yntent / yn godes laws for to be bownd’. 43 This religious defense is a key characteristic of the sequence, and it is both 34 35 36 37 38 39 40 41 42 43
BL MS Add. 17492, 26v, ‘Wyth sorowful syghes’, line 15. Line 2. Line 1. Line 5. Lines 6–8. Lines 7 and 8. Lines 9–10. Line 10. Irish, ‘Gender and Politics’, 91. ‘Wyth sorowful syghes’, lines 11–12.
210
Gender and Position-Taking in Henrician Verse
bolstered and complicated by the unique religious atmosphere of the 1530s in England. The speaker needs both to appeal to the law of God, which he likely believes to be universal, and to negotiate the issue of that threatening ‘they’, who have changed the terms of that universal law. The knot of complication is dense: Henry VIII claims he is representing God’s law, and so the speaker cannot appear to say that Henry’s will has become God’s law; God’s law must not disagree with Henry’s will, lest the speaker appear to question the King’s claims. The appeal, then, is both religious and politic. If the lovers have been joined by the religious law that Henry aims to advance, then the King cannot actually fully condemn the contracted marriage. The speaker mitigates his criticism by casting the contract in terms of ‘yntent’. 44 The lovers have intended to follow God’s law, so the King should validate the planned marriage. However, if he chooses not to do so, the poem leaves open the possibility that this is not because the King is wrong, but because the lovers have failed in this intent. Remley makes the points regarding Howard’s poetic voice that ‘its vocabulary is traditional, and, in many cases, specif ically medieval in origin’, 45 and he uses ‘terms drawn from the conventional vocabulary’. 46 In this traditional vocabulary, Howard evokes the system of court service of which courtly love has become a part and a mirror, as explored in earlier chapters. In doing so, Howard partly engages with the charge against him, but he also emphasizes the newness of this charge. He is accused of violating this system—and in some of the poems, as we will see, Howard admits to some part of the charges, acknowledging that he has loved above his station. However, this aspirational love is itself an essential trope in the courtly love tradition. And there are two important factors that distinguish and validate Howard: his love is not adulterous and his love has been ratified by religious solemnization. In appealing to these traditions, Howard attempts to justify himself and to indict the failure of those who should be acknowledging his service by protecting him. The fealty of the courtier creates an obligation in his lord; throughout the sequence, Howard suggests that these obligations have not been met. If he has violated the terms of service, his crime is nonetheless demonstrably new: the crime has been created in order to charge Howard. The terms of obligation that he sees as not being met are more traditional—Henry has violated the terms of courtly relationships even as he appropriates that vocabulary for his courtly propaganda, and 44 Line 11. 45 Remley, ‘Mary Shelton’, 52–53. 46 Remley, 53.
Resistance and Unit y in the Dougl as-Howard Exchange
211
the speaker in these verses implicitly appeals to him to meet his end of their bargain. Simultaneously, however, the speaker also moves into an appeal to the larger public; this strain of appeal may partly compel the transcriber’s act of recording. Howard and Douglas are successful, that is, in appealing to the larger sympathy of the courtly audience; at least one member of that audience then seeks to preserve their complaint. Such preservation is inherently an act of resistance, because such sympathy is counter to the King’s actions and aims. The poem again covers that resistance, this time by implicitly including the King in that sympathetic audience. The final stanza appeals that Ther doth not lyue no lovyng hart but wyll lament ower greuous woo and pray to god to ease owre smart and shortly togyther that we my may goo47
The audience, in this final stanza, is inherently comprehensive. Unlike other verses, this one does not limit this claim to those who hear or read the poem or story; rather, ‘no lovyng hart’ can fail to support the lovers. 48 In this case, those hearts include both the powerless members of the audience, who can only pray to God; the councilors and powerful allies who can appeal to Henry, now proclaimed the country’s immediate intermediary with God; and Henry himself, who must pray to God for advice as to his own actions and plans in this case. The bottom of this page preserves one of the marks whose interpretation is so difficult in Devonshire, but whose presence has necessarily influenced the reception of these entries. The mark ‘fynis ma r h’ has been read in a wide variety of ways, though usually as a signature, however a given critic has interpreted the various ligatures.49 As implied by my own transcription here, I agree with Remley and Southall that the mark is likely Mary Shelton’s, both because of my reading of those letters I consider legible and because I concur with Remley’s correlation between this hand and Shelton’s other signatures.50 The context of the poem makes it unlikely that this mark is meant to refer 47 ‘Wyth sorowful syghes’, lines 17–20. 48 Line 17. 49 BL MS Add. 17492, 26v. Heale, who reads the marks as ‘Margrt’, notes that ‘the name is added in a different hand and presumably refers to [Douglas …]. The speaker of the poem is male, so the poem is not being attributed to [Douglas]’. (Devonshire, 99) 50 Remley, ‘Mary Shelton’, 54.
212
Gender and Position-Taking in Henrician Verse
to the poem’s speaker or subject. The speaker is almost certainly a man, given the feminine pronouns assigned to the beloved and their explicitly marital relationship, and the critical consensus that these verses apply to the imprisonment following the Douglas-Howard affair precludes Shelton as the intended recipient. As such, I agree with Remley’s claim that reading Shelton as the transcriptionist of the exchange, authored by Howard and Douglas, is a reasonable and helpful approach. Proceeding from this interpretation raises a question about the placement of this particular mark. This signature can be read as an implicit response to the final stanza of the poem, marking Shelton as both transcriber and respondent. To the claim that no one with a heart can help but pity the lovers, Shelton offers herself as witness. She then also marks herself as the transcriber who proves her sympathy through complicity. Although the vagueness of the shapes does recall Shelton’s use of complex spellings to cloud her own culpability, the act of making any mark at all in the context of these verses acknowledges an assumption of some measure of risk. The implication is that the injustice is grave enough to demand some kind of response and to warrant this act of preservation—Shelton’s signature draws attention to the act of transcribing, its risks, and the reason she feels she must take on these risks. The next poem in the sequence moves into a quite different tone, but one which underscores the reasons the lovers’ cause can be approached with sympathy. Before moving into a discussion that valorizes the lovers’ merit and the worthiness of their love, the poem begins with a valediction of the joys to be found in love. In contrast to the desperation that characterizes the entry on 26v, this first stanza opens, ‘What thyng shold cawse me to be sad / as longe ye reyoyce wyth hart’51 and seems almost a direct repudiation of the despair of the preceding entry; the poem might be read as a response from Douglas to Howard’s apparent decline. The second stanza emphasizes the mutual fidelity of the lovers, yff I shuld wryte and make report what faythfulnes in yow I fynd the terme off lyfe yt were to short wyth penne yn letters yt to bynd wherefor wher as as ye be so kynd as for my part yt ys but dewe lyke case to yow to be as true52 51 BL MS Add. 17492, 27r, ‘What thyng shold cawse’, lines 1–2. 52 Lines 8–14.
Resistance and Unit y in the Dougl as-Howard Exchange
213
The speaker here uses an echo of the Thynne Chaucer, from which this same hand will transcribe selections later in the manuscript, in the first four lines’ acknowledgement of the limitations of writing to preserve the worth of the beloved. From this insufficiency, the poem moves into one of the many presentations of fidelity as a reciprocal obligation between the two lovers; offered loyalty and service which exceed description, the lover’s ‘part’ is necessarily to return that love.53 The metaphorical connection between the lovers’ service to each other and their courtly service then once more reaches out to suggest the failure of those in power to fulfill their part and offer the lovers their ‘dewe’—protection in exchange for loving service.54 The placement of ‘for let them thynke and let them say’ in the third line of the third stanza then allows the line a double meaning.55 In one sense, it continues the discussion of methods of coercion established in the first two lines of the stanza: ‘My loue truly shall not decay / for thretnyng nor for punysment’.56 Whatever ‘they’ may think, say, threaten, or punish, the lover remains true. The line also, however, invites in a ‘they’ who will also ‘thynke […] and say / toward yow alone I am full bent’, as the fourth and fifth lines follow.57 This technique collapses the two ‘them’ referents, as well; the ‘them’ of the first reading becomes the same as that of the second, and those who threaten and punish will ultimately be forced to admit the lovers’ fealty and honesty. This strategy is repeated throughout the sequence and offers particularly fruitful evidence of the kind of protected position-taking practiced in the manuscript. The writers leave themselves various ‘outs’, but they also tend to offer an ‘out’ from their critique. Here, that out implies that the threatening ‘they’ can be reunited into the larger, communal ‘they’ which affirms the couple’s love, once ‘their’ false suspicions have been put to rest. Following a further religious appeal, ‘desyryng god that off hys grace / to send no tyme hys wyll and plesor / and shortly to get hus owt off thys place’,58 the poem closes with an image that emphasizes the lovers’ fealty to, and dependence on, each other: ‘then shal I be yn as good case / as a hawke that getes owt off hys mue / and strayt doth seke hys trust so trwe’.59 53 54 55 56 57 58 59
Line 3. Line 13. Line 17. Line 15–16. Lines 17–18. Lines 23–26. Lines 26–28.
214
Gender and Position-Taking in Henrician Verse
Remley has pointed out that this ‘striking simile […] is arguably Chaucerian in derivation’;60 as such, the image reinforces the claims of precedent and tradition. The simile also strengthens the connection between the courtly lover and the courtier. If freed, the lover will ‘strayt […] seke’ the beloved, leaving one prison for a more voluntary bondage.61 The established metaphorical connection between love and courtly service, though, also implies that if free, the lover will return to service to the King, proving desert of freedom by, in real terms, largely surrendering it. When the lover promises to stay ‘trusty and trw[e]’, those qualities speak directly to the charges of treason.62 The lover’s suit is true, both in terms of contracting the marriage for the ‘right’ reasons, not solely for political power, and in terms of not stemming from any treasonous intent. The repetition of ‘my part’ and ‘your part’ within this poem and across the sequence suggests both the manipulation of roles and the claiming of certain rights. The pronoun use in the commentary for the Social Edition reflects the frequent assignment to Howard, but an argument can be made for reading the poem in Douglas’s voice, as well. First, no gendered language demands that the speaker be understood as a man; the f inal figure does use masculine pronouns, but all of these refer explicitly to the hawk. Second, the poem does seem to offer a response to the laments of the previous poem—and that response aligns strongly with the advice of the first, also ungendered, entry in this set. Thematically, at least, there is an argument for at minimum a reconsideration of these poems that views the entries on 26r and 27r as written by or in the voice of Margaret Douglas, sandwiching the quite different entry more explicitly in Thomas Howard’s voice. The reading makes a certain amount of historic sense, as well, as Margaret’s comparative equanimity could be influenced by her much more comfortable positon: she was relatively assured of forgiveness and she was moved out of the Tower after only a few months of imprisonment. In contrast, Howard’s whole case rested on proving his desperate love, and his situation generally was much more dire, as his eventual death would prove. If the first three poems are indeed an exchange between Douglas and Howard, rather than a sustained sequence by Howard alone, then in the fourth poem Howard may take some heart from Douglas’s advice, though the tone still slides, occasionally, into notes of his earlier despair. While the 60 Remley, ‘Mary Shelton’, 53. 61 ‘What thyng shold cawse’, line 28. 62 Line 21.
Resistance and Unit y in the Dougl as-Howard Exchange
215
first stanza is a lament ‘that men be so vngent / [t]o order me so creuelly’,63 the second stanza moves into a more assertive and resistant stance: They kno my hart ys set so sure that all ther wordes can not prevayle Tho that the thynke me to allure wyth doubyll tonge and flaterynge tayle64
The third stanza combines the element of lamentation with the stronger claims of fealty and right, as the speaker decries that alas me thynke the do me wronge That they wold haue me to resyne my tytly tytle wych ys good and stronge that I am yowrs and yow ar myne65
While the first line of the stanza echoes the self-pity of the second poem, the final lines of the stanza are particularly confident. The speaker offers a clear appeal, in legal language, to the rights implicit in the act of marriage. The unnamed ‘they’ become usurpers of the speaker’s title. The title itself, further, is delineated in terms of reciprocity; although the third line of the stanza refers only to the speaker’s title, he goes on to describe the terms: ‘that I am yowrs and yow ar myne’.66 Further, the line balances against itself, emphasizing the theme of concordance. Although ‘I am yowrs’ comes first, its position after ‘that’ means that the metrically and visually matched ‘myne’ has more emphasis, at the end of the line. Thus, the two phrases share the markers of importance and emphasis, as much as possible. The two lovers—the ‘I’ and ‘yow’—are countered against that third, outside pronoun referent, ‘they’. In the next lines, the balancing continues, but this time the speaker’s agency is set against the outsider’s will, as ‘the wold’ strikes an internal rhyme with ‘I shold’: I thynke the wold that I shold swere your company for to forsake 63 64 65 66
BL MS Add. 17492, 27v, ‘Alas that men’, lines 1–2. Lines 5–8. Lines 9–12. Line 12.
