Studies in Variation, Contacts and Change in English
Tudor women writing: Multimodal style and identity in the English letters and prose of Queen Katherine Parr and Princess Elizabeth
This paper investigates how the stylistic choices of sixteenth-century female writers can be understood as an act of identity when read in their local and public contexts. Focusing on the mid-sixteenth-century writings of two elite women, Princess Elizabeth, and her stepmother, Queen Katherine Parr, I investigate the extent to which each woman develops a distinctive “voice” for their religious prose in contrast to their vernacular epistolary writing, and how such stylistic efforts can be interpreted within the contemporary constraints of permitted female literary practice and vernacular language norms. The analysis uses the sociolinguistic concept of style to explore how features combine to achieve particular stylistic goals, as realised within written documents. Material features (such as layout and ink colours) combine with morphosyntactic properties to suggest that each woman had an idea of a specific vernacular style for her religious prose writing, as opposed to her correspondence, and that Parr and Elizabeth shared a similar concept of what this style should constitute. The article concludes by considering how quantitative stylometric findings cohere with the fine-grained analyses, and identifies similarities between Parr’s and Elizabeth’s prose and Coverdale’s English Bible, suggesting its stylistic influence on their work.
The opening dedication of Princess Elizabeth’s 1544 New Year’s gift to her stepmother Queen Katherine Parr illustrates how material and linguistic resources intersect in the construction of identity in the written mode. She writes: “TO OUR MOSTE NOBLE AND virtuous quene KATHERIN, Elizabeth her humble daughter wisheth perpetuall felicitie and euerlastyng ioye”. The eleven-year old princess manipulates her hand-written script to visually foreground the superlative address, mimicking conventions of print (cf. Ellis 2009: 166). Her words are inscribed in a humanist italic hand that showcases the contemporary, continental nature of her education. The language is informed by rhetorical principles of the day, including superlative address (“moste noble”), diminutive self-reference (“humble daughter”) and synonymia with alternating Latinate and Germanic noun phrases (“perpetuall felicitie and euerlastyng ioye”) (Adamson 1999: 547). She further articulates her relationship with the recipient, Katherine Parr, through a pronoun shift, switching from an inclusive first-personal plural (“our…quene”) to the distancing and deferential third-person self-reference (“her…daughter”). And, at the morphosyntactic level, she uses the third-person singular verb ending –eth (“wisheth”), rather than the innovative (less formal) –es ending. In combination, such features convey her knowledge of socially-appropriate communicative conventions, and, therefore, associated character traits; that is, she presents herself a modest, literate and deferential daughter.
Of course, Elizabeth would become one of England’s best-known and longest-reigning monarchs, challenging social conventions by remaining unmarried until her death in 1603. Yet, when writing her 1544 dedication, Elizabeth was newly placed on the European marriage market, following her reinstatement as a legitimate child of Henry VIII. As the king’s youngest daughter, it was imperative that she secured her status and future within the royal court, and it is in this context that her careful dedication should be read. In producing the dedication, as part of an English translation, Elizabeth was participating in a newly emerging set of cultural practices associated with Humanism, Protestantism, gendered literacies, and the emergence of English as a written, literary language, and therefore signalling her legitimacy within such an elite cultural community. Queen Katherine Parr was no accidental recipient, either. Parr played a considerable role in the advancement of the English-language Reformation in the mid-sixteenth century, in part through the publication of her own religious, vernacular writings (e.g. Lamentation of a Sinner, 1547) and gift manuscripts (e.g. Prayers, or Meditations c.1545); works that similarly operated to establish Henry VIII’s sixth wife as a legitimate figure at the royal Court (Mueller 2011: 171).
Previous discussions of the writings of both Elizabeth and Parr have focussed on the content and context of their religious prose. Underpinning this scholarship is an attempt to read the identities of these women through their writing, whether in the expression of parental anxieties on the part of the motherless Elizabeth (Prescott 1985), or the motivation to assert royal authority in the case of Katherine Parr (Mueller 2011: 171). The present investigation develops this interest in their writing and identity using a sociolinguistic and stylistic lens. The analysis explores how a linguistic analysis, using the sociolinguistic concept of style (Coupland 2008), can enrich our understanding of how these women used the linguistic and material resources of mid-sixteenth-century written English for identity work (defined by Beech as the process through which “images and representations…become imbued with meaning […] taken as being part of one’s identity”, as perceived by the individual and third-parties (2008: 52); for a survey of the application of “identity” in sociolinguistic scholarship, more specifically, see Edwards 2009: Chapter 2). The “rise” of the English language in this period, newly associated with religious, literary and Humanist enterprise, gave the vernacular new significance in written communicative contexts (Adamson 1999: 543–4), both in terms of what it was used for, and how it was used.
Contemporary sociolinguistics (e.g. Coupland 2008, Eckert 2000) emphasises that the association between a linguistic form and social meaning is fluid and contextual, and that it is with repertoires of features, style, rather than singular forms, that identity work is performed. Working analogously, this study refocuses the investigation of style to consider historical written language, encompassing material and linguistic features across different levels of the language system, to interrogate the writings of Queen Katherine Parr and Princess Elizabeth. There are three interrelated aspects of inquiry. Firstly, what are the stylistic properties of the epistolary and religious prose texts produced by Parr and Elizabeth in the mid-sixteenth century? Secondly, how do the linguistic features identified as style markers for each writer relate to contemporary macro-level genre and sociolinguistic norms? And, thirdly, how might we interpret these similarities and differences in style from the perspective of identity?
The concept of style has proved highly valuable for our understanding of how linguistic variation intersects with identity construction, and the relationship between synchronic language acts and diachronic trajectories of change (Eckert 2000). Present-day investigations generally focus on spoken language which entails that, for scholars interested in style and identity from historical perspectives, the frameworks require augmentation to account for the differences in data (i.e. written texts), as well as – to the best of our knowledge – different social systems in which the language is used. The disciplines of historical sociolinguistics, pragmatics and stylistics have developed precisely to accommodate these different theoretical and practical challenges.
Macro-level inquiries within these fields generally focus on the register (genre) and sociolinguistic distribution of specific linguistic constructions, e.g. the studies of the third-person singular verb ending –eth and –s undertaken by Kytö (1993), Rissanen (1991), Nevalainen and Raumolin-Brunberg (2003), and Gries and Hilpert (2010), among others. However, investigations of language and identity necessitate a finer-grained, micro-level analysis of specific individuals and communicative contexts (e.g. Raumolin-Brunberg 2005; Evans 2015), and such work has indicated how present-day studies of style find their equivalents in the analysis of repertoires of written communication; for instance, in the linguistic and material features of Early Modern correspondence (Williams 2014).
In Early Modern England, the stratification of education and literacy entailed that acts of writing were salient and socially meaningful. This encompassed the language used, the stylistic choices for that language, and the genres in which the writing was produced. To be “literate” generally meant one knowledgeable of Latin, not “merely” someone who could work in the English vernacular (Ferguson 2003: 16), although the expansion of English in form and function challenged this concept. Rhetoric was central to this new fashioning of English, with the classical system of oratory repurposed for written modes, including letter-writing, treatises and poetry (see Adamson 1999). One consequence was an anxiety over the capabilities of English to achieve the necessary eloquence for these higher-status social functions, with consternation over the etymological origins of lexis (known as the inkhorn controversy; see Nevalainen 1999), concern for the variability of orthography (Salmon 1999), and the pressing of English syntax into rhetoric structures developed for Latin grammar, such as the periodic sentence (Adamson 1999: 583). As Morini (2013) has shown in his analysis of etymological choices in translations of Virgil’s Aeneid, a writer’s language and style choices signalled more than a desire to communicate meaning. It would state their ideological standing in regards to contemporary debates of English language, and, by implication, their ability to participate within such debates.
An important dimension for the study of style in Early Modern English, and especially relevant to the present investigation, is gender. Sociolinguistically, gender is a significant factor in language change in both modern and historical periods. Women have been found to adhere to standard language variants, as well as leading in the uptake of innovative forms (what Labov terms the gender paradox (1990)). In historical periods, the absence of a supralocalized standard language entails that this so-called universal is difficult to evaluate, although the hypothesised establishment of localised language norms (Bergs 2005) is both plausible and useful (Evans 2015). Women show a tendency to innovate in language change from below, with changes from above the preserve of their male contemporaries – a trend attributed to the different educational opportunities and language usage (that is, writing experience) for men and women in the period (Nevalainen and Raumolin-Brunberg 2003: 130–1). The relationship between gender and the patterns of linguistic variation and change illustrates how linguistic style at an individual level must span both conscious and unconscious language practices, which are inevitably bound to macro-level cultural norms, values and structures.
The link between social structures and language use is perhaps most acute in relation to literacy. In the sixteenth century, opportunities for literacy, which would permit the production of written language, were shaped by cultural values of gender. Education was functional, not aspirational (Pollock 1989), and “reinforced existing social distinctions” (Clarke 2001: 19). Hence, many women would receive tuition in domestic concerns such as needlework, rather than languages. Yet, some educationalists advocated female education to keep their purportedly “weaker minds” occupied. This resulted in a paradoxical situation that women granted access to knowledge for their betterment could potentially disrupt the distribution of gendered social roles (Clarke 2001: 19); a situation exacerbated by the rise in humanist thinking which privileged individual access to knowledge. A stylistic analysis of female-authored texts therefore needs to recognise the conventions governing language use, and the expectations (pressures or constraints) placed on textual production.
Such conventions include the choice of language itself. The vernacular was generally the preserve of women, following Cicero’s De Oratore, whereas classical and European languages were male domains; the gendering of linguistic codes consequently imposed a public/private communication divide along similarly gendered lines (Clarke 2001: 29–30; Ferguson 2003: 4). Another, related division is evident in the aesthetic appearance of written language. The new, continental italic script was initially deemed suitable for women, with secretary hand conceived as its masculine counterpart (Goldberg 1997). Genres were also gendered, with some deemed more acceptable for female authorship than others. The acceptable spaces for female writing were, therefore, quite carefully delineated. This is relevant to the works of Parr and Elizabeth, each of whom produced texts within the culturally-accepted parameters; namely, English-language translations, devotional writing, and correspondence.
With regards to translation, the widely-held (or, at least, most vocal) view of Early Modern women’s translation at the time was that it was “a mechanical exercise, one that would occupy the mind and body much as embroidery did” (Morini 2006: 22). Following their Early Modern predecessors, many modern scholars have echoed this view, and Elizabeth’s childhood translations are often cited as a representative example (Morini 2006). However, more recent (feminist) thinking has argued for a more positive and nuanced view of women’s translation in the period, citing its cultural value and the scope for original expression. Uman posits that translation “gave women entry into the rich literary culture of the Renaissance” (2012: 3). Similarly, Smith (2013: 33) argues that female translators were recognized for their “ability, skill and dexterity”, and translation allowed women to participate “in literary culture in a way that did not openly challenge social or literary power arrangements” (Smith 2013: 27), whilst nevertheless offering a space for them to accrue valuable social status. Religious writing, particularly devotional prose, is another genre appropriate for sixteenth-century women writers. Within the educated elite, the Reformation granted women an opportunity for “heightened cultural agency” through the production of “culturally influential texts” (Coles 2008: 7–8). Coles argues that “some women were […] central to the development of a Protestant literary tradition” (2008: 1); a group that includes Katherine Parr and Elizabeth. Letter-writing was also a genre in which women were active producers. Correspondence was arguably open to a greater number of women due to its non-literary status, although composition was informed by humanist conventions from the continent, as well as the legacy of the medieval ars dictaminis (see Davies 1967). 
The use of written language in Early Modern England was socially and stylistically meaningful, informed by and interpreted via expectations surrounding gender, genre, and social background. The present study is interested in uniting work on Early Modern women’s writing with the insights afforded by sociolinguistic conceptions of style, in order to see how identity can be read across the linguistic levels (the fabric, if you like) of their texts. Roberts (2007) persuasively argues that any discussion of female writing should pay attention to the use of stylistic and formal devices, in order to assess a writer’s literary competence and familiarity with literary culture. Focussing on these two, rather exceptional, women, I wish to assess how Parr and Elizabeth use the linguistic and material resources of mid-sixteenth-century English, viewed explicitly as written language acts, and evaluate the implications for their identity as Tudor women writers.