216
Gender and Position-Taking in Henrician Verse
but ons ther ys no worldly fere shal cawse me such anothe to make67
Here, either Shelton as a transcriber chooses to introduce her strategy of complex spellings, or the original author has used a similar strategy: ‘I thynke the wold that I shold swere / your company for to forsake’ is just difficult enough to open up a possible alternative.68 The initial reading, of course, is ‘I think they would have me swear to forsake your company’.69 Alternatively, though, the spelling ‘the wold’ allows ‘I think thee would have me swear to forsake your company’—that is, perhaps the beloved suggests a measure of self-preservation. The lover’s response, though, works neatly for both interpretations: there is no worldly fear that could cause him to make such an oath. The doubled interpretation may allow Howard a moment of communication with his bride without fracturing the united front which they have presented to their accusers thus far; he can acknowledge her fears for his safety and return his own redoubled loyalty, without explicitly allowing their accusers to see any possible weaknesses in a defense that rests largely on unity and arguments for the sacramental and legal bonds of marriage. Following the establishment of legal claim, the speaker moves into a stanza that emphasizes the claims of affection; from here, though, he moves back to appeals to larger systems and powers. The fifth stanza is one of the many where the speakers specifically appeal their case to God, ffor I do trust ere yt be longe that god off hys benyngnyte wyll send us ryght where we haue wrong for servyng hym thus faythfulye70
Irish sees this stanza in terms of stark complaint, saying that ‘in the face of their increasingly dire circumstances it also becomes increasingly likely that the restoration of their union will be mediated, not by Henry’s earthly authority, but by divine mandate’.71 Certainly, the religious appeal is central to the lovers’ claims regarding the rights of their contracted marriage. 67 Lines 13–16. 68 Lines 13–14. 69 This is the reading Heale chooses in her edition, where the line reads ‘I think they would that I should swear’ (Devonshire, 101, ‘Alas that men’, line 13). 70 ‘Alas that men’, lines 17–20. 71 Irish, ‘Gender and Politics’, 93.
Resistance and Unit y in the Dougl as-Howard Exchange
217
However, while the language here is clearly religious, the language of service may also make an implicit appeal to Henry, himself. That is, rather than an explicit condemnation of the King that abandons hope of help from that quarter, the savvy courtiers may make use of Henry’s new role as head of the church to offer yet another appeal. The conflation of God and king is not a markedly unusual one, and the language here could be used both to petition to Henry and to offer him a more comfortable or laudable role as an out from his place in this unpopular situation. Howard and Douglas, in this formulation, affirm their unity through the use of the ‘we’ pronoun, but simultaneously affirm that they are united in service to a beneficent Lord who will both emerge on the side of ‘ryght’—morally and legally—and whose actions are motivated by his ‘benyngnyte’.72 Whatever Howard’s opinions of the King’s actual character, he would almost certainly have known that such a description and such a parallel would appeal hugely to Henry’s ego and his frank need to be liked, or, more aptly, adored. The next entry is one of the most structured in the sequence, and I suggest that its structure must be related, in some way, to Douglas’s preference throughout the manuscript for poetic structures based in repetition. Defining the exact relationship may be impossible; perhaps Howard writes the poem using a preferred strategy of his lover’s as a tribute to her, or possibly her later preference for this style was because of memories of her would-behusband’s use. The implication, in either case, draws the poem more clearly into the strategies of the manuscript as a whole—if Howard is intentionally echoing the structure for Douglas, the poem emphasizes their unity; if Douglas later gravitates towards the style in memory of Howard, then her entries throughout the manuscript may constitute, to some degree, implicit eulogies for her almost-husband. Further, of course, the two possibilities are not mutually exclusive; if Howard echoed Douglas’s favorite style in his compositions to her, she could also well have associated that style with him after the traumatic events surrounding their brief contract and his death. The repeated use of ‘my none’ for ‘mine own’ in the verses is striking in context, implying the negation of the contract even as it asserts the relationship codified therein.73 As his contracted wife, Douglas is Howard’s ‘own’; as a courtier under a charge of treason—and so a threat of attainder—Howard has nothing, and Douglas becomes a symbol of that absence, both its cause and its effect. The poem also returns to the image of writing materials, though in this case the speaker does not necessarily claim insufficiency. 72 ‘Alas that men’, lines 19 and 18. 73 BL MS Add. 17492, 28r, ‘Who hath more cawse’.
218
Gender and Position-Taking in Henrician Verse
Where elsewhere, the image, perhaps taken from the Thynne Chaucer, is always evoked as incapable of recording the lovers’ fealty, here the heart is as determined as can be recorded.74 The lines may be read as reflecting a willingness to record whatever is necessary to mark the truth of the statement. Where some poems mock writing as incapable of containing the truth of the lovers’ vows, writing here is capable of recording something of the heart’s determination. This moment may also invest some power back into the resistance recorded by these verses, indicating that writing is not sufficient, but is still a true method of recording, expressing, and preserving something of the truth. Although scholars attribute many other poems in the manuscript to Douglas, only one poem in this particular exchange is universally treated as Douglas’s.75 The assumption of a woman’s voice is here undeniable; from the first stanza, the speaker claims a clearly gendered identity: I may well say with Ioyfull harte as neuer woman myght say beforn that I haue takyn to my part the faythfullyst louer that ever was born76
Given the consensus that these poems were part of the prison exchange and my own analysis, paired with Remley’s, that the entire exchange is then later transcribed by Shelton, I treat Douglas as the author of this particular poem. This argument is, of course, also influenced by my earlier point that more than just this entry may be Douglas’s; the evidence could easily suggest that her role in the entries as a whole has been undersold, as the Social Edition editors would also seem to argue. This entry, particularly, seems to show strong evidence of Douglas’s composing strategies. In addition to the clear woman speaker, the poem echoes the four-line, ABAB rhyme scheme which Douglas often uses in the poems explicitly identified as hers. The woman speaker’s description also certainly fits the situation of the Douglas-Howard exchange, as the beloved man seems more endangered than the speaker: 74 Lines 15–16. 75 This is the only poem in this sequence that Heale identif ies as ‘probably composed by [Douglas]’ (Devonshire, 102); the Social Edition is an exception in noting that Douglas ‘could also have composed’ ‘Wyth sorowful syghes’ on 26v and in attributing ‘To yowr gentyll letters’ to Douglas, as well. 76 Douglas, BL MS Add. 17492, 28v, ‘I may well say’, lines 1–4.
Resistance and Unit y in the Dougl as-Howard Exchange
219
Wyth thretnyng great he hath ben sayd off payne and yke off punnysment yt all fere asyde he hath layed to loue me best was hys yntent77
The poem seems, here, to respond to the outrage of Howard’s entries describing efforts to convince him to renounce the match; Douglas knows that her lover has been threatened and cajoled, but her response is that his aims were never treasonous or dishonorable. His aim was only to ‘loue [her] best’,78 and so the denial which the unnamed ‘they’ expects of him cannot be forthcoming; indeed, Douglas thus characterizes Howard’s only potentially treasonable acts (the planned marriage to her and so denial of the King’s will) as inherently markers of loyalty. From this defense of her lover’s nature, the speaker then moves into a bold statement of her own commitment, and her defiant insistence that everything she has belongs to the man she plans to marry. She not only expresses no regret or reservation, but goes on to declare that yff I had more more he shold haue and that I kno he knowys full well to loue hym best vnto my graue off that he may both bye and sell79
Douglas engages the criticism of their disparate stations only to deny its validity,80 and her wording mounts a significant defense in terms of the values of the day. Even as her language acknowledges the station and money which union with her could bring Howard, she casts these benefits as inherently his right as her husband-to-be, rather than acknowledging any possible ulterior motivation. The poem ends with lines that emphasize both the lovers’ unity and, once more, the religious validity of their claims. Many critics have noted the unity of theme and voice between this poem and the other entries. Heale observes that ‘the phrases of gratitude and fidelity mirror, in some cases word for word, those gendered male in other poems’.81 Remley points 77 Douglas, lines 9–12. 78 Douglas, line 12. 79 Douglas, lines 17–20. 80 Heale offers several footnotes on the lovers’ disparate social statutes as relevant to this sequence (Devonshire, 97 and 104). 81 Heale, ‘Female Voices’, 16.
220
Gender and Position-Taking in Henrician Verse
out that Douglas ‘adapts entire lines of [Howard’s] poems to her own use, introducing her own echoic effects’.82 Irish writes that ‘the poem written in the voice of Margaret Douglas is a virtual pastiche of the lyrics presented throughout this prison sequence’.83 Irish continues that, thus, ‘Douglas [literalizes] the trope of oneness and indivisibility that guides the sequence as a whole’.84 There is a very real possibility that the echoes of theme and technique in this poem are also partially of Douglas’s own poems; the lovers together echo the themes which are most central to their arguments and share those techniques which are most useful for emphasizing these themes. This meaning may be particularly noticed in the verse’s last line, which promises to ‘bryng us shortly both in one lyne’,85 offering a poetic space that literalizes the speaker’s wish. As a further example, the epistolary hints of the closing of Douglas’s poem on 28v are echoed in the explicit declaration of the same letter/poem hybrid at the beginning of Howard’s next poem. 86 By emphasizing the communicative natures of the poems, Douglas and Howard highlight not only the closeness of their own marital relationship, but also the loyalty of their friends in courtly circles, who must participate in some way in this exchange of notes. Howard opens his poem To yowr gentyll letters an answere to resyte both I and my penne there to wyll aply and thowgh that I can not yor goodnes aquyte In ryme and myter elegantly yet do I meane as faythfully87
Once more the poem speaks to the insufficiency of words; the repetition of this trope makes more immediate the lovers’ situation, unable to communicate with each other except through this insufficient medium, as the first two lines of the poem emphasize. The element of communication through others, then, is extended beyond this cooperative circle as the speaker continues: 82 Remley, ‘Mary Shelton’, 53. 83 Irish, ‘Gender and Politics’, 95. 84 Irish, 95. 85 Douglas, ‘I may well say’, line 24. 86 The Social Edition attributes this poem to Douglas, but Heale also reads the poem as Howard’s (Devonshire, 104). Baron seems to agree with Bond’s early argument for reading the poem as Howard’s response to Douglas’s preceding verse (‘Fitzroy’s Hand’, 325). 87 BL MS Add. 17492, 29r, ‘To yowr gentyll letters’, lines 1–5.
Resistance and Unit y in the Dougl as-Howard Exchange
221
And where as ye wyll contynew myne To reporte for me ye may be bold That yff I had lyves as argus had yne yet soner all them lyse I wold then to be tempte for fere or for gold yow to refuse or to forsake wych ys my faythful and louyng make88
The assertion that Douglas may report of Howard that he maintains his own claim may suggest that this entry dates after the amelioration of Douglas’s circumstances, when she was moved from the Tower to Syon Abbey. The final line once again uses the flexible spelling of the period to offer two interpretations. The clearer sense is that Howard will not refuse his faithful and loving mate, and ‘spouse’ is one meaning of the clear ‘make’ spelling on the page. While the first reference may be to Douglas as his make, the alternate definition of ‘make’ as ‘relating to the form or composition of something physical or immaterial’ was already well-established by the time of Howard’s writing.89 Thus, Howard’s ‘make’, his composition and character, are also faithful and loving. His fealty to his wife-to-be may, indeed, be the only way for Howard to prove his fealty to his King. Howard will not refuse or forsake Douglas because his make is faithful and loving. In evoking the rules of love service which Henry had used to define his court, Howard appeals to the King that his love is true and that he is a person of true character—a good courtier who will serve his King as faithfully as his wife. When the transcriber of these verses records these pleas of fealty, quite likely after Howard’s death but at best during the time of his imprisonment, the indictment of the King’s refusal to acknowledge these loyalties, or to play by the rules which he has established for courtly love, is inescapable. The final entries related to this exchange shift noticeably in content, even as that shift creates a link to later entries in the Devonshire. Apparently building from those images and lines adapted from the Thynne Chaucer, the lovers now seem to exchange adapted and excerpted verses from Troilus and Criseyde. Markers within the poems suggest that Howard was responsible for the changes; Irish observes that As a tale of secret love wrenched apart by geopolitical machinations, Troilus and Criseyde has some obvious similarity to the real-life romance of 88 ‘To yowr gentyll letters’, lines 8–14. 89 OED, s.v. ‘make’.
222
Gender and Position-Taking in Henrician Verse
Thomas Howard and Margaret Douglas, and it is not surprising that Lord Thomas found in Chaucer a complaint rhetoric—and, indeed, a rhetorical situation more generally—remarkably apt for the events of his own life.90
The first of these selections is a short entry, written as a couplet, which builds on the earlier poem’s increasing tone of accusation: And now my pen alas / wyth wyche I wryte quaketh for drede / off that I muste endyte91
Irish argues that using this excerpt as a preface activates an associative network in which ‘Lord Thomas adopts the mantle of Troilus, and implicitly aligns their respective fates’.92 Through this context, Irish approaches the importance of the word choice ‘endyte’, which ‘is flexible enough to entail both the act of composition and the act of transcription’;93 through this word choice, Howard becomes both Chaucer the poet and Troilus the lover, while the context also means that ‘endyte […] slides into its morphological cousin indict’.94 Irish details the etymological relationship between the words, including the use of ‘indite’ and ‘indyte’ in the sense of ‘indict’ in contemporary grammars and dictionaries; he concludes that, when the couplet is read in Howard’s voice, ‘it is hard to imagine that the couplet doesn’t entail this sense of accusation, a charge against the captors who tore him from his rightful wife’.95 Thus, Howard joins the technique of slippery spelling, used throughout the manuscript, to his own more specific manipulation of tradition, both literary and courtly. The excerpts which follow this couplet continue to manipulate that network, while Howard furthers his own use of the transcriptions through space and omission. Howard leaves space, throughout these excerpts, where Criseyde’s name should occur, as in the very first stanza: O very lord / o loue / o god alas That knowest best myn hert / & al my thowght […] 90 Irish, ‘Gender and Politics’, 97. 91 BL MS. Add, 17492, 29v, ‘And now my pen’. The shorter quotation is embedded here to eliminate possible confusion over the slash marks present in the text. 92 Irish, ‘Gender and Politics’, 98. 93 Irish, 98. 94 Irish, 99, italics in original. 95 Irish, 99.