Born in 1512, Parr’s education followed a humanist curriculum in reading, writing, modern and classical languages, and mathematics (Porter 2010: 29–31; Mueller 2011: 5–6). Widowed twice by the age of thirty (Mueller 2011: 10), she married Henry VIII at Hampton Court in July 1543 to become queen of England (Mueller 2011: 11–12). During this time, and in the period following Henry’s death in 1547 until her own in September 1548, Parr established her identity as a pious and learned woman, linked to the reformist religion. It was in this capacity that Parr’s prose works were produced. Two texts with the most secure claims to authorship are considered here: Prayers or Meditations (henceforth Prayers) (1545) and Lamentation of a Sinner (henceforth Lamentation) (1547). Parr worked closely with Thomas Cranmer to create Church of England forms of worship, with Cranmer publishing the Litany (an English-language service book) in May 1544. Parr’s Prayers was a private prayer book designed to accompany the Litany, and the published text includes Parr’s name on the title page, “the earliest such occurrence for a woman author in England” (Mueller 2011: 369–370). Prayers was produced with the authorisation, even solicitation, of the King; an appointment which may reflect the contemporary association between women and the vernacular. English, in Henry’s eyes at least, was likely not considered the “proper tongue” for learned men to “conduct theological discussion” (Coles 2008: 64).
Prayers is an abridgement of an English translation, produced by Richard Whitford, of Thomas à Kempis’s De Imitatione Christi. Mueller argues that Parr’s text is “a determined, sustained act of intertextual appropriation that constitutes a genuine claim to authorship” (Mueller 2011: 372). Certainly it proved popular, seeing twenty-four print editions over sixty-eight years (Coles 2008: 52). Significantly for the present comparative analysis, there is an autograph manuscript version of this work, held at Kendal, which is thought to have been produced as a gift manuscript for a daughter of Sir Brian Tuke, secretary to King Henry VIII (Mueller 2011: 384). Due to the greater degree of agency in the production of a manuscript, and therefore its more reliable capturing of Parr’s own stylistic choices, this is the version of Prayers used for analysis. The order of production – manuscript versus printed book – is not clear, although the manuscript was likely created at some point in 1544 or 1545, contemporary with the two gift manuscript translations Elizabeth produced for her stepmother (see discussion in Mueller 2011: 369–386).
Lamentation was published in 1547, although the references to a still-alive Henry VIII suggest a composition date of 1546 (Mueller 2011: 425). The prose work offers a formula for conversion, which is dependent on Parr’s “status as both a woman and a public figure” (Coles 2008: 64). It strikes a deferential, self-abasing posture, which led William Cecil, who wrote the prefatory letter, to apologise for the queen’s self-critical tone. Mueller identifies contemporary English religious prose works, such as those produced by William Tyndale and Thomas Cranmer, as influential models, but sees Lamentation as “the most free-standing of Parr’s works” (Mueller 2011: 427). There are also parallels with Elizabeth’s debasing vocabulary in her translation of Navarre (Mueller 2011: 429). Alongside religious prose, Parr’s extant English writings also include thirteen autograph letters written to Henry VIII and Sir Thomas Seymour (uncle of King Edward VI), among others.
Parr’s step-daughter, Princess Elizabeth, was the youngest daughter of Henry VIII, and the only surviving child from his second marriage to Anne Boleyn. Born in 1533, Elizabeth received an exceptional humanist education. As a female royal heir, it was imperative that she achieve “good letters and godly learning” (Mueller and Scodel 2009: 2), following a curriculum in Latin, French, Italian, Greek, as well as reading and composition. By the late 1540s, Elizabeth was securing a favourable reputation for her “virtue, faith, science and experience of language and letters”, which were deemed exemplary for one of her “noble youth and femininity” (Bale 1548: sig. 7r-v). In later life, she sustained her intellectual reputation through patronage, as well as producing her own works of translation (Benkert 2001).
The relationship between Parr and Elizabeth has been described as mutually and positively influential (King 1985), an assessment that is largely based on the written communication and textual production linking the two. Elizabeth produced two prose translations in the mid-1540s, following the gift tradition of occasional pieces, dedicating them to Katherine Parr. The Mirror or Glasse of the Synneful Soul (1544) (henceforth Mirror) is a translation of Marguerite de Navarre’s Le miroir de l’âme pécheresse. The second, How we ought to know God (1545) (henceforth God), translates the first, lengthy chapter of Calvin’s French language Institution de la religion Chrestienne. Ellis speculates that these works were produced with the objective of “making [Elizabeth’s] position at court more secure” (2009: 158), exemplifying the significance of pious intellectual activity, and the work of written language, for the identity and social position of elite women in the period. Notably, the English translations likely formed part of a quartet of translations presented as New Year’s Gifts by Elizabeth in concurrent years (1544–1545). The other works, translations into French, Italian and Latin, were dedicated to King Henry VIII. One such text, the translation of Parr’s own Prayers or Meditations, survives, but the other, a translation of Erasmus, is now lost. She would also undertake a Latin translation of a sermon by Bernard Ochino for her brother, Edward VI (Woudhuysen 2007: 11; Mueller and Scodel 2009).
The two English translations are typical of pre-Reformation piety in their Christ-centeredness and meditative style. Yet, Gibson suggests that they form part of a wider literary movement that propelled the theological shift from the old religion to the new Protestantism. Elizabeth, along with Parr and other women, “helped set the parameters for a good deal of pious writing in prose and poetry in Early Modern England” (Gibson 2004: 33–4). Ellis concurs, noting that even though Elizabeth stresses her translations as possessing “more narrowly literary interests, the very times in which they were being produced connect them to radical trends” (Ellis 2009: 164). The subsequent (and, we can presume, unforeseen) publication of Mirror by John Bale in 1548 is one example of the appropriation of these texts for reformist action. Admittedly, the prose works of Parr and Elizabeth are not wholly comparable, as Elizabeth’s writings are more explicitly translations of singular source texts. That said, Early Modern conceptions of authorship and adaptation meant translations afforded considerable opportunity for originality, with scope for “generic innovation… and stylistic and lexical invention” (Clarke 2001: 13).
Elizabeth’s adolescent canon also includes autograph correspondence, in which Parr’s influence is shown as a frequent recipient. The letters to Parr include Elizabeth’s earliest extant letter, an Italian epistle composed in 1543 that, whilst largely a schoolroom exercise, signals Parr’s role as a focal point for Elizabeth to communicate her educational advancement (Perry 1990: 26–8). Her other letters to Parr are in English, starting with the prefatory epistle to Mirror in 1544, and two further letters in June and July 1548.
The epistolary relationship between the two individuals is also shown, indirectly, in a letter sent from Parr to Thomas Seymour, which is endorsed in Elizabeth’s italic hand: “nolito me tangere” (roughly: let him not touch me [the letter]). Quite why, or how, Elizabeth was given the opportunity to add this endorsement to a letter between her step-mother and her new husband is subject to speculation, although Mueller hypothesizes that Elizabeth was entrusted to pass the letter to a messenger for delivery (Mueller 2011: 169, 171 fn.146). That it alludes to Christ’s words to Mary Magdalene in the gardens, as given in the Vulgate is perhaps indicative of the place of faith in the relationship between Elizabeth and her step-mother.
Studies of Elizabeth and Parr’s writings take various perspectives. Literary scholars have acknowledged the scope for female agency within the religious prose and translations (Clarke 2001), the fashioning of vernacular self-expression (Mueller 2011), and the influence of humanistic models of style (Booth 2013). Linguists, on the other hand, have generally focussed on the correspondence of these women, which is included in the Corpus of Early English Correspondence, mapping their participation in language change in the early modern period, such as morphosyntactic variants (e.g. Raumolin-Brunberg 2005; Evans 2013; Evans 2015) or pragmatic features, such as address terms and self-reference (e.g. Nevala 2004). What emerges from both areas of research is the trend for innovation, with each woman in the vanguard of written, literary and linguistic expression; as Ellis observes of Elizabeth’s English translations, for instance, “they were all ‘firsts’” (Ellis 2009: 162).
The details of Elizabeth and Parr’s autograph works produced in the 1540s, and used for the following analysis, are shown in Table 1. The aim of the present study is to take a synchronic snapshot of the language and style in these mid-sixteenth-century writings, and use this data to explore how each writer used linguistic and material resources for style and identity work in different genres; that is, in sociolinguistic terms, to explore evidence of the written-language style-shifting. The approach combines close and distant reading techniques, using quantitative and qualitative analysis, and reflects on how each woman’s appropriation of the newly-elevated vernacular fits with social and genre conventions of the period, as well as reflecting more local, intrafamilial practices. Significantly, in the prefatory letter to Parr discussed above, Elizabeth apologizes that her text of Mirror contains many flaws, and asks Parr to ensure that “no other, (but your highnes onely) shal rede it, or se it” [my emphasis]. The equivocation between reading (interpreting) and seeing (perception) suggests that the styling of this text, and indeed all the manuscripts considered here, crosses linguistic and material borders, reflecting their production as objects of personal correspondence or gift manuscripts: texts and materials for private consumption. The analysis therefore considers the style of these works as multimodal, looking first at material features and secondly at linguistic forms, to see how they cohere as acts of written identity.
Electronic versions of the texts were prepared using the original manuscript and (when necessary) print originals, wherever possible cross-checking my keyings with recent original-spelling and modernized scholarly editions. The attention to individual stylistic practices entails that the reliability, or authenticity, of the source material is paramount. Autograph status applies to all texts considered here, with the exception of Parr’s Lamentation, where analysis is based on the printed text. In this case, additional care is needed when interpreting the language due to the possibility of printer interference although the extent to which this applies to morphosyntactic variables, such as negation or positive do, can be considered fairly low. The discussion of materiality considers only the manuscript materials, in which each woman’s agency can be more confidently attested.
Interest in social materiality, which conceives of writing as “an embodied act…shaped by physical and social contexts” (Smith 2014: 14), encourages a reading that considers the script, layout, ornamentation and orthography of a text, as features that can signal as much about the writer’s learning, social status, and gender as the language itself. As Daybell stresses, although “content and style as they are registered linguistically” will, and should, remain central to analysis, it is important to account for how “social and cultural meaning was encoded into the very fabric” of manuscript texts (Daybell 2012: 106–7). The visual properties of a document are encountered first, before the linguistic meaning is processed, and therefore play an integral role in signaling the social position, literacy, and identity of that writer, subject to the generic norms of, in this case, Early Modern epistolary or literary style.
Starting with the materiality of the correspondence, an obvious example of the complementarity in material and linguistic resources for epistolary identity construction is the signature. Parr and Elizabeth each rendered their name to be authorizing (the primary function of a signature), difficult to forge (due to connectors and decorative elements), and indicative of their status as Tudor royals. Katherine Parr works to differentiate herself from previous Queen Katherines, by adding her initials “KP” after her full name. Elizabeth, too, has her own ornamental flourish, a quatrefoil, that signals her royal lineage to Henry (Pryor 2003: 12).
The rendering of these signatures is partly informed by the script used by each writer. The hands of Elizabeth I and Katherine Parr have been previously described (see Woudhuysen 2007; Mueller 2011); therefore, the present discussion focusses on evidence of each writer’s awareness, and consequential practices, of the social meaning of their script. Parr’s English correspondence uses a mixed, or “bastard” secretary hand, which Mueller suggests may have been taught to her by her mother, Maud, whose own hand is very similar (2011: 6).  This is a reflection of Parr’s age. She would have learnt to write in the 1510s, before italic script was fully established for the vernacular. Parr appears to have been aware of the humanist associations of italic script, and of her own deficiencies with it. A Latin letter to Princess Mary was drafted by Parr, but copied out by a young Elizabeth (Mueller 2011: 86, fn. 39), and Prince Edward’s comments on the development of Parr’s writing, “la beaute de voz lettres” (Mueller 2011: 113), in the epistles she sent him suggests that advancing her penmanship was an important tool in cultivating her identity as an educated queen. A draft letter in Latin survives on the lower half of a letter from Edward to Parr, leading Mueller to posit that the italic script is “presumably KP’s” (Mueller 2011: 126). If so, then it suggests Parr achieved her goal for Latin writing but did not make the script-shift to italic for her vernacular compositions (see Kaislaniemi, forthcoming). By contrast, Elizabeth was fully trained, and highly competent, in the italic hand. In her letters, the hand is a signal of her schooling and literacy, particularly in the “schoolroom” letters written to her family. Notably, Elizabeth does not differentiate her italic script between languages: the vernacular is firmly inscribed in a humanist hand (see Goldberg 1997). Elizabeth was not the only young elite to use italic for English in the 1540s, of course, but she was among the vanguard, contributing, perhaps, to the increasing social profile of the language in literary, religious and (later in life) political domains.