Resistance and Unit y in the Dougl as-Howard Exchange
223
Syns ye [ ] / & me hau fully brought Into your grace / and both our hat hertes sealed howe may ye suffer alas yt be repealed96
The echo of the Henry/God conflation is particularly strong in these verses; the lover’s appeal is ostensibly to God, of course, but once more the more direct, worldly appeal is to his newly declared primary intermediary in England. The f inal lines show the divorce that has occurred between heavenly and earthly divine will: God has joined these two together, and Henry has put them asunder. When Howard omits Criseyde’s name, he also omits Margaret’s, even as he invites the substitution. Irish calls this omission ‘an act of symbolic defiance, in which Lord Thomas, whose body belonged increasingly to the crown, chose actively to inscribe his wife’s name, not into the pages of the manuscript, but into the book of his heart’;97 Murray sees in the spaces ‘the very rooms of the Tower itself: indiscriminate spaces waiting—with implacable blankness—for their next unwitting inmates’.98 I would add that the spaces can encompass both of these resistant interpretations while also echoing the poets’ use of ‘my none’ for ‘mine own’—Margaret is absent from him because of their bond, and the relationship that calls the exchange which preserves Howard’s voice into existence is simultaneously the relationship that incarcerates and ultimately erases his body. The failures of the ‘they’ who threaten the lovers have made the bonds which should define courtly relationships into traps which erase personhood.
B. Contextual Affect and Effect The Douglas-Howard exchange shades practically the entire manuscript; it seems likely that the actual affair shaded Henry’s court for some time. Particularly, Douglas’s choices throughout the manuscript seem influenced by her contemporary struggles. While the exchange between Shelton and Douglas discussed in the previous chapter is one such example, the entry which precedes the next set of poems entered by the hand TH2 is perhaps an even clearer one. In the poem, likely an original composition,99 Douglas’s 96 BL MS Add. 17492, 29v, ‘O very lord’, lines 1–7. 97 Irish, ‘Gender and Politics’, 100. 98 Murray, ‘Prisoner’, 35. 99 Heale identifies the work as ‘probably’ by Douglas (Devonshire, 246); the editors of the Social Edition offer a more conservative ‘possibly’.
224
Gender and Position-Taking in Henrician Verse
speaker dramatizes a kind of vocal will, a speech in which she justifies her life and records her final wishes.100 Heale says of the poem that ‘unless, or even if, we imagine her reciting these lines to her father, the Earl of Angus and assembled friends, the poem already represents a version of any real events mediated through the discourse of romance and shaped in verse’;101 she says this makes it ‘impossible to separate the private and the public, the factual and the fictional’.102 I would note, further, that Margaret had been living in London, in the household of Henry’s eldest daughter, for several years.103 A compelling suggestion, as also offered by the editors of the Social Edition, is that Douglas intends ‘ffather Dere’, in her opening stanza, to refer to, and perhaps to appeal to, Henry.104 If this is the case, a doubled reading of the phrase is almost inevitable. On the one hand, Douglas—desperate and out of favor—needs Henry to understand her appeal sincerely. She does need his favor; she is not, in fact, dying, and only with his approval can she regain any kind of power or efficacious agency. On the other hand, the situation demands a more ironic reading if the ‘Dere’ and ‘swet’ father is also the one responsible for the tragedy at hand.105 In fact, this ironic reading may well be strengthened by the necessity of the ‘sincere’ mask. Douglas’s contemporaries would read, simultaneously, her dependence on Henry and her resentment and fear of him; they would understand the terms by which Douglas, and essentially all courtiers, existed, bound to the King regardless of his cruelties, and even more aware of those cruelties because of the impossibility of escaping the system. Following this imagined, recriminatory ‘will’, the hand which transcribed the Douglas-Howard poems picks up once more. These verses are separated from Douglas’s complaint by several blank pages, then entered, like the Douglas-Howard exchange, apparently all at once or at least in a consistent style—the hand is regular, the ink and spacing fairly consistent, and several pages show evidence of pencil ruling. Irish records the general consensus that ‘the identical handwriting and the usage of Thynne’s 1532 Chaucer suggest that the two sequences could emerge from the same period of 100 Douglas, BL MS Add. 17492, 88r, ‘Now that ye be assemblled heer’. 101 Heale, ‘Female Voices’, 17. 102 Heale, 17. 103 Marshall, ‘Douglas’, Oxford DNB. In the early 1530s, Henry set Douglas up in his eldest daughter’s household; moreover, as Marshall notes, Henry ‘for a time treated his niece as heiress presumptive’, strengthening the filial sense of the relationship. As noted in Chapter Four, Douglas eventually became part of Parr’s household, as well. 104 Douglas, ‘Now that ye be assemblled heer’, line 3. 105 Douglas, lines 3 and 17.
Resistance and Unit y in the Dougl as-Howard Exchange
225
inscription’; 106 because Irish believes Howard to be the transcriber, he suggests Howard’s imprisonment as potentially providing the dates of entry. While I find Shelton a more compelling candidate for the copyist, the connection between the two sections is still relevant. While some critics have read the transcriptions, many from Thynne, as ‘typical’ entries in any commonplace book, Remley’s ‘reading […] suggests that the transcription of these medieval borrowings was undertaken in a less convivial spirit’.107 He argues that, ‘rather than participating in some sort of courtly amusement, Shelton […] finds a voice for her indignation at the treatment of women of her time by hypocritical lovers’.108 I would add to that reading that the copyist’s selective use of this hand—or, if the copyist is not Shelton, then their generally selective entries in the Devonshire—reinforces and magnifies the indignation Remley identifies. By suggesting connections to the Douglas-Howard affair, through placement, style, and source material, the writer suggests that this indignation is directed at the King whose own love life was so far from the courtly ideal, and yet who exerted such judgment and control over those around him who strayed from those ideals; this is more typical of the writings in the manuscript than less meaningful copy work would be. Many of these verses ‘read as miniature volleys in the medieval defense-ofwomen tradition’,109 whether through manipulation of the original or simply through selective transcription. The first excerpt is a carefully punctuated verse; alternate punctuation allows for alternate, less laudatory readings of the lines. Womans harte vnto no creweltye enclynyd ys / . but they be charytable pytuous deuoute ful off humylyte shamefast debonayre / a and amyable dredeful / and off wordes measurable what women these haue not parauenture folowyth not the way off her nature110
A series of flourishes on the page then separates this poem from the following, longer poem, which begins 106 Irish, ‘Gender and Politics’, 103. 107 Remley, ‘Mary Shelton’, 55. For further readings on the complex constructions of gendered identities in the Devonshire, see Shirley, ‘Reading Gender’. 108 Remley, ‘Mary Shelton’, 56. 109 Irish, ‘Gender and Politics’, 102. 110 BL MS Add. 17492, 89v, ‘Womans harte vnto no creweltye’.
226
Gender and Position-Taking in Henrician Verse
ys thys afayre avaunte / ys thys honor a man hymselfe accuse thus and diffame ys yt good to confesse hymself a traytour and bryng a woman to sclaundrous name and tell how he her body hath don shame no worshyppe may he thus to hym conquer but great dysclaunder vnto hym and her 111
In the context of Henry’s court, the transcription may at first seem to have a significant number of referents; in fact, drawing attention to that field may be part of the purpose of the transcription. While the poem may use ‘traytour’ metaphorically, in the sense of a betrayal of the terms of a love affair, men in Henry’s court were branded traitors much more literally for their confession of affairs. The insistence on the dishonor this brings upon the men seems to emphasize this element of the transcriber’s cultural context. However, even as the initial reading may seem to encompass any number of affairs at the court, a closer reading begins to eliminate those possible references, one by one. In the case of the Anne Boleyn affair, perhaps the most notable and most immediately destructive of the scandals at Henry’s court, not one of the noblemen accused did ‘confess’; only Mark Smeaton offered anything like a confession. Further possible references to disastrous love affairs—the Douglas-Howard exchange or perhaps, to a lesser degree, the more successful marriage between Charles Brandon and Mary Tudor—are negated by the man’s accusation of the woman, when in most of the marriage scandals at Henry’s court the lovers stayed remarkably steadfast in their loyalties to one another. The affairs of Henry’s fifth queen, Katherine Howard, do fit this description quite nicely—the men involved almost all confessed, and the affairs may partly have been brought to light due to breathtakingly misguided braggadocio—but most study of the Devonshire concludes that the bulk of the writing was entered in the 1530s, slightly before Katherine Howard’s short marriage to the King. The point, then, may not actually be that the lyric can refer to so many people, but rather that the lyric refers to essentially no one who has actually been punished for this kind of conduct at Henry’s court. Indeed, perhaps the only person who has publicly accused women of misconduct which also involved himself is Henry. Further, this reading opens up a new interpretation of the lyric, which does not, explicitly, refer to the cause of the shame as an affair outside of marriage. One reading available to the contemporary audience would have understood this verse as referring to the very public 111 BL MS Add. 17492, 89v, ‘Ys thys afayre avaunte’, lines 1–7.
Resistance and Unit y in the Dougl as-Howard Exchange
227
trials and depositions of Henry’s wives. Connected to the first poem, this second emphasizes the importance of perspective—of remembering who is controlling the narrative. This reading introduces a much greater logic for the stanza that follows: To her nay / yet was yt no reprefe for all for vertue was that she wrowght but he that brwed hath all thys myschefe that spake so fayre / & falsely inward thowght hys be the sclawnder as yt by reason ought and vnto her thanke perpatuel that in suche a nede helpe can so well u 112
In another context, the claim that the woman lover is blameless because ‘all for vertue was that she wrowght’ would be practically nonsensical.113 The writer could argue in the woman’s defense that she acted from love, that she acted naively, or that she acted with good intentions, but none of those would erase the inherent sin, for the typical early modern aristocratic reader, of an affair outside of marriage, whether the woman in question was married or not. The poem could, in a contrasting interpretation, be read as implying that the accused woman is innocent, and the man is spreading a false tale, thus absolving the woman of blame. The poem does not, however, explicitly exonerate the woman; it only explicitly condemns the bragging man. Still, then, in an application to Henry the lines take on their clearest and most reasonable rhetorical shape. The lines charge the general hypocrisy of men, certainly, but they also code a charge of singularly catastrophic hypocrisy against the man who would control the system—in terms partially authorized by his own early verse constructions. The poems which are entered next, on 90r, have already been discussed in the opening to this work, and I wish here only to add the final entry on the page to that context and discussion: Alas what shuld yt be to yow preiudyce yff that a man do loue yow faythfully to yowr worshyp eschewyng euery vyce so am I yowrs and wylbe ueryly I chalenge nowght of ryght / and reason why 112 ‘Ys thys afayre avaunte’, lines 8–14. 113 Line 9.
228
Gender and Position-Taking in Henrician Verse
for I am hole submyt vnto yowr servyce ryght as ye lyst yt be ryght so wyll I to bynd myself were I was at lyberty 114
While most courtly love lyrics also act as poetry on courtly service, in this case there is particularly ‘little […] that strictly necessitates an eroticized reading of the stanza in isolation […] if one temporarily disassociates the language of service from its figurative, eroticized form […] the lines sound equally as a profession of political loyalty’.115 The entry may serve two distinct purposes in the context of the accompanying lines on women’s virtue and the larger position-taking context of the manuscript. First, the poem defuses some of the resistant energy of the earlier verses. In this way, it both appeals to Henry’s vanity and protects the less acceptable energies preserved in the poem. Second, though, the poem may, to the informed reader, offer an even further criticism of Henry, who has condemned faithful women who were also faithful to him, in service. That these entries are made following several blank pages further emphasizes that this grouping on the page is more than economy or necessity; the transcriber wants the three poems read together, but also maintains their separation through the use of a single flourish between each set of lines. Several more blank pages then separate these entries from the next set of lines entered by TH2, but the scribal hand, the theme, and the source material all construct connections between the entries, as, indeed, does their position which, though separate from each other, is more extremely divorced from the rest of the text. The first of these poeticizes Jason’s betrayal of Medea;116 the criticism of the unfaithful lover is both typical and topical, particularly when that lover is a famous, but compromised, hero who has tarnished his reputation by failing to value his wife appropriately. Indeed, so long as these entries were made after the Douglas-Howard affair, the implicit criticism of Henry already has two strong referents. The mention of the conquered ‘flece off gold’ may be seen as a sidelong allusion to the triumphs of the Field of Cloth of Gold, where Katherine of Aragon was Henry’s celebrated consort.117 Further, Katherine was acting as regent during the English victory at the Battle of Flodden; Henry quite literally ‘got victory’
114 115 116 117
BL MS Add. 17492, 90r, ‘Alas what shuld yt be’. Irish, ‘Gender and Politics’, 105. BL MS Add. 17492, 91r, ‘How frendly was medea’. ‘How frendly was medea’, line 2.
Resistance and Unit y in the Dougl as-Howard Exchange
229
from her.118 However, Medea is not herself an uncompromisingly virtuous character, perhaps making Anne Boleyn an equally likely candidate for the readers’ recollection as they engage with this verse. Both women could also be viewed as having gotten Henry ‘great pryce and name’:119 Katherine through her diplomatic connections and the humanist strengths for which the court became renowned under her co-reign, Anne through her role in establishing Henry as head of the Church in England and for furthering that same humanist reputation. The poem implicitly participates in the traditions the King himself had established, which claimed that only the true lover was truly noble, but it turns those conventions against both the mocked Jason and the King who has abandoned the ideals he once created. From the mythical, heroic generalized plaint on Medea and Jason, though, the transcriber moves into a first-person quotation from Chaucer, and one of the few which does not require any manipulation to serve the thematic ends of the grouping. The complaint is originally spoken by a woman, and the complaint is against men, rather than women. In lines like ‘wher ys the truthe off man who hath yt slayne / she that them loueth shall them fynde as fast / as in a tempest ys a rotten maste’120 changeability becomes a masculine trait, rather than a feminine—or, at minimum, it becomes a trait that is true of men, and women become the party victimized by men’s fickleness. The subject, in an immediate sense, is more clearly love than in some of the other entries, though the larger criticism of men’s fickleness and breaches of truth still seems, thematically, to relate to Henry, as well. The poem that follows, then, maintains the speaking ‘I’, but also moves into critiques that are less definitely posed in terms of love and, if not about love, are also more threatening and more immediate: yff yt be so that ye so creuel be that off my death yow lysteth nowght to retch that ys so trewe and worthy / as ye se no more than off a mocker or a wretch yff ye be suche yowr beaute may not stretch to make amendes off ss so crewel a dede Auysement ys good before the nede121
118 Line 4 characterizes the woman ‘by whom vyctorye he gate’. 119 ‘How frendly was medea’, line 7. 120 BL MS Add. 17492, 91r, ‘for thowgh I had yow’, lines 5–7. 121 BL MS Add. 17492, 91r, ‘Yff yt be so’.
230
Gender and Position-Taking in Henrician Verse
The first lines clearly establish the speaker’s fealty and desert; the speaker is ‘trewe and worthy’.122 However, the unperceptive and ungrateful object of the speaker’s complaint does not demonstrate appropriate value, instead thinking no more of the speaker than of some ‘lower’ person. Certainly, the language of courtly love and courtly service overlap here; further, the poem reflects some of the tensions which caused courtiers to push back against Henry’s increasing centralization of power, particularly the sense that this placed their value down among the other subjects, rather than granting them greater scope. The fifth line may imply a gendered subject that counters the usual slant of this section of entries, since the ‘beaute’ that fails to cover these failings may seem likely to be intended as a feminine attribute. However, the larger context of the poetry has established a woman’s voice for this series, and no explicit language here reverses that attribution.123 The passage continues onto 91v, but in the entries on 91v the writer changes the separating and distinguishing flourishes used in the earlier sections, moving into less-formed dashes. Perhaps mimicking Howard’s alteration of Chaucer to strengthen the associations between the two sections, ‘Shelton alters the pronouns in Pandarus’s conversation with Criseyde to turn the application of the lines to her own circumstances’.124 If Shelton earlier copies the Douglas-Howard exchange, with its many references from the 1532 Thynne Chaucer, as an act of resistance against Henry, then these later transcriptions from the Thynne may operate in a similar vein. Drawing on the same themes and the same lovers, Shelton may also use the shift in her writing hand to signal the connection. The consistent marks between these stanzas further the link. In these poems, though, Shelton also makes great use of her established strategy of slippage in spelling and meaning: Wo worthe the fayre gemme vertulesse wo worthe that herbe also that dothe no bote wo worthe the beaute that ys routhlesse wo worth that wyght that trede eche vnder fote and ye that ben off beauty croppe and rote Iff therwythall in yow be no routhe than ys yt harme that ye lyuen by my trouthe125 122 Line 3. 123 See, also the comparison of Henry VIII to Judith as his country’s ‘beautie’ in Morley’s Exposition, discussed in the second chapter. 124 Remley, ‘Mary Shelton’, 56–57. 125 BL MS Add. 17492, 91v, ‘Wo worthe the fayre gemme’.
Resistance and Unit y in the Dougl as-Howard Exchange
231
The woe/woo slippage is particularly noticeable in this passage, especially following a page full of excerpts on the untrustworthiness of men. Once more, the ‘I’ and ‘you’ of the poem remain ungendered, and the context may offer the only indication as to a speaker. The poem which shares this page builds off of these themes and techniques. for loue ys yet the moste stormy lyfe ryght off hymself / that euer was begonne for euer some mystrust / or nyce stryfe there ys in loue / some cloude ouer the sonne thereto we wetched women nothyng conne whan to vs ys wo / but wepe and syt and thyngke our wreake ys thys / our owne wo to doo drynke126
If the reader has assumed ‘woo’ for ‘wo’ in the preceding verse, then this poem revises that impression—here, the more obvious sense is woe. The more subversive sense, though, may be ‘woo’—that is, that this poem may point out the lie of courtly love. Under the terms of the system, wooing is offered to a woman as a tribute, even as, in fact, the woman has little power in the relationship, and her reactions may be determined by any number of factors far outside of her emotions or even her control. Certainly, the trail of Anne Boleyn and the accusation of Margaret Douglas made quite clear that courtly gestures of love, offered to a woman, could easily become the source of weeping and ‘wreake’. The defense of women throughout this section is based in a traditional understanding of certain valuations of privilege and of religious right. The poems’ defenses of love offer strong echoes of many of Henry VIII’s own verses, emphasizing the note of indictment that hand TH2 has integrated throughout her selections. In echoing Henry’s own sentiment, the courtiers coopt his terms of resistance and make them their own, and they imply that this appropriation is necessary precisely because the King has failed his own standards. Henry has supplanted the traditional rights of both the church and the aristocracy, and the verses that seem to bear clearest relation to the events of the Douglas-Howard affair frequently express resistance to this usurpation. In the focus on claiming the primacy of a vowed marriage, the writers reinstate certain religious ‘truths’ as beyond the reach of the King—significantly, the same category of ‘truth’ which has driven the King to redefine the religion of the kingdom. In a focus on the obligations of courtly service, the writers 126 BL MS Add. 17492, 91v, ‘for loue ys yet’.
232
Gender and Position-Taking in Henrician Verse
express both resistance to newly imposed norms and resentment at the exploitation and abandonment of a supposedly reciprocal arrangement. The difficulty of these analyses lies in expressing what I do and do not intend by these connections and conclusions. My argument, here, is not that any single author composed any single poem about exactly and only the situation to which I have linked the verse. Nor is that any single transcriber intended to create one—and only one—reference in their contribution. Rather, I hope to reactivate some of the interpretative work and possibility of the manuscript’s reader/writer community. Exploring these interpretive connotations offers us one way to access ‘the social context that shaped [verse manuscripts] and the system through which they were originally produced, circulated, altered, collected, and preserved’.127 As acknowledged at the opening of this final section, meaning here is less clear because of the embeddedness of these systems. If specific interpretations are more muddied, though, the systems of exchange are clear; in fact, some of that muddiness likely accrues from writers’ greater awareness of and reliance on specific interpretive frameworks shared by their readers. What Deborah Solomon identifies as the ‘provocatively cryptic side to handwriting’128 dramatizes, as she articulates, the balance we must always keep in mind of, on the one hand, intimacy, immediacy, and specificity, and, on the other, obscurity, distance, and inaccessibility. Moreover, in the communal manuscript, where many hands participate, the collaborative nature of the text emphasizes the many composing, recording, and reading lenses that were at work on the texts from their earliest compilation. That community appropriates the codes of courtly love and service, authorized by Henry himself in the early years of his court, to energize and protect their position-taking verse.
Works Cited Baron, Helen. ‘Mary (Howard) Fitzroy’s Hand in the Devonshire Manuscript’. The Review of English Studies 45, no. 179 (1994): 318–335. Carlson, David R. ‘The Henrician Courtier Writing in Manuscript and Print: Wyatt, Surrey, Bryan, and Others’. In A Companion to Tudor Literature, edited by Kent Cartwright, 151–177. Malden, MA: Wiley-Blackwell, 2010. The Devonshire Manuscript. 1525?–1567? MS BL Add. 17492. British Lib., London.
127 Marotti, English Renaissance Lyric, 2. 128 Solomon, ‘Representations’, 679.
Resistance and Unit y in the Dougl as-Howard Exchange
233
Harrier, Richard. ‘A Printed Source for The Devonshire Manuscript’. The Review of English Studies 11 (1979): 179–194. Heale, Elizabeth. ‘“Desiring Women Writing”: Female Voices and Courtly “Balets” in some Early Tudor Manuscript Albums’. In Early Modern Women’s Manuscript Writing, edited by Victoria E. Burke and Jonathan Gibson, 8–31. Burlington, VT: Ashgate, 2004. Heale, Elizabeth, ed. The Devonshire Manuscript: A Women’s Book of Courtly Poetry. Toronto: Center for Reformation and Renaissance Studies, 2012. Heale, Elizabeth. ‘Women and the Courtly Love Lyric: The Devonshire MS (BL Additional 17492)’. The Modern Language Review 90, no. 2 (1995): 296–313. Irish, Bradley J. ‘Gender and Politics in the Henrician Court: The Douglas-Howard Lyrics in the Devonshire Manuscript (BL Add 17492)’. Renaissance Quarterly 64 (2011): 79–114. Marotti, Arthur F. Manuscript, Print, and the English Renaissance Lyric. Ithaca: Cornell UP, 1995. Murray, Molly. ‘The Prisoner, the Lover, and the Poet: The Devonshire Manuscript and Early Tudor Carcerality’. Renaissance and Reformation 35, no. 1 (2012): 17–41. The Oxford Dictionary of National Biography Online. Oxford UP, 2015. Web. https:// www.oxforddnb.com/. Accessed 31 May 2021. The Oxford English Dictionary Online. Oxford UP, 2019. Web. https://www.oed. com/. Accessed 31 May 2021. Pincombe, Mark and Cathy Shrank. ‘Prologue: The Travails of Tudor Literature’. In The Oxford Handbook of Tudor Literature, 1485–1603, edited by Mike Pincombe and Cathy Shrank, 1–17. New York: Oxford UP, 2009. Remley, Paul G. ‘Mary Shelton and Her Tudor Literary Milieu’. In Rethinking the Henrician Era: Essays on Early Tudor Texts and Contexts, edited by Peter C. Herman, 40–77. Urbana: U of Illinois P, 1994. Shirley, Christopher. ‘The Devonshire Manuscript: Reading Gender in the Henrician Court’. English Literary Renaissance 45, no. 1 (2015): 32–59. A Social Edition of the Devonshire Manuscript MS (BL Add. MS 17492). Wikibooks, The Free Textbook Project. https://en.wikibooks.org/w/index.php?title=The_Devonshire_ Manuscript& oldid= 3469218. Accessed 1 July 2022. Solomon, Deborah. ‘Representations of Lyric Intimacy in Manuscript and Print Versions of Wyatt’s “They flee from me”’. Modern Philology 111, no. 4 (2014): 668–682. Southall, Raymond. The Courtly Maker: An Essay on the Poetry of Wyatt and His Contemporaries. Oxford: Basil Blackwell, 1964. The Workes of Geffray Chaucer Newly Printed, with Dyuers Workes Whiche Were Neuer in Print Before. Ed. William Thynne. London: Thomas Godfray, 1532.
Conclusion Abstract: This conclusion emphasizes the need to shift our framework to consider Henrician verse work as position-taking, enabling a move away from a reification of individual authorship and allowing more space to consider both communal production and the prominence of women in court spaces for poetic production. Three threads come together in the work: the extension of our understanding of verse as political positiontaking, combined with the work on manuscript poetry that emphasizes both multiple authorship and form as part of verse’s action, both creates and is reinforced by a framework that restores women to their place in Henrician courtly verse work. To demasculinize this literary history is to better understand the politics of communally read, communally created position-taking verse. Keywords: Henrician translation; early modern women’s writing; Tudor verse transcription; Devonshire Manuscript; gendered canon formation; courtly love lyric
In the introduction to their edited collection of early modern letters, Robert J. Clements and Lorna Levant include the almost warning note that, in reading the collection as a whole, the ‘servility imposed by patrons on their greatest Humanists, writers, artists, and musicians emerges in a disturbing dimension’.1 The comment forewarns the reader of the shock of seeing, for example, Wyatt’s groveling, self-serving letter to Henry VIII, apparently written immediately after Anne Boleyn’s execution. In the letter, Wyatt claims to have warned the King that Anne was ‘a bad woman’ whom the King should not marry.2 More shockingly, he then goes on to accuse Anne of likely sleeping with a groom, after leaving her bed mid-tryst with Wyatt himself, and claims that within the same week he ‘had [his] way with her, 1 2
Clements and Levant, ‘Introduction’, Renaissance Letters, xii. Wyatt, ‘To Henry VIII’, in Renaissance Letters, eds. Clements and Levant, 397–398.
Quoss-Moore, R.M., Gender and Position-Taking in Henrician Verse: Tradition, Translation, and Transcription. Amsterdam: Amsterdam University Press, 2023 doi 10.5117/9789463723534_con
236
Gender and Position-Taking in Henrician Verse
and, if your Majesty, when you banished me, had permitted me to speak, I should have told you what I now write’.3 The surprise of the letter, of course, depends entirely on the accepted content of Wyatt’s poetry. Setting aside the general modern exoneration of Anne Boleyn, how to reconcile this damning letter with the content of critical poetry like ‘Who lyst his welthe and eas Retayne’ or even the less political, but still far from servile, ‘Who so list to hounte’? The answer lies in a complex negotiation of simultaneously separating and relating political, personal, and poetic strategies as understood and used by the Henrician court poet. In their invocations of traditional elements, Henry’s courtly makers sought to establish a precedent for their criticisms of, and reactions against, the King. These larger aims were largely shaped by the perceived decline of the King into despotism. The perception of having lost certain freedoms gave subjects, particularly educated subjects, a clear force against which to use their artistic energies. While these forces may have had little immediate effect on Henry’s increasing tyranny, the courtier-poets who found themselves limited to increasingly circumscribed techniques for criticism helped lay the groundwork for new ideological investments in certain modes of intellectual freedom. Even in claiming a certain privilege for the authority of the poet, most court poets acknowledge that the voice that tries to challenge the system functions from within, benefits from, and is tainted by that same system. The frustrations and anxiety of that complicity shape courtier poets’ work with traditional tropes and themes, with translation, and with transcription. We can trace these processes in each poet’s works and in manuscript collections—but the creation is not, as I identify in the introduction, of a self, a singular poetic voice. Rather, it is of voices which can convey the concerns, anxieties, and values of particular philosophical and sociopolitical positions. Each specific position is occupied by an individual, but the individual aims to express their position-taking efforts in changing social contexts, all subject to forces of communal meaning-making and interpretation. Shifting our framework to consider this work as position-taking moves us away from the reification of individual authorship to which the selffashioning terminology may have unintentionally contributed, allowing more space to consider both communal production and the prominence of women in court spaces for poetic production. Three threads come together: the extension of our understanding of verse as political position-taking, combined with the work on manuscript poetry that, f irst, emphasizes 3
Wyatt, ‘To Henry VIII’.