In Early Modern correspondence, layout was perhaps the most significant material property, signaling the writer’s relationship and social status with the recipient. In recognition of the price of paper, the greater the amount of white space on a page, the greater the respect towards the recipient (or, on occasion, the wealth and flamboyance of the author) (Daybell 2012: 89–91). Vives, writing in the 1530s, remarks that the configuration of a letter’s text on a page should be sensitive to “the rank of the person to whom it is written” (1534: 113).  Thus, it was conventional to have a wide left margin, and to write to the edge of the right side of the sheet. Deference towards the recipient was also signaled by the location of the subscription and signature, with the bottom (right-hand) corner the most respectful location when writing to a superior, even when the main body of the letter terminated only half-way down the sheet (Daybell 2012: 91). Letters to a subordinate, and/or an intimate, were to visually mark their status through a decrease in white space. The subscription could be located underneath the main body or, for the social elite, at the top left of the page.
Parr adheres to these social conventions of epistolary layout. Her signature and subscription is consistently pressed into the bottom right-hand corner in letters addressed to her male superiors, such as Henry VIII (e.g. BL Lansdowne 1236 f.9) and her fourth husband, Thomas Seymour (e.g. CP 133/3, f.6r-7v). Unfortunately, letters to other recipients survive either as copies or drafts, with the subscriptions omitted, although their absence from the main text may imply a similar supine positioning on the page (see, e.g. the letter of condolence to Lady Wriothesley, BL Lansdowne 76, f.182r; Mueller 2011: 80–1). Parr’s letters to her husbands also make judicious use of white space, with a left margin and, when the length of the letter permits, a gap between the main body and the sign-off. Her deference is combined with the formulaic subscriptions, which position Parr in her role as “humble obedient loving wife and servant” (Mueller 2011: 68–9).  Other extant correspondence further illustrates Parr’s epistolary literacy, with her scribal letters produced in her capacity as queen signed at the top-left, thus foregrounding (linearly, in terms of reading) her authority in these contexts (Mueller 2011: 60). Therefore, as queen and as queen dowager, Parr uses epistolary conventions to carefully signal her social identity, underpinning and reinforcing the letter’s message.
Elizabeth’s correspondence shows a similar grasp of the material dimensions of Early Modern letter-writing. Her adolescent correspondence includes letters to Parr, Edward VI and Mary I, and her positioning of the subscriptions consistently indicates a subservient and respectful relationship with her family members. These letters, too, are explicit in their use of white space, with wide left-hand margins and substantial gaps between the main body and the closing subscription. Her letter to Parr, written in 1548, is a visual model of humanist, deferential letter-writing (see image in Perry 1990: 55). That the adolescent, conscientious Elizabeth took such pains over the layout of her letters may not be particularly surprising. However, there are two examples that foreground the social significance that these visual properties must have possessed. Firstly, a letter to Mary I known as the “Tide” letter, which was written when Elizabeth was accused of treason and in serious danger, works stringently within contemporary epistolary conventions, linguistically and visually, as she attempts to persuade her sister to see her in person (see Evans, 2016). Elizabeth concludes this letter on a second sheet using ten lines, and leaves a large gap on the page to place her closing expression and signature in the bottom right corner. One of the best-known features of this famous letter are the eleven diagonal lines drawn across the white space – whether by Elizabeth or by a third-party (such as the letter bearer) is unclear – to prevent interference. This arguably demonstrates the risk Elizabeth was prepared to take (i.e. through forged additions to her correspondence) in order to follow epistolary layout conventions, suggesting their social import for her identity and relationship with her sister. 
Another example of the significance the adolescent Elizabeth placed on epistolary layout is the change in practice following her accession. In her royal autograph letters Elizabeth’s hand shifts to a mixed script of italic and secretary graphs, and white space is no longer such a striking feature of the page. It may be that such changes were partly practical, with the secretary letterforms better-suited to the rapid textual output feasibly required in her role as sovereign; but they also signal a re-evaluation of what Elizabeth, as queen, felt was necessary for her epistolary identity, in which pleasing aesthetics, and their association with gender and literacy, were no longer so essential – what Williams (2011) has called, in relation to George Talbot’s near-incomprehensible hand, “uglyography” .
Parr and Elizabeth’s careful adherence to the visual conventions of letter-writing signal their appreciation of the significance of these features for identity work, both in terms of relationship building, social identity, and for achieving their communicative goals. Their prose gift manuscripts show a similar attention to the social meaning of material elements, but these features are differently configured, reflecting their placement in devotional and religious manuscript writings.
Boffey (2012) has recently drawn attention to the continued production of presentation manuscripts in the fifteenth and sixteenth centuries, even when print versions of the same text were available. The reason for the sustained role of such works was, she suggests, their uniqueness, which was “more valuable and thus more flattering and respectful to its dedicatee” (2012: 57). Yet print also carried an authority, which led to the appearance of manuscripts carefully modelled on their print prototypes, including modulation of hands to reflect changes in typeface, and from the late fifteenth-century, print woodcuts were pasted into scribal devotional works (Boffey 2012: 58, 73). Although pastings and hand-switches are not found in Parr and Elizabeth’s texts, both writers draw on other material resources that seemingly respond to contemporary social evaluations of both media, demonstrating their familiarity and competence with each format. This can be seen in three related properties: external features, script choices and punctuation.
Parr’s tiny volume of Prayers, which measures only four by six centimeters, is bound in solid silver, with “raised bands simulating a leather binding” (Mueller 2011: 384). Elizabeth’s prose manuscripts are bound in embroidered covers, each of which inscribes the target reader into the exterior, with the initials “K.P.” raised up from the base cover. Elizabeth also includes spinal cord-marks which evoke “a bound manuscript volume such as one might find in a library” (Quilligan 2000: 208). Thus, the external materiality, literally the fabric of the work, signals their status as literary texts, with the associated cultural weight.
The rich, external fabric is matched by decorative choices on the internal pages that draw on manuscript and print traditions. Parr uses differently coloured inks to frame her black-inked secretary script. This inked ornamentation includes a red border, infilled line-initial letters (Valentine 1965: 78, 83), and red and blue rectangles at line-ends, which contain vine and leaf motifs, known as arabesques (Valentine 1965: 39). Vine and leaf illustrations were common in medieval manuscripts, particularly those of a religious and devotional nature, because of their symbolic connection to communion, as well as having classical links (Fisher 2004: 5–6). Parr’s use of colour, too, is evocative of medieval Books of Hours, which would often include rich decorative borders with botanical ornamentation (Fisher 2004: 14). (There is some similarity in Parr’s ornamentation and Elizabeth’s use of flowers in her embroidered cover design.) The selection of decorative paratextual features therefore situates Parr’s gift within medieval and devotional manuscript traditions, even as the text challenges these traditions through its vernacular promotion of private devotion. Parr’s manuscript has been read as “a transitional item in the history of the book”, where:
Coloured inks are not used in Elizabeth’s English prose works, although the coloured thread offers some visual similarities.  However, Elizabeth does use the “aldine” leaf (also known as the vine leaf, ivy leaf, hedera or floral heart (Vervliet 2012: 3)), inscribing the heart-shaped leaf at line ends throughout Mirror. By 1544, the vine leaf had an established tradition in both manuscript and printed modes. Originating in classical inscriptions as a word divider, period, or line filler, its use was maintained into medieval manuscripts where it was used to signal the start of a passage or paragraph (Vervliet 2012: 6). In print, the vine leaf became “one of the most favoured decorative designs in Renaissance typography” (Vervliet 2012: 5) from the early sixteenth century. In selecting this symbol, therefore, the text appears to be responding to conventions of both media, demonstrating Elizabeth’s familiarity and competence with each format. Its presence on the title page of one edition of her source text, published 1531, may also explain its implementation in Elizabeth’s manuscript, although this is not the edition Elizabeth is thought to have used (Mueller and Scodel 2009: 29). By comparison, the manuscript of God has little ornamentation, a stylistic choice that may reflect the logocentric message of the text. Indeed, Elizabeth’s emphasizes this message in her preface, arguing for the primacy of language, compared to visual arts, because it offers an immediacy of meaning and engagement with God (Mueller and Scodel 2009: 214; Ellis 2009: 169–170), suggesting how material choices were deliberately chosen to style these literary works, and convey the abilities and values of their writer.
The prose works thus draw on material components that signal their status as literary works, emphasizing the hand-crafted efforts of their writer as well as their legitimacy through assimilation of print conventions. Material features therefore have significance for identity work in both genres considered here, constructing a set of visual and aesthetic cues that would signal to a reader the writer’s education, writing experience, piety, and social status. Each woman shows a mastery of the material conventions for each genre, and their adherence to such conventions positions Parr and Elizabeth as exceptional women, not in their transgression from these norms, but rather in their exemplification as learned women writers cast in the new, humanist mold. Each shifts their material practice in order to utilize the social significance of the visual forms of a given genre, integrating small aspects of individuality (the design of the signature, the selection of ornamentation) into the material construction of their written texts. Their material stylistic accomplishments are something that would be seen, before they were read: tactile and arresting objects that represent each author as appropriately deferential, pious, and literate. In sum, the material identities constituted by the correspondence and prose each align their writer with the new feminized ideals of mid-sixteenth century England.
The code choice of the correspondence and religious works likewise aligns Parr and Elizabeth with gendered norms of the period. Despite their knowledge of classical and European languages, Parr and Elizabeth choose English as the language for their prose, and much of their correspondence. Parr’s work in the vernacular has been read as evidence in her reformist beliefs, as well as perhaps reflecting less confidence in her linguistic abilities in other languages. English also dominates the adolescent canon of Elizabeth, although she produced correspondence in a greater range of languages than her stepmother. Nevertheless, this, too, may have a gendered distribution, with more letters written in Latin to her brother and her father, whereas those to Parr and her sister Mary are generally in English (an exception to this is her French preface to God). Her English letters are nevertheless carefully crafted, drawing on medieval ars dictaminis and Humanist models of letter-writing to flatter the recipient, and demonstrate Elizabeth’s understanding of epistolary construction and classical learning (Booth 2013: 19–28; he also notes that Elizabeth’s display of formal epistolary structuring is shared with the Latin letters of her younger brother, Edward. See an example of Elizabeth’s correspondence to her brother at the British Library).
Elizabeth’s understanding of the gendering of translation and literacy in the period is further suggested by her prefacing description of the target language of God as “nostre langue maternelle”. The presentation of the vernacular as a shared mother tongue carries plural meanings in this context, evoking the familial relationship between Elizabeth and Parr, their shared regal identity in the English Tudor lineage, and their religious identity. Indeed, Elizabeth’s awareness of her dedicatee’s influential position in reform is reflected in the statement that “the organ of your [Parr] royal voice may be the true instrument of His word” (Mueller and Scodel 2009: 219). By conducting her own Englishing, Elizabeth is paying tribute to her stepmother, and following her lead in promoting and perhaps also shaping the place of the English language in religious, literary culture. But this raises the question of how it was deployed: what “variety” or “style” of English did each writer use, and what meaning did this have for their identity as Tudor women writers? The following analysis concentrates on morphosyntactic features underdoing change during the period; features that are not linked to content so much as the structural style of a piece.  Macro-level studies (e.g. Rissanen, Kytö and Palander-Collin 1993) have shown the importance of genre in the diffusion of language change. The present analysis explores some well-known linguistic variables to assess whether stylistic variation is evident in the contemporaneous language of Parr and Elizabeth’s correspondence and prose, and the possible significance of this for their identities as women writers.
The Early Modern period witnessed many changes in English morphology and syntax. One major development concerns English second-person pronouns, both in form and in social function. In Middle English, the second-person pronoun system was numerical: ye (subj) and you (obj) were used for plural referents, with thou (subj) and thee (obj) applied to singular referents. In Early Modern English, the you/thou distinction shifted to mark social relations, you used for polite and deferential address, and thou more intimate (and, subsequently, insolent) address (Lass 1999: 152–4). A formal change, the generalization of you to all case positions, also took place during the late fifteenth and sixteenth centuries (Bergs 2005: 129; Lass 1999: 154), although the outgoing form, ye, survived longer in more formal registers.
The writings of Parr and Elizabeth largely support the observed linguistic trends for these pronominal changes (see Table 2 and 3. For a more thorough account of this, and other morphosyntactic changes at the mid-century Tudor court, see Evans 2015). In their correspondence, both writers consistently use the polite you form. This includes even the relatively intimate letters between Parr and Thomas Seymour:
Elizabeth’s letters show no evidence of ye, suggesting that the generalization of you was well developed in her written English. Parr, on the other hand, uses both you and ye, at a roughly 50% split, a distribution that can plausibly be attributed to generational differences.