Conclusion
237
multiple authorship and, second, emphasizes form as part of verse’s action, both creates and is reinforced by a framework that restores women to their place in Henrician courtly verse work. To demasculinize this literary history is to better understand the politics of communally read, communally created position-taking verse. Beyond the manuscript, we must consider how this embedded positiontaking influenced choices in print transmission and the readers’ receptions of those choices. Those choices have directly underpinned our field’s canon formations, even as, recursively, the assumptions made in that canon formation have distorted our understandings of those choices. Megan Heffernan’s Making the Miscellany, released while I was revising this project, offers one excellent access point for the kind of reconsideration we must undertake; Heffernan brilliantly explicates the distorting weight of the miscellany as a concept in our field that does not align with, and so has often misrepresented, the work of form as understood by early modern readers. I would add to this sense of distortion the weight that early studies of Songes and Sonettes placed on Wyatt’s and Surrey’s work. We read Songes and Sonettes because of the prominence it was assigned early in the formation of our fields;4 we read Wyatt and Surrey because Tottel named them in Songes and Sonettes. Because of that naming, we have read their texts as single-authored and constructed Henrician verse as a masculine space, constructions that do not hold up to the fuller literary history. While the history of our own field then partially explains how Wyatt and Surrey became the most studied users of translation in the service of Henrician position-taking, that fuller literary history reveals that they are not alone or original in their effort to use translation as an expression of resistance. Indeed, the longstanding focus on these two men has obscured the roles of courtly women writers in creating a position-taking poetic space through translation and transcription. Women were among the most prominent collectors and transcribers in the manuscript culture of the Henrician court, and their creation and use of that culture bears particular scrutiny as a site of position-taking. The analysis of the centrality of women to production of texts in early modern England opens up several key elements to understanding the coding that took place in such texts, particularly complicating any attempt to simplify the Petrarchan love lyric as an expression of men’s frustration. I hope I have offered, here, one model for joining the attention paid to manuscript creation, the scholarly work to reintegrate women’s writing, and the work in early book history, in order to 4 Heffernan, Making the Miscellany, 19–20.
238
Gender and Position-Taking in Henrician Verse
interrogate both the codes of versified position-taking and what happened to that verse when it moved between distinct audiences. Particularly, my concern is in applying these frameworks to work before 1550, as a great deal of the existing scholarship on the relationships between manuscript and print production thus far has prioritized works from the latter half of the sixteenth century and from the seventeenth. Pulling that framework back particularly reveals a moment, in the early sixteenth century, where the functions of gender, manuscript and print conventions, and verse forms relate in different and sometimes less binary ways than we have previously understood. This work attempts to offer some response to the many and diverse calls to restore women’s writing to the early modern canon and, concomitantly, to re-read our literary histories through the historical presences revealed by that restoration. My hope is that such work provides a stronger foundation for future engagements with Margaret Douglas, Mary Shelton, Mary Fitzroy, and the other Henrician court poets whose contributions have been diminished because of their gender. Still, such engagement will need to contend with that tricky balance of assignment, to restore women’s roles, and understandings of a more communal mode of writing, to create more accurate literary histories and understandings of how literature gets made and read across cultures and time. Demasculinization of our literary canon must address both the erasure of women and the distortions of practice. I particularly hope that this work offers a contribution to frameworks for future engagements with these particular networks—helping us see the connective tissue between Mary Fitzroy, Henry Howard, Margaret Douglas, and Gawin Douglas, for example—and for analyses of these position-taking systems that consider the evolutions of particular verses as they transfer between hands and modes of production, as in ‘My ywtheffol days ar past’ which I here argue is first written down by Mary Shelton, but which accrues meaning and value as it transfers between manuscripts and into print. During a recent conference presentation, I was asked whether I believe any works in Songes and Sonettes are by women. I hope the work, here, on women’s place in manuscript writing, contextualizes and justifies my still-essentially-instinctive ‘Yes!’ By instinctive, though, I do not mean un- or ill-formed. On the one hand, I want to acknowledge the importance of proving that there is something more to that claim than instinct. That work will involve applying sociolinguistics and paleography to identify women’s hands, revisiting extensive archival material to see which poems show evidence of composition or revision from women’s hands, and intensive work in tracing manuscript and book histories to justify any claim about
Conclusion
239
how a particular verse might make its way through different production contexts. Moreover, we cannot stop at Songes and Sonettes without reifying the distorting energies I mention above. However, on the other hand, a core premise of this work is that we also, simultaneously, need to start from the assumptions of that ‘yes’—not just in Songes and Sonettes, but in every early print poetry collection —by understanding how central women were to communal, courtly verse production and by acknowledging how much material in courtly verse arose from models that were opposed to single authorship. While there is a contradiction, here, in the impulse to identify women’s writing and the need to better contextualize verse within communal forces of production, that contradiction is embodied in the Henrican verses considered throughout this work. Our better pedagogies will both integrate Douglas and Shelton as courtly verse producers working alongside the oft-studied Wyatt and Surrey and recontextualize all these individual authors as collaborative producers whose work was communally informed.5 The scope of such work underlines that our work must be every bit as communal as their work was. That this question arose about Tottel’s, specifically, offers a point of entry for my concern with Henrician verse. Henrician literature broadly merits greater attention than it sometimes attracts, precisely because of the same in-between-ness that may lead many a survey of British literatures to jump from Chaucer to Shakespeare. As an inflection point, Henrician print and manuscript cultures can offer us one answer for why Chaucer or Shakespeare, at all: from whence Chaucer’s reputation and its association with English-ness? How does print influence our view of the importance and authority of the writer? For what reasons, associated or heterogenous, did the broader field of literary studies anchor itself in misleading concepts of individual genius and single authorship? Because of its position alongside the rise of the printing press and a period of stabilization (just preceding a further expansion) of English monarchy, borders, and national identity, Henrician literature can reveal important sources for the rise of England’s nationalistic identity—and that identity’s association with masculinity, self-sufficiency, and certain kinds of literary endeavor—even as it reveals the counterpoint sources which allow us to trouble and de-stabilize these falsely phallogocentric literary histories. 5 Decades of work in Shakespeare studies, particularly, can offer us models for working with themes of attribution and collaboration coequally—and such work often speaks quite directly to the difficulty of working to restore these collaborative spaces while identifying drives towards representations of individual authorship (like Surrey’s).
240
Gender and Position-Taking in Henrician Verse
My hope is that the material here provides some platform for further re-working of our approaches to Henrician writing—and, particularly, for reconsidering how gender works in reading and writing networks. While this kind of work has been and is ongoing, the early sixteenth century offers an especially fruitful and thus far, I would argue, undertreated corpus that may predate the calcification of certain expectations about who wrote what kind of things, why, and what worth that conferred on writing. The continuing work in manuscript studies has already enabled extensive reconsiderations, like those I engaged here, of works traditionally attributed to Skelton, Wyatt, or Surrey—and, centrally, our approaches to attribution as a practice. My close reading work, throughout, demonstrates some of the re-reading that a new systemic framework enables. There is, though, much more work ahead to undo an inaccurate conflation of masculinity and power—and, particularly, of masculinity and poetic authority. That work will be archival, historiographical, pedagogical, and, so, necessarily, political, encouraging our scholarship practices to reveal and interrogate their own position-taking force.
Works Cited Clements, Robert J. and Lorna Levant, eds. Renaissance Letters: Revelations of a World Reborn. New York: New York UP, 1976. Heffernan, Megan. Making the Miscellany: Poetry, Print, and the History of the Book in Early Modern England. Philadelphia: U of Pennsylvania P, 2021. Wyatt, Thomas, Sir. ‘To Henry VIII’. 1536. In Clements and Levant, Renaissance Letters, 397–398.
Bibliography Anglo, Sydney. Spectacle, Pageantry, and Early Tudor Policy. Oxford: Clarendon Press, 1969. Baron, Helen. ‘Mary (Howard) Fitzroy’s Hand in the Devonshire Manuscript’. The Review of English Studies 45, no. 179 (1994): 318–335. Bates, Catherine, ed. A Companion to Renaissance Poetry. Hoboken, NJ: Wiley, 2018. Beal, Peter. In Praise of Scribes: Manuscripts and their Makers in Seventeenth-Century England. Oxford: Clarendon, 1998. Beilen, Elaine V., ed. Early Tudor Women Writers. Surrey: Ashgate, 2009. Beilen, Elaine V. Redeeming Eve: Women Writers of the Renaissance. Princeton UP: Princeton, NJ, 1987. Blanchard, W. Scott. ‘Skelton: The Voice of the Mob in Sanctuary’. In Herman, Rethinking, 123–144. Bourdieu, Pierre. The Field of Cultural Production. New York: Columbia UP, 1993. Brigden, Susan. New Worlds, Lost Worlds: The Rule of the Tudors, 1485–1603. New York: Penguin, 2002. Brigden, Susan. Thomas Wyatt: The Heart’s Forest. London: Faber and Faber, 2012. Burke, Victoria E. ‘Materiality and Form in the Seventeenth-Century Miscellanies of Anne Southwell, Elizabeth Hastings, and Jane Truesdale’. In English Manuscript Studies 1100–1700, Vol. 16: Manuscript Miscellanies c. 1450–1700, edited by Peter Beal and A.S.G. Edwards, 219–241. London: British Library, 2011. Burke, Victoria E. and Jonathan Gibson, eds. Early Modern Women’s Manuscript Writing. Burlington, VT: Ashgate, 2004. Burke, Victoria E. and Jonathan Gibson. ‘Introduction’. In Burke and Gibson, Manuscript Writing, 1–8. Butler, Judith. Gender Trouble. New York: Routledge Classics, 2006. Carlson, David R. ‘The Henrician Courtier Writing in Manuscript and Print: Wyatt, Surrey, Bryan, and Others’. In A Companion to Tudor Literature, edited by Kent Cartwright, 151–177. Malden, MA: Wiley-Blackwell, 2010. Carlson, David R. ‘Skelton, Garnesche, and Henry VIII: Revels and Erudition at Court’. The Review of English Studies 66, no. 274 (2015): 240–258. Castiglione, Baldesar. The Courtier. Translated by George Bull. New York: Penguin, 2003. Childs, Jessie. Henry VIII’s Last Victim: The Life and Times of Henry Howard, Earl of Surrey. New York: St. Martin’s, 2007. Clarke, Danielle. ‘Translation’. In The Cambridge Companion to Early Modern Women’s Writing, edited by Laura Lunger Knoppers, 167–180. Cambridge: Cambridge UP, 2009.
242
Gender and Position-Taking in Henrician Verse
Clarke, Elizabeth. ‘Women’s Manuscript Miscellanies in Early Modern England’. In Teaching Tudor and Stuart Women Writers, edited by Susanne Woods and Margaret P. Hannay, 52–60. New York: MLA, 2000. Clements, Robert J. and Lorna Levant, eds. Renaissance Letters: Revelations of a World Reborn. New York: New York UP, 1976. Clymer, Lorna. ‘Introduction’. In Ritual, Routine, and Regime: Repetition in Early Modern British and European Cultures, edited by Lorna Clymer. Toronto: U of Toronto P, 2006. Crane, Mary Thomas. Framing Authority: Sayings, Self, and Society in SixteenthCentury England. Princeton UP: Princeton, 1993. Crane, Mary Thomas. ‘Roman World, Egyptian Earth: Cognitive Difference and Empire in Shakespeare’s Antony and Cleopatra’. Comparative Drama 43 (2009): 1–17. Davidson, Peter. ‘Spatial Texts: Women as Devisers of Environments and Iconographies’. In Phillippy, History, 186–202. Day, Alexandra. ‘Literary Gifts: Performance and Collaboration in the Arundel/ Lumley Family Manuscripts’. In Pender, Gender, Authorship, 125–148. The Devonshire Manuscript. 1525?–1567? MS BL Add. 17492. British Lib., London. Dolven, Jeff. ‘Reading Wyatt for the Style’. Modern Philology 105, no. 1 (2007): 65–86. Domenichelli, Mario. ‘Sir Thomas Wyatt’s Translations from Petrarch and the “Lingua Cortigiana”’. English Studies in Italy 15, no. 1 (2002): 65–86. Donawerth, Jane. ‘Women’s Poetry and the Tudor-Stuart System of Gift Exchange’. In Women, Writing, and the Reproduction of Culture in Tudor and Stuart Britain, edited by Mary Burke, Jane Donawerth, Linda L. Dove, & Karen Nelson, 3–18. Syracuse, NY: Syracuse UP, 2000. Edwards, A.S.G. ‘Print and Manuscript: The Text and Canon of Surrey’s Lyric Verse’. In In the Prayse of Writing: Early Modern Manuscript Studies. Essays in Honor of Peter Beal, edited by S.P. Cerasano and Steven W. May, 25–43. London: British Library, 2012. Elizabeth I, trans. ‘Translation of Marguerite de Navarre’s Le Miroir de l’âme Pécheresse’. In Elizabeth I: Translations, 1544–1589, edited by Janel Mueller and Joshua Scodel, 40–128. Chicago: U of Chicago P, 2009. Erasmus. ‘To Henry VIII’. 9 September 1517. Letter VI-4. In Clements and Levant, Renaissance Letters, 255–257. Erasmus. ‘To Ulrich von Hutten’. 23 July 1517. Letter I-7. In Clements and Levant, Renaissance Letters, 20–25. Ezell, Margaret J. M. ‘Invisibility Politics: Aphra Behn, Esther Inglish and the Fortunes of Women’s Works’. In Phillippy, History, 27–45. Ezell, Margaret J.M. Social Authorship and the Advent of Print. Baltimore: Johns Hopkins UP, 1999.