However, the pronouns in the prose works show a different configuration. Firstly, reflecting the intimacy of devotional writing, both Parr and Elizabeth use second-person pronoun thou/thee in their apostrophic address to God:
You/ye is consequently far less frequent, with only twelve tokens in total across these works. However, in contrast with the correspondence, the distribution of ye/you follows the Middle English numerical system (example 5). This would likely have felt archaic even to its sixteenth-century readers, and perhaps even to the writers themselves, as suggested by a correction found in Elizabeth’s manuscript of Mirror (example 6). The comparable shift to more archaic, linguistically conservative pronominal practice, giving thou a central role in contrast to each writer’s usual “speech-like” epistolary practice, suggests a shared consensus regarding vernacular devotional prose style.
Other stylistic properties further suggest that Parr and Elizabeth had similar conceptions of what constituted appropriate epistolary or religious prose style, and the contrasting properties of these. The second-person pronoun choice in the prose works can be understood as contributing to a conservative, or archaic style, in contrast to their epistolarly writing. Other morphosyntactic variables can be similarly interpreted based on their use and distribution. For example, the replacement of third-person singular verb ending -eth by -es is a well-documented change in Early Modern English, shaped by a range of internal and external factors (e.g. Kytö 1993; Gries and Hilpert 2010). One significant feature of the change is lexical diffusion, with the auxiliary verbs have and do lagging behind other (lexical) verbs throughout the progression of the change (Nevalainen and Raumolin-Brunberg 2003: 67–8). Another association, relevant to the present analysis, is that -eth was stylistically more formal as the outgoing variant, although recent quantitative work suggests this evaluation developed only in the seventeenth century, at least in correspondence (Gries and Hilpert 2010: 307).
The distribution of -eth and -es in the correspondence supports previous work regarding the temporal and sociolinguistic development of the change (Table 4). Parr is a conservative user, congruent with her age, generation and gender, with -es accounting for only 2.1% (one token) of lexical verb forms in her autograph letters, and no examples with have/do (examples 7 and 8). Conversely, Elizabeth’s youth appears to predispose her to more innovative linguistic preferences, using -es 30% of the time with lexical verbs, and 20% of the time with have (although never with do) (example 9).
By comparison, the religious prose writings of Parr and Elizabeth use -eth for all verb forms (example 10). This contrast can be read as stylistically meaningful, perhaps motivated by the more conservative stylistic properties of eth, to complement the archaism of their pronoun choices. Although Gries and Hilpert (2010) date the formality association of -eth to the seventeenth century, this dating applies to correspondence. We might expect that religious language texts are more sensitive to stylistic values than the less formal “speech like” style of epistolary writing, which may therefore include the traditional, non-innovative verb-ending.
Sentential negation also contributes to a conservative style in the religious prose, when compared with the epistolary writing. In Early Modern English, multiple negation, the use of two or more negative particles within a clause, was gradually supplanted by single negation (one negative particle, not, never, with a non-assertive form such as any, ever). Single negation emerged in administrative texts in late Middle English, where precision and clarity of expression were paramount (Kallel 2007: 27), although the stigmatization of multiple negation did not develop until the seventeenth century.
The results for this variable are shown in Table 5. In her correspondence, Parr shows a preference for single negation. The figures need to be treated with caution, however, as there are only four tokens overall. In Elizabeth’s letters, the young princess shows herself again to be linguistically innovative, with 13 of 14 negative constructions featuring non-assertive forms (cf. Evans 2015). In their religious prose, however, multiple negation is more established, constituting 12 of 21 tokens in Parr’s writing, and 13 of 29 tokens in Elizabeth’s prose (examples 11 and 12). Although the increase in multiple negators may be attributed to source language interference in the case of the latter, the comparable pattern in Parr’s works again suggests a more deliberate, and perhaps conservative, motivation for the stylistic shift. This is further borne out by the presence of the negator ne (example 12) in Parr’s Prayers, a form that was in decline from the turn of the fifteenth-century in Early Modern English (Lass 1999: 269–270). The negator is not found in Parr’s correspondence.
Viewed accumulatively, and by contrasting the epistolary and religious prose, the aforementioned morphosyntactic features indicate that a conservative or archaic vernacular style was perceived as an appropriate quality for religious prose, with correspondence being the more innovative genre (affirming its continued positioning as being the more “speech like” Early Modern genre, and therefore well suited to exploring processes of language change (see Nevalainen and Raumolin-Brunberg 2003)). The implementation of the second-person pronoun system by number, rather than social properties of the referent, is especially striking, and does suggest “conscious” thought on behalf of Elizabeth and Parr in the linguistic dimension of these works.
However, further analysis also reveals that archaism is not the only stylistic “theme” pursued in the religious texts. The pronoun choices, as well as being archaically styled, also create an interactive aspect, evocative of spoken address. This reflects each text’s construction of a personified narrator, engaging both with God and with the reader (on the gendering of these speakers see Coles 2008; Mueller and Scodel 2009). The narratorial voice of the religious prose is also vivified through the use of exclamatives, including alas, yea and o, in apostrophic or reflexive address. By contrast, the explicitly interactive correspondence uses exclamatives infrequently, with only two instances (yea), which occur in the aforementioned “Tide” letter from Elizabeth to Mary. The exclamatives can be seen to add an emphatic, oratorical dimension to the religious writing, in keeping with its private and devotional purpose, that appears less suited to the direct and perhaps more practical style of the correspondence.
The interactive, oratorical and emphatic style theme can also be associated with other morphosyntactic properties, namely positive and negative declarative do. Auxiliary do was an innovation in sixteenth-century English. For Early Modern writers, positive do provided syntactic and expressive clarity, as well as adding discursive emphasis to key sections of a text. These functions distribute differently across text-types, indicating different stylistic weight. Positive do occurs more frequently in formal text-types, seemingly disambiguating verb phrases in complex sentences that may be classed as more literary, or written, whereas in more speech-like genres, such as witness depositions, do-clusters foreground meaning at the discursive level (topic or narrative peak) (Rissanen 1999: 240–1).