Bibliogr aphy
243
Fortescue, John, Sir. On the Laws and Governance of England. Edited by Shelley Lockwood. Cambridge: Cambridge UP, 1997. Foucault, Michel. Discipline and Punish: The Birth of the Prison. Translated by Alan Sheridan. New York: Vintage Books, 1995. Frye, Susan. Pens and Needles: Women’s Textualities in Early Modern England. Philadelphia: U of Penn P, 2010. Gibson, Jonathan. ‘Katherine Parr, Princess Elizabeth, and the Crucified Christ’. In Burke and Gibson, Manuscript Writing, 33–49. Gibson, Jonathan. ‘Miscellanies in Manuscript and Print’. In Bates, Renaissance Poetry, 103–114. Goldberg, Jonathan. ‘The Female Pen: Writing as a Woman’. In Language Machines: Technologies of Literary and Cultural Production, edited by Jeffrey Masten, Peter Stallybrass, and Nancy J. Vickers, 17–38. New York: Routledge, 1997. Greenblatt, Stephen. Renaissance Self-Fashioning. Chicago: U of Chicago P, 1980. Greene, Roland. ‘Petrarchism among the Discourses of Imperialism’. In America in European Consciousness, 1493–1750, edited by Karen Ordahl Kupperman, 130–165. Chapel Hill: U of North Carolina P, 1995. Griffiths, Jane. ‘Having the Last Word: Manuscript, Print, and the Envoy in the Poetry of John Skelton’. In Pincombe and Shrank, Tudor Literature, 69–86. Griffiths, Jane. John Skelton and Poetic Authority: Defining the Liberty to Speak. New York: Oxford UP, 2006. Halasz, Alexandra. ‘Wyatt’s David’. Texas Studies in Literature and Language 30, no. 3 (1988): 320–344. Hall, Kim F. Things of Darkness: Economies of Race and Gender in Early Modern England. Ithaca: Cornell UP, 1995. Hamlin, Hannibal. Psalm Culture and Early Modern English Literature. Cambridge: Cambridge UP, 2004. Hannay, Margaret Patterson, ed. Silent but for the Word: Tudor Women as Patrons, Translators, and Writers of Religious Works. Kent: Kent State UP, 1985. Harrier, Richard. ‘A Printed Source for The Devonshire Manuscript’. The Review of English Studies 11 (1979): 179–194. Harris, William O. ‘“Love That Doth Raine”: Surrey’s Creative Imitation’. Modern Philology 66, no. 4 (1969): 298–305. Heale, Elizabeth. ‘“Desiring Women Writing”: Female Voices and Courtly “Balets” in some Early Tudor Manuscript Albums’. In Burke and Gibson, Manuscript Writing, 8–31. Heale, Elizabeth, ed. The Devonshire Manuscript: A Women’s Book of Courtly Poetry. Toronto: Center for Reformation and Renaissance Studies, 2012. Heale, Elizabeth. ‘Women and the Courtly Love Lyric: The Devonshire MS (BL Additional 17492)’. The Modern Language Review 90, no. 2 (1995): 296–313.
244
Gender and Position-Taking in Henrician Verse
Heffernan, Megan. Making the Miscellany: Poetry, Print, and the History of the Book in Early Modern England. Philadelphia: U of Pennsylvania P, 2021. Herman, Peter C., ed. Reading Monarchs Writing: The Poetry of Henry VIII, Mary Stuart, Elizabeth I, and James VI/I. Tempe, AZ: MRTS, 2002. Herman, Peter C., ed. Rethinking the Henrician Era: Essays on Early Tudor Texts and Contexts. Urbana: U of Illinois P, 1994. Herman, Peter C. Royal Poetrie: Monarchic Verse and the Political Imaginary of Early Modern England. Ithaca: Cornell UP, 2010. Herman, Peter C. and Ray G. Siemens. ‘Henry VIII and the Poetry of Politics’. In Herman, Reading Monarchs, 11–34. Herman, Peter C. and Ray G. Siemens. ‘Introduction’. In Herman, Reading Monarchs, 1–10. Hobbs, Mary. Early Seventeenth-Century Verse Miscellany Manuscripts. Aldershot: Scolar Press, 1992. Holton, Amanda and Tom MacFaul, eds. Tottel’s Miscellany. By Sir Thomas Wyatt; Henry Howard, Earl of Surrey; et al. New York: Penguin, 2011. Howard, Henry, Earl of Surrey, trans. The Aeneid. By Virgil. Edited by Florence H. Ridley. Berkeley: U of California P, 1963. Howard, Skiles. ‘“Ascending the Riche Mount”: Performing Hierarchy and Gender in the Henrician Masque’. In Herman, Rethinking, 16–39. Hunt, Alice. The Drama of Coronation. Cambridge: Cambridge UP, 2008. Irish, Bradley J. ‘Gender and Politics in the Henrician Court: The Douglas-Howard Lyrics in the Devonshire Manuscript (BL Add 17492)’. Renaissance Quarterly 64 (2011): 79–114. Irish, Bradley J. Emotion in the Tudor Court: Literature, History, and Early Modern Feeling. Evanston: Northwestern UP, 2018. Jeffrey, David Lyle. ‘Courtly Love and Christian Marriage: Chretien de Troyes, Chaucer, and Henry VIII’. Christianity and Literature 59, no. 3 (2010): 515–530. Kesselring, K. J. ‘No Greater Provocation? Adultery and the Mitigation of Murder in English Law’. Law and History Review 34, no. 1 (2016): 199–225. Kilgore, Robert. ‘The Politics of King David in Early Modern English Verse’. Studies in Philology 111, no. 3 (2014): 411–441. King, John N. ‘Henry VIII as David: The King’s Image and Reformation Politics’. In Herman, Rethinking, 78–91. King, John N. ‘Patronage and Piety: The Influence of Catherine Parr’. In Hannay, Silent, 43–60. King, John N. Tudor Royal Iconography: Literature and Art in an Age of Religious Crisis. Princeton: Princeton UP, 1989. King’oo, Claire Costley. ‘Rightful Penitence and the Publication of Wyatt’s Certayne Psalmes’. In Psalms in the Early Modern World, edited by Linda
Bibliogr aphy
245
Phyllis Austern, Kari Boyd McBride, and David L. Orvis, 155–174. Burlington, VT: Routledge, 2011. King’oo, Claire Costley. Miserere Mei: The Penitential Psalms in Late Medieval and Early Modern England. Notre Dame, Indiana: U of Notre Dame P, 2012. Legg, Leopold G. Wickham, ed. English Coronation Records. Westminster: Archibald Constable, 1901. Letters and Papers, Foreign and Domestic, Henry VIII, Volumes 1–21. Edited by J.S. Brewer, James Gairdner, and R.H. Brodie. London: HMSO, 1862–1920. Loades, David. Henry VIII: Court, Church, and Conflict. Kew: National Archives, 2007. Lockey, Brian. Law and Empire in English Renaissance Literature. New York: Cambridge UP, 2006. Loomba, Ania. ‘The Theatre and the Space of the Other in Antony and Cleopatra’. In Shakespeare’s Late Tragedies: A Collection of Critical Essays, edited by Susanne L. Wofford, 235–248. Upper Saddle River, NJ: Prentice Hall, 1996. Love, Harold. Scribal Production in Seventeenth-Century England. Oxford: Clarendon, 1993. Lumley, Jane, trans. Iphigeneia at Aulus. By Euripides. Edited by Harold H. Child. England: Chiswick Press, 1909. Marotti, Arthur F. ‘“Love Is Not Love”: Elizabethan Sonnet Sequences and the Social Order’. English Literary History 49, no. 2 (1982): 396–428. Marotti, Arthur F. Manuscript, Print, and the English Renaissance Lyric. Ithaca: Cornell UP, 1995. Marotti, Arthur F. and Steven W. May. ‘Manuscript Culture: Circulation and Transmission’. In Bates, Renaissance Poetry. 78–102. May, Steven W. ‘Popularizing Courtly Poetry: Tottel’s Miscellany and Its Progeny’. In Pincombe and Shrank, Tudor Literature, 418–433. Morley, Henry Parker, Lord. The Exposition and Declaration of the Psalme, Deus Ultionum Dominus. London: Thomas Berthelet, 1539. Mueller, Janel. ‘Complications of Intertextuality: John Fisher, Katherine Parr, and “The Book of the Crucifix”’. In Beilen, Women Writers, 75–108. Mueller, Janel. ‘Devotion as Difference: Intertextuality in Queen Katherine Parr’s Prayers or Meditations (1545)’. Huntington Library Quarterly 53, no. 3 (1990): 171–197. Mueller, Janel, ed. Katherine Parr: Complete Works and Correspondence. Chicago: U of Chicago P, 2011. Muir, Kenneth. Life and Letters of Sir Thomas Wyatt. Liverpool: Liverpool UP, 1963. Muir, Kenneth and Patricia Thomson, eds. Collected Poems of Sir Thomas Wyatt. Liverpool: Liverpool UP, 1969. Murphy, Beverly. Bastard Prince: Henry VIII’s Lost Son. Gloucestershire: The History Press, 2004.
246
Gender and Position-Taking in Henrician Verse
Murray, Molly. ‘The Prisoner, the Lover, and the Poet: The Devonshire Manuscript and Early Tudor Carcerality’. Renaissance and Reformation 35, no. 1 (2012): 17–41. Nevalainen, Terttu. ‘Gender Differences in the Evolution of Standard English: Evidence from the Corpus of Early English Correspondence’. Journal of English Linguistics 28, no. 1 (2000): 38–59. Nevalainen, Terttu and Helena Raumolin-Brunberg. Historical Sociolinguistics: Language Change in Tudor and Stuart England. New York: Pearson, 2003. Norbrook, David. Poetry and Politics in the English Renaissance. New York: Oxford UP, 2002. North, Marcy L. The Anonymous Renaissance: Cultures of Discretion in Tudor-Stuart England. Chicago: U of Chicago P, 2003. The Oxford Dictionary of National Biography Online. Oxford UP, 2015. Web. https:// www.oxforddnb.com/. Accessed 30 June 2022. The Oxford English Dictionary Online. Oxford UP, 2019. Web. https://www.oed. com/. Accessed 31 May 2021. Park, Jennifer. ‘Discandying Cleopatra: Preserving Cleopatra’s Infinite Variety in Shakespeare’s Antony and Cleopatra’. Studies in Philology 113 (2016): 595–633. Parr, Katherine. Prayers or Medytacions, Wherein the Mynd is Stirred, Paciently to Suffre All Afflictions […] London: Berthelet, 6 November 1545. Parr, Katherine. Psalms or Prayers Taken Out of Holye Scripture. London: Berthelet, 1544. Pender, Patricia, ed. Gender, Authorship, and Early Modern Women’s Collaboration. New York: Palgrave Macmillan, 2017. Phillippy, Patricia, ed. A History of Early Modern Women’s Writing. Cambridge: Cambridge UP, 2018. Phillippy, Patricia. ‘Introduction: Sparking Multiplicity’. In Phillippy, History, 1–24. Pincombe, Mike and Cathy Shrank, eds. The Oxford Handbook of Tudor Literature, 1485–1603. New York: Oxford UP, 2009. Pincombe, Mark and Cathy Shrank. ‘Prologue: The Travails of Tudor Literature’. In Pincombe and Shrank, Tudor Literature, 1–17. Powell, Jason. ‘Marginalia, Authorship, and Editing in the Manuscripts of Thomas Wyatt’s Verse’. In English Manuscript Studies 1100–1700, Vol. 15: Tudor Manuscripts 1485–1603, edited by Peter Beal and A.S.G. Edwards, 1–40. London: British Library, 2009. Powell, Jason. ‘Thomas Wyatt and Francis Bryan: Plainness and Dissimulation’. In Pincombe and Shrank, Tudor Literature, 187–202. Prescott, Anne Lake. ‘The 2011 Josephine Waters Bennett Lecture: From the Sheephook to the Scepter: The Ambiguities of David’s Rise to the Throne’. Renaissance Quarterly 65, no. 1 (2012): 1–30.
Bibliogr aphy
247
Prescott, Anne Lake. ‘The Pearl of the Valois and Elizabeth I: Marguerite de Navarre’s Miroir and Tudor England’. In Hannay, Silent, 61–76. Remley, Paul G. ‘Mary Shelton and Her Tudor Literary Milieu’. In Herman, Rethinking, 40–77. Richardson, David A. ‘Humanist Intent in Surrey’s Aeneid’. English Literary Renaissance 6 (1976): 204–219. Rupp, Susanne. ‘Performing Court: The Music of Henry VIII’. Journal for the Study of British Cultures 12, no. 1 (2005): 105–114. Sacks, Peter. ‘The Face of the Sonnet: Wyatt and Some Early Features of the Tradition’. In Green Thoughts, Green Shades: Essays by Contemporary Poets on the Early Modern Lyric, edited by Jonathan F.S. Post, 17–40. Berkeley: U of California P, 2002. Scarisbrick, J.J. Henry VIII. Berkeley: U of California P, 1968. Seneca. Phaedra. In Seneca: Tragedies, Volume I. Translated by John G. Fitch. Loeb Classical Library 62. Cambridge, MA: Harvard UP, 2018. Sessions, W.A. ‘Surrey’s Wyatt: Autumn 1542 and the New Poet’. In Herman, Rethinking, 168–192. Shirley, Christopher. ‘The Devonshire Manuscript: Reading Gender in the Henrician Court’. English Literary Renaissance 45, no. 1 (2015): 32–59. Siemens, Raymond G. ‘Henry VIII as Writer and Lyricist’. The Musical Quarterly 92, no. 1 (2009): 136–166. Siemens, Raymond G., ed. The Lyrics of the Henry VIII Manuscript. Grand Rapids, MI: English Renaissance Text Society, 2013. Simpson, James. ‘Breaking the Vacuum: Ricardian and Henrician Ovidianism’. Journal of Medieval and Early Modern Studies 29, no. 2 (1999): 325–355. Skelton, John. Certayne Bokes, Compyled by Mayster Skelton, Poet Laureat. London: Richard Lant, 1545. Skelton, John. The Complete English Poems. Ed. John Scattergood. New Haven: Yale UP, 1983. Snook, Edith. Women, Reading, and the Cultural Politics of Early Modern England. Aldershot, UK: Ashgate, 2005. Snyder, Susan. ‘Guilty Sisters: Marguerite de Navarre, Elizabeth of England, and the Miroir de l’âme Pécheresse’. Renaissance Quarterly 50, no. 2 (1997): 443–458. A Social Edition of the Devonshire Manuscript MS (BL Add. MS 17492). Wikibooks, The Free Textbook Project. https://en.wikibooks.org/w/index.php?title=The_Devonshire_ Manuscript& oldid= 3469218. Accessed 1 July 2022. Solomon, Deborah. ‘Representations of Lyric Intimacy in Manuscript and Print Versions of Wyatt’s “They flee from me”’. Modern Philology 111, no. 4 (2014): 668–682. Songes and Sonettes, Written by the Right Honorable Lorde Henry Howard Late Earle of Surrey, and Other. Ed. Richard Tottel. London: Tottel, 1557.
248
Gender and Position-Taking in Henrician Verse
Southall, Raymond. The Courtly Maker: An Essay on the Poetry of Wyatt and His Contemporaries. Oxford: Basil Blackwell, 1964. Southall, Raymond. ‘The Date of Wyatt’s Psalms’. English Studies 71, no.6 (1990): 496–500. Stemmler, Theo. ‘The Songs and Love-Letters of Henry VIII: On the Flexibility of Literary Genres’. In Henry VIII in History, Historiography, and Literature, edited by Uwe Baumann. Frankfurt: Peter Lang, 1992. 97–111. Stevenson, Jane and Peter Davidson. Early Modern Women Poets (1520–1700): An Anthology. Oxford: Oxford UP, 2001. Streitberger, W.R. Court Revels, 1485–1559. Toronto: U of Toronto P, 1994. Tadmor, Naomi. The Social Universe of the English Bible: Scripture, Society, and Culture in Early Modern England. Cambridge: Cambridge UP, 2010. Thomson, Patricia. ‘The First English Petrarchans’. Huntington Library Quarterly 22, no. 2 (1959): 85–105. Tudeau-Clayton, Margaret. ‘Supplementing the Aeneid in Early Modern England: Translation, Imitation, Commentary’. International Journal of the Classical Tradition 4, no. 4 (1998): 507–525. Tudeau-Clayton, Margaret. ‘What is My Nation?: Language, Verse, and Politics in Tudor Translations of Virgil’s Aeneid’. In Pincombe and Shrank, Tudor Literature, 389–403. Twombly, Robert G. ‘Wyatt’s Paraphrase of the Penitential Psalms of David’. Texas Studies in Literature and Language 12, no. 3 (1970): 345–380. Uman, Deborah. Women as Translators in Early Modern England. Newark, Delaware: U of Delaware P, 2012. Virgil. Eclogues. Georgics. Aeneid: Books 1–6. Translated by H. Rushton Fairclough. Loeb Classical Library 63. Cambridge, MA: Harvard UP, 1916. Walker, Greg. Writing under Tyranny: English Literature and the Henrician Reformation. Oxford: Oxford UP, 2005. Warner, Christopher J. The Making and Marketing of Tottel’s Miscellany, 1557: Songs and Sonnets in the Summer of the Martyr’s Fires. Burlington, VT: Ashgate, 2013. Weir, Allison. Henry VIII: The King and His Court. New York: Ballantine, 2001. White, Micheline. ‘Katherine Parr, Henry VIII, and Royal Literary Collaboration’. In Pender, Gender, Authorship, 23–46. White, Micheline. ‘The Psalms, War, and Royal Iconography: Katherine Parr’s Psalms or Prayers (1544) and Henry VIII as David’. Renaissance Studies 29, no. 4 (2015): 554–575. White, Micheline. ‘Women Writers and Literary-Religious Circles in the Elizabethan West Country: Anne Dowriche, Anne Lock Prowse, Anne Lock Moyle, Ursula Fulford, and Elizabeth Rous’. Modern Philology 103, no. 2 (2005): 187–214.
Bibliogr aphy
249
The Workes of Geffray Chaucer Newly Printed, with Dyuers Workes Whiche Were Neuer in Print Before. Ed. William Thynne. London: Thomas Godfray, 1532. Woudhuysen, H.R. Sir Philip Sidney and the Circulation of Manuscripts, 1558–1640. Oxford: Clarendon, 1996. Wyatt, Thomas, Sir. Certayne Psalmes Chosen out of the Psalter of David / Commonlye Called thee. VII Penytentiall Psalmes, Drawen into Englyshe Meter by Sir Thomas Wyat, wherunto is Added a Prolage of Auctore before Euery Psalme, Very Pleasaunt & Profttable to the Godly Reader. London: Thomas Rayuald and John Harryington, 1549. Wyatt, Thomas, Sir. Collected Poems of Sir Thomas Wyatt. Edited by Kenneth Muir and Patricia Thomson. Liverpool: Liverpool UP, 1969. Wyatt, Thomas, Sir. ‘To Henry VIII’. 1536. In Clements and Levant, Renaissance Letters, 397–398. Zaharia, Oana-Alis. ‘“De interpretatione recta…”: Early Modern Theories of Translation’. American, British, and Canadian Studies 23, no. 1 (2012): 5–24.
Index anonymity, uses of 16–17, 114, 165, 170–171, 176, 205 aristocracy see class or courtiers audience see readers authority 15, 23, 46, 49, 53, 63, 67–68, 88–89, 114, 122, 129, 131, 134, 139, 142, 143 and gender 23, 71–72, 79, 115, 126, 139–140, 142–143, 147, 148, 151, 157–159, 240 poetic and rhetorical 15, 47–48, 57, 63, 66–68, 71, 72–73, 79, 88, 95, 113–114, 126, 127, 138, 143, 147, 157–159, 170–171, 186, 227, 232, 236, 239–240 royal 15, 30, 34–36, 41–42, 45, 49, 62, 64, 78, 123, 131, 139–140, 143, 216 see also power authorship 13, 16, 47, 140, 143, 145, 166, 169, 172, 173 n. 45, 183–184, 196–197, 236–237, 239 Baron, Helen 10, 175, 186–187, 220 n. 86 canon formation 13–14, 24, 73 n. 54, 147, 173 n. 45, 189, 197, 237 masculinization in 13–14, 17, 25, 97 n. 43, 106 n. 87, 126, 144, 158, 184, 197, 237–238, 240 carceral imagery 69, 78, 93–94, 150–151, 206–207, 213–214, 223 Carlson, David 19, 30 n. 5, 35, 52–53, 55–56, 117 n. 13, 169, 170–171, 177, 202 Chaucer, Thynne edition of 10–12, 202, 212–214, 217–218, 221–223, 224–225, 229–231 class 36–37, 47–48, 55, 72, 87–89, 91, 93, 96–98, 110, 118, 120, 124–125, 126, 137, 142 and language use 96–99, 103, 116, 159, 181–182 codes 16–18, 22, 63, 88–89, 102, 159, 232 poetic 10, 15, 16, 17, 20, 22–23, 24–25, 38, 40, 46, 57–58, 63, 67–68, 74–76, 77, 83–84, 85–87, 88–89, 99, 102, 109–110, 114, 159, 164, 166, 197, 204–205, 232, 237–238 social and courtly 14, 16, 17, 30, 31, 32–33, 34, 35, 43, 45, 52–53, 63, 72, 75, 79, 89, 91–92, 98, 101, 153 n. 149, 232 see also tradition; translation; transcription communal verse production 12–13, 18–19, 23, 25, 35, 73 n. 55, 73–74, 92, 97 n. 43, 98–99, 102–103, 114, 126, 138, 164–167, 183, 184, 189, 194, 196–197, 205–206, 232, 236–237, 238, 239 rejection or avoidance of 23, 35, 74–75, 79, 99, 102–103, 126, 159 context and courtly performance 9–10, 91–92, 150, 176, 187, 195–196 in manuscript writing and placement 10– 13, 16–17, 18, 24, 140, 145, 164–169, 177, 181,
184–185, 187–188, 189, 189–194, 194–196, 202–203, 205, 206–223, 224–231 of print vs. manuscript production 73–74, 79, 86–89, 94, 94–95, 185 n. 90 of verse at court 9–10, 15, 16, 46, 57–58, 68–70, 74 n. 62, 78, 88, 91–92, 92–93, 96, 109, 146 n. 119, 174, 184, 204–205, 205–206, 223–224 courtiers 10, 11, 16, 24, 57–58, 74–75, 85–86, 89, 92, 98–99, 110, 166, 169, 175, 205, 236 and critique or counsel 14, 36–37, 38–39, 42, 43, 45, 46, 49, 51–53, 56, 57, 62–63, 67–68, 71, 73–74, 84, 85, 114–115, 127–128, 129, 130, 132, 146 and service 30, 39, 50, 68–70, 92–93, 101, 108, 176, 193–194, 196, 207–208, 214, 217, 221, 228 resentment of 12, 15, 32–33, 39–40, 71–72, 76, 77–78, 84–85, 88, 93–94, 95–96, 96–97, 103, 107, 108–109, 177, 182–183, 183, 185, 208, 224, 230–231 traditional privileges of 20, 22, 32–33, 45, 45–47, 48, 62–63, 84, 119 n. 19, 127, 201–202, 210–211, 213, 230, 231–232 courtly love lyric 12, 21–22, 22–23, 34, 47, 50, 51, 67–68, 88–89, 109, 194, 221, 231 and chivalry 35, 36–37, 39–40, 45, 57, 78 and service 22, 32–33, 40, 67–68, 84–85, 92–94, 94–95, 102, 110, 182–183, 185, 202, 207–208, 210–211, 214, 228, 230, 232, 237–238 court masques 29–30, 32–33, 33–34, 35, 64 David (biblical king) 63–64, 73, 127–138 and Henry VIII 22, 63–67, 127, 128–129, 129–137, 139 and poetic authority 127–128, 129, 131, 136, 138 see also Parker, Henry (Lord Morley), Exposition and Declaration of the Psalme; tradition, of biblical reference; Wyatt, Thomas, Certayne Psalmes defense of women 11–12, 120–121, 194, 229–230, 231–232 deniability 16, 23–24, 48–49, 57, 70–71, 85–86, 88, 144, 164, 165, 166, 171, 176, 184, 187, 196, 197, 205, 232 rejection or avoidance of 66–67, 68, 75, 79, 205 Devonshire Manuscript 10–13, 14, 16–20, 21 n. 35, 22, 23–24, 73 n. 55, 78, 79, 85, 89 n. 17, 106–109, 158, 163–171, 171–183, 183–189, 189–197, 201–206, 206–223, 223–232 A Social Edition 10–11 n. 6, 20, 106, 108 n. 98, 166 n. 16, 184 n. 87, 185 n. 94, 191 n. 114
252
Gender and Position-Taking in Henrician Verse
& 115, 202 n. 5, 203 n. 6, 214, 218, 220 n. 86, 223 n. 99, 224 see also Douglas-Howard exchange dialogues (poetic) see exchanges, poetic or literary Douglas-Howard exchange 10, 195–196, 201–202, 204, 205–206, 206–223 Douglas-Howard marriage 9–10, 78, 171, 174, 186 n. 95, 195, 201–202, 205–206, 208, 214, 223–224, 225, 226, 231 Douglas, Margaret 13, 15, 185–186, 211 n. 49, 223, 238 as writer 10, 24, 184, 194–196, 197, 202, 206, 207 n. 27, 206–208, 211, 212–214, 217, 218–220, 223–224, 231–232, 239 attributions to 208, 212, 214, 218 biographical details 9–10, 62, 140, 185–186, 221 Elizabeth I as writer 23, 115, 144–146, 154, 158, 159, 163–164 biographical details 148–149, 150–151, 153 Miroir 145, 146–153 exchanges, poetic or literary 11, 24, 52–56, 94, 143, 146–147, 153 n. 149, 154 n. 150 & 153, 165, 166 n. 17, 167, 189–194, 194–196, 201, 202, 204, 205–206, 206–223, 232 feminization 20–21, 39–40, 84–85, 90–91, 99, 115 n. 5, 122 n. 28, 123, 141–143, 147, 159, 188, 208 Fitzroy, Henry 9, 118–119, 185–186 Fitzroy, Mary 13, 185–186, 238 as writer 175, 186–188 attributions to 175, 186 biographical details 9 n. 1, 186 Fortescue, John 45–46, 62 gender 10–13, 13–14, 15, 16 –18, 19, 20–21, 23, 25, 32–33, 34, 36, 39–40, 54–55, 57–58, 65, 75, 84–85, 90–91, 96, 98–99, 104–105, 106 n. 87, 107, 108–109, 114–115, 117–125, 126, 138–139, 140–143, 144–145, 146–149, 150–153, 154–158, 158–159, 163–164, 165–166, 168–169, 171, 175, 180–181, 182–183, 183–185, 185–186, 188, 191, 194, 196, 197, 201–202, 208, 214, 218–219, 224, 225–231, 236–240 grief 74–75, 76–77, 90–91, 118–119, 157, 183, 188, 194–196 Heale, Elizabeth 10, 11, 20, 21 n. 35 & 36, 106–107, 108 n. 98, 166 n. 16, 175 n. 57, 178 n. 65, 183, 184 n. 87, 192 n. 115, 202–203, 208, 211 n. 49, 216 n. 69, 218 n. 75, 219, 220 n. 86, 223–224 Henry VIII as king 9–10, 14, 22, 29–30, 30–32, 33–34, 40–41, 44, 45, 48, 53, 56–57, 57–58,
61–62, 66–67, 72, 75, 84, 95, 97–98, 101 n. 64, 185–186, 201–202, 235, 236 as writer 15, 19, 21–22, 33, 33–36, 36–45, 67–68, 79, 139–140, 159, 231, 232 attributions to 34–35, 37 n. 49 comparisons to Caesar 74, 75, 89–90, 91 comparisons to David see David (biblical king), and Henry VIII poetry about and literary references to 12, 109, 48–52, 52–53, 64–66, 68–71, 71, 73–74, 75–77, 77–79, 89–92, 93–94, 95–96, 103–104, 107, 108—109, 123, 126, 127–137, 139, 141–142, 146, 150–153, 186, 195–196, 208–211, 213–214, 216–217, 220–221, 221–223, 223–224, 225–227, 227–228, 228–230, 231–232 Herman, Peter 19, 21, 30 n. 5, 31, 33–34, 35–36, 37, 41, 45, 84 n. 1 Howard, Henry, see Surrey, Earl of (Henry Howard) Howard, Thomas 62, 196, 206 as writer 24, 201–202, 206, 207, 208–211, 214–217, 220–221, 221–223 attributions to 10–11, 203, 225 biographical details 9–10 references to 78–79, 202, 217, 219–220, 224, 225, 230 humanism 14, 24, 31–32, 35, 49–50, 52, 64, 84, 89, 97–98, 101, 115 n. 5, 116–117, 127, 130, 154 n. 153, 229 Irish, Bradley 11–12, 20, 119 n. 19, 167, 202–203, 207, 209, 216, 220, 221–222, 223, 224–225 kingship 22, 29–30, 30–31, 31–32, 33, 35–36, 36–37, 38, 39, 40, 41, 43, 45–46, 56, 57, 62–63, 63–68, 68–70, 72, 73–74, 75–77, 92, 101, 123, 125–126, 127–138, 139–140, 151–152, 210–211 language 16, 32 n. 20, 49–52, 56–57, 71, 85–86, 90, 96–104, 106, 116–118, 141, 143, 145–146, 148–152, 156–157, 180–181, 208, 215–217, 228, 230 law, English 30–31, 43, 45–46, 56–57, 62–63, 84 n. 2, 101, 124, 208, 209–210, 214, 215–216, 217–218 Henry VIII’s modifications to 9–10, 14, 30–31, 33–34, 35, 45, 62 n. 7, 72, 79, 151–152, 185–186 love 117, 171–172, 174, 176, 179–180, 226–227, 228–229 and gender 39–40, 75–76, 84–85, 103–105, 108–109, 141–143, 180 n. 76, 182–183, 188, 191, 194, 230–231 and service 10–13, 17, 22–23, 32, 50–51, 67, 77–78, 84–86, 88, 94, 95, 100–101, 107–108, 108–109, 110, 182–183, 185, 194, 228, 230 and youth 35–39, 40, 45, 57
Index
familial 146–149, 150–153 marital 9–10, 34, 123–124, 141–143, 152–153, 155–157, 195–196, 202, 206–223, 223–224, 226, 231–232 of God 39, 70 n. 38, 132–133, 141–143, 147, 149–153, 185 Lumley, Jane as writer 23, 113, 115, 144–146, 153–158, 164 biographical details 145–146, 154 Iphigeneia 145–146, 153–158 manuscript writing 23–24, 35, 57–58, 66–67, 73–74, 77, 99, 102, 110, 117, 124, 126, 154, 156, 164–167, 172–173, 177, 183, 189–190, 205–206, 213, 222, 225, 232, 236–237 materiality of 10, 12, 16–19, 34–35, 73, 89 n. 17, 108 n. 101, 127, 154 n. 150, 167–170, 173, 177, 181–182, 184–185, 185–186, 186–188, 189–194, 202, 204–205, 232 relationship to print 13–14, 18, 19–20, 21, 24, 74–75, 79, 86–89, 106, 115, 124, 138–139, 140, 143, 159, 170, 172, 184, 185 n. 90, 237–239 roles of women in 13–14, 17–18, 19, 20–21, 46, 76, 85, 98, 109, 115, 138–139, 140, 143, 144–145, 147, 154, 158–159, 163–164, 165–167, 168, 182 n. 83, 184, 197, 237–239 see also Devonshire Manuscript Marotti, Arthur 17–18, 73 n. 54, 85 n. 3, 87, 88–89, 95 n. 38, 164–165, 167, 189 masculinity 14, 17, 20, 21, 23, 32–33, 36, 38–40, 54–55, 57, 75, 85, 90–91, 99, 104–105, 115, 117, 120–122, 126, 144, 150, 158, 188, 229, 239–240 and martial imagery/military action 42, 53–54, 64 n. 17, 75–76, 78, 91, 92–93, 104–105, 118–120, 139, 188 nationalism, English 34, 45–46, 53–55, 63, 66, 90–91, 101, 102 n. 67, 116, 123, 126, 239 New Historicism 12–13, 17, 25, 163, 236–237 Parker, Henry (Lord Morley) biographical details 64 n. 19 Exposition and Declaration of the Psalme 64–66, 230 n. 123 Parr, Katherine as queen 139–141, 142, 143, 145, 146–147, 148, 168 as writer 20 n. 32 & 33, 23, 63 n. 11, 66, 70 n. 38, 114, 115, 138–143, 144, 145, 147, 148, 154, 158, 159 Prayers or Medytacions 115, 140, 141–143 Psalms or Prayers 115, 139–140, 143 performance 20, 29–31, 32–33, 33–34, 35–36, 37, 39–40, 50–51, 52–53, 56, 57, 61, 63, 70, 72, 74–75, 84–85, 88, 89, 90–92, 94, 98, 100–102, 104, 107, 114–115, 117, 130–131, 135–136, 138–139, 143, 150, 157–158, 172–173, 175, 187, 188, 195–196, 197, 201–202, 207–208, 209
253 Petrarch, translations of 22, 50, 84, 85–86, 88 n. 14, 93–94, 96–105, 110, 113, 126, 206, 237 placement see context, in manuscript writing and placement position-taking 10, 12–13, 14, 15–16, 17, 19, 20, 21–25, 36, 40, 46–48, 49, 52, 57–58, 62–63, 66–68, 71, 73–74, 75, 76, 77, 79, 84–86, 91–93, 96–99, 103, 109–110, 113–115, 116–117, 126, 139–140, 141, 143, 145, 146 n. 119, 147, 154 n. 150, 155, 158, 158–159, 163–167, 168–169, 171, 182 n. 83, 187, 188, 196–197, 201–202, 204–206, 213, 228, 232, 236–238, 240 power 23, 25, 34, 36, 40, 41, 86, 105, 114–115, 116–117, 149–150, 152, 176, 192, 216 and gender 20–21, 23, 34, 36, 39–40, 57, 84–85, 96, 104–105, 115, 138–139, 141–142, 146, 147, 148, 154, 158–159, 180 n. 76, 182–183, 188, 193–194, 197, 214, 224, 231, 240 cultural 15, 33, 46, 47, 56, 57–58, 67–68, 71, 114, 116–117, 129, 139–140, 147, 148, 167, 170–171, 218, 240 political 15, 20, 22, 32, 40, 42, 43, 45, 48, 49, 54, 57, 69, 77, 91, 93, 105, 120, 140, 171, 185, 211, 213, 224 royal 15, 22, 30–32, 33–34, 35, 36–37, 38, 40–42, 43, 44–45, 48, 57, 62–63, 70, 72, 85, 93, 98, 101, 134–135, 151–152, 202, 230 see also authority Psalms, Penitential in early modern culture 64–65, 66, 70 n. 38, 114, 126–128, 132 n. 69, 137–138, 139, 153 see also Wyatt, Thomas, Certayne Psalmes queenship 34, 70, 109, 124, 125–126, 139–142, 142–143, 147, 153, 168, 228–229, 231 readers 11, 13, 15–16, 18, 19–20, 23, 24, 25, 43–44, 47 n. 85, 49, 67, 69, 78, 79, 84, 90–92, 96, 99, 102, 108–109, 109–110, 117, 129, 134, 138, 139, 140, 141, 142–143, 148, 151, 152, 164–165, 170, 173, 174, 175–176, 189, 191, 192–194, 205, 206, 211, 226–227, 228–229, 231, 232, 237–238 courtly vs. public 19–20, 24, 79, 87–89, 109–110, 185 n. 90, 189, 238 Remley, Paul 10, 11, 20, 202, 203–204, 204–206, 210, 211–212, 214, 218, 219–220, 225, 230 repetition 18, 42, 44, 51, 73, 84, 102, 108, 122 n. 28, 125, 130, 150–151, 155–157, 172–173, 177, 180 n. 76, 184, 189–194, 196, 207, 213, 214, 217, 220, 223 Shelton, Mary 13, 238 as writer 175, 183–185, 187, 191, 193–194, 194–196, 197, 205–206, 207 n. 27, 212, 216, 225–231, 239 attributions to 11, 183–184, 187, 191, 203–204, 189–190, 203–204, 211–212, 218, 225, 238
254
Gender and Position-Taking in Henrician Verse
Siemens, Raymond 19, 21, 30 n. 5, 33–35, 37, 40, 45, 67 Skelton, John 15, 19, 139, 240 ‘Against Garnesche’ 52–56 as poet 46–48, 48–57, 57–58, 126 biographical details 47–48, 55–56 ‘Speke Parrot’ 48–52 spelling 18, 51, 107, 124, 126, 173–174, 174–176, 179–181, 184, 186–188, 192–193, 197, 205, 212, 216, 221, 222, 230–231 Surrey, Earl of (Henry Howard) 35, 39, 46–47, 66–67, 71, 71–72, 96–97, 98, 110, 116, 129, 138 Aenied, translation of 23, 24, 113–114, 114–115, 115–126, 137, 147 as writer 20, 22, 46–47, 54, 67, 73–75, 75–77, 77–79, 85, 97–99, 99–101, 103–105, 126, 139, 143, 144, 159, 180 n. 76, 186, 188, 239–240 biographical details 9 n. 1, 57, 71, 72–73, 77–78, 115–116, 117–118, 120, 126 n. 42 in Tottel 73–74, 79, 86–89, 99–100, 109, 170, 237 surveillance 37, 130, 171–174, 175–176, 177, 201 Tottel’s (Songes and Sonettes) 13–14, 18, 19–20, 24, 73–74, 79, 86–89, 94, 95, 102, 170, 184–185, 237, 238–239 titles in 79, 87–89, 92, 94, 94–95, 99, 106, 109, 170 tradition 10, 11, 12–13, 15, 16, 19, 31, 32–33, 40, 41, 45, 46, 48–49, 52–53, 57–58, 63, 75, 79, 90, 117, 142, 145, 147–149, 153–154, 202, 210–211, 214, 222, 231–232, 236 and the courtly love lyric 21–22, 22–23, 33–35, 35–36, 45, 50, 77, 84, 85–86, 88, 92, 95, 101, 107, 182, 194, 206, 210, 229 see also courtly love lyric of biblical reference 22, 64–68, 113–114, 126–128, 130, 137–138, 139 see also David (biblical king); Parker, Henry (Lord Morley), Exposition and Declaration of the Psalme; Wyatt, Thomas, Certayne Psalmes of Classical reference 22, 68–70, 73–74, 113–114, 116–117, 119–120, 126, 157–158 see also Henry VIII, comparisons to Caesar; Lumley, Jane, Iphigeneia; Surrey, Earl of (Henry Howard), Aeneid, translation of
transcription 10–13, 16, 19, 22, 23–24, 79, 85, 106–107, 117, 140, 159, 163–164, 165–171, 172–173, 174–175, 181–182, 183, 186–188, 189, 194–195, 205–206, 211, 212, 221, 222, 225–228, 228–231, 236, 237 alterations in 10–12, 23, 101–102, 140, 166, 169–170, 177–181, 190–191, 222–223, 230 see also Chaucer, Thynne edition of; context, in manuscript writing and placement; Devonshire Manuscript; manuscript writing translation 10, 12–13, 16, 19, 22–23, 24, 67, 68, 70, 77, 79, 83–84, 84–86, 96, 110, 113–115, 115–117, 117–118, 120, 121 n. 23, 123–125, 126, 139–140, 144–146, 147, 150, 151–152, 153, 153 n. 147, 154, 157–158, 158–159, 164–165, 166, 236, 237 biblical 23, 62, 73, 114, 126–128, 130, 137–138 of courtly love lyrics 22–23, 50, 84–86, 88 n. 14, 92, 92–93, 99–105, 106, 108, 177 see also Elizabeth I, Miroir; Lumley, Jane, Iphigeneia; Petrarch, translations of; Surrey, Earl of (Henry Howard), Aeneid, translation of; Wyatt, Thomas, Certayne Psalmes Walker, Greg 14, 30 n. 5, 31–32, 47–49, 61–63, 66–67, 71, 71–73, 75, 77, 84 n. 2, 93 n. 30, 98 n. 48, 126 n. 42, 129 n. 58, 130 n. 62, 134 White, Micheline 20 n. 33, 63 n. 11, 64 n. 18, 65 n. 20, 66, 138, 139, 143, 168 Wyatt, Thomas 13, 19, 35, 39, 46, 130 as writer 20, 22, 23, 66–67, 68–71, 77, 85, 89–92, 92–94, 94–96, 96–99, 100–105, 109, 115, 144, 169, 236, 239 attributions to 106, 108–109, 129, 171 n. 40, 171–172, 173 n. 45, 177, 184, 190, 191 n. 113, 194, 240 biographical details 68 n. 34, 93, 98, 235–236 Certayne Psalmes 23, 70 n. 38, 113–114, 114–115, 126–138, 139, 147, 153 in Tottel 73–74, 86–89, 92, 94–95, 99–100, 106, 170, 237 youth 29, 35–37, 38, 40, 40–42, 43–45, 50–51, 57, 69, 122–123