Positive do is a marked variant in Elizabeth’s correspondence and adolescent translations, occurring exceptionally frequently in the latter (see Table 6). The preponderance of do in her prose translations is seemingly linked to Elizabeth’s employment of the form for both rhetorical and discursive functions. The original French texts offer little explanation for the frequency. Although Palsgrave (1530: fol. xxi) suggests that do may be an English equivalent for the French past imperfect tense, due to its ability to mark habitual action, there is little to suggest that this is a primary cause in Elizabeth’s English writing. Large-scale corpus studies likewise offer little evidence that do was used for this function in Early Modern English (Rissanen 1991: 322–3).
One explanation for the high frequency is that Elizabeth was attempting to emulate (and over-hitting) the prose styles of other genres which drew upon the form’s emphatic and intensifying functions. Rissanen (1999: 240) reports that do was “favoured in the records of court trials […] and, to a somewhat lesser extent, in sermons. Both text types are based on argumentative spoken discourse in highly formal situations”. In these genres, “[d]o in itself is not necessarily emphatic [at the level of the verb phrase], but it adds to the intensity and emphasis of the utterance” (Rissanen 1999: 240). From the perspective of Elizabeth’s translations, the association of do with sermons, formal (spoken) prose with a religious, proselytizing subject, is a possible stylistic influence (example 13).
Interestingly, and in contrast to the trends so far established, Katherine Parr does not show the same preference for the variant; in fact, do is slightly more frequent in her correspondence. In Parr’s prose, do occurs in contexts associated with rhetorical do (example 14), such as hypotheticals, subordinate clauses and sentences with intervening material between subject and verb phrase (Rissanen 1999: 241), rather than the clustering associated with discursive foregrounding do. Whatever the stylistic effect Elizabeth was aiming for in her religious prose writing, in regards to positive do, it appears that this was not shared with her stepmother.
The distribution of negative declarative do, on the other hand, shows greater similarity in the epistolary and prose styles of each writer; negative do occurs more frequently in the latter, quite considerably so in Elizabeth’s prose (75% to 17.6% in the correspondence) (Table 7). There are a few caveats in regards to the frequencies for Parr, which also suggest a preference for negative do in her religious prose (13.1% to 0% in her correspondence). Firstly, the token count for her correspondence is very low (only 8 constructions) and, secondly, negative do occurs only in the printed prose work, Lamentation, rather than the manuscript fragment of Prayers. This could indicate third-party interference is responsible for negative do in the print text. However, there is a plausible stylistic motivation to the distribution of negative do in the data, linked to the self-abasement of the speaker that characterizes the devotional, vernacular prose, and in particular the agentive, human voice (examples 15 and 16).
The morphosyntactic analysis identifies features that show stylistic variation across the correspondence and prose writings of both Parr and Elizabeth, with patterns in distribution that can be aligned with two broad stylistic effects. The correspondence is the more innovative genre, capturing changes in morphosyntax that can be equated with correspondence’s greater informality and speech-like style. The innovative linguistic features, it should be noted, are framed by a compliant adherence to (what we understand to be) contemporary epistolary materiality conventions, such as layout.
The religious prose writings, on the other hand, include more conservative forms, perhaps reflecting a perceived link between traditional language use, and the weighty, credible subject matter. These are interwoven with intensifying features associated with interaction and spoken expression. The crafted prose styles of these works, which allude to spoken genres and previous linguistic eras, are framed by rich material signs that signal a transitional modality, perhaps paralleling the transient, reformist ideologies of their texts.
The distinctiveness of the prose writings, when compared with the epistolary style of each writer, and the general coherence in this styling of the devotional writing between the two women, raises questions about the stylistic models that Parr and Elizabeth were drawing on when producing their literary texts. Their influence on each other is possible, and likely. However, literary scholars have also speculated as to the influence of biblical works on the stylization of these vernacular prose works. Writers, especially women, took inspiration from the (newly vernacularized) Bible, in which “[v]irtually all literary forms and genres could be identified in some guise”, and, Clarke suggests, served as a “useful pretext for experimentation” in developing literary expression (2001: 148). More specifically, Ellis (2009: 177–8) suggests that Elizabeth worked closely with an English Bible when completing her translations, based on the citation information she provides for the psalms and other biblical material. Parr, too, is thought to have been influenced by the New Testament in terms of the structuring and sentiments of her religious writing, as well as the sermons of Bishop Fisher (Coles 2008: 68). Such hypotheses relate more broadly to the ahistorical question of how individuals develop diverse writing styles which interface with and inform genre norms within a speech community (e.g. Argamon et al. 2003: 342–3).
For the final analytic step of this study, therefore, I apply computational methods of stylistic analysis, more typical of distant reading methods (Moretti 2013), to investigate this particular hypothesis, in recognition that it may reveal profitable avenues for future inquiry. This analysis focusses solely on the linguistic stylometric properties of the letters and prose works discussed here. The inclusion of material properties, which could potentially be incorporated into quantitative measures of similarity and difference between documents, seems plausible, but is not pursued here.
Burrows’s delta (2002) is a statistical measure designed to identify the degree of similarity between texts based on the most frequent words (MFW). Developed for literary authorship analysis, delta establishes the likeness between texts by establishing z-scores for these frequent words (not discriminating between function and content words), the sum of which is known as delta. As a method, it offers another way of exploring style as a repertoire of markers, because it identifies those linguistic features that, in combination, mark out one textual work from another.
Delta was developed for authorship analysis (Burrows 2002), and has subsequently been tested and refined for different historical periods and literary genres (e.g. Hoover 2004, Smith and Aldridge 2011). However, as a measure, like many stylometric tests, it is known to be sensitive to genre and topic, as well as the authorship signal (Antonia and Jordan 2014; López-Escobedo et al. 2013). Given that the current investigation is interested in stylistic similarity, rather than authorship per se (although the question of collusion and support in the production of the prose works by Parr and Elizabeth is an interesting one), delta should allow any systematic differences or similarity in style to emerge. A similar computational approach was taken by Connors (2006) to investigate the stylistic distinctiveness of Elizabeth Cary’s Mariam.
The test conducted here compares Parr and Elizabeth’s writings with contextually-relevant religious works, controlling, in a sense, for topic. Specifically, it compares the women’s texts with Cranmer’s Litany (1544) and Exhortation (1544), the first ten chapters of Coverdale’s 1540 Bible, and the sermons of Latimer and Fisher. The latter two were taken from the Helsinki Corpus, with the others collected from EEBO-TCP. The test was run using Stylo for R (Eder and Rybicki 2011). Following Smith and Aldridge (2011), who assessed the most effective calibration of word vectors and text length, the top 300 MFWs were used for comparison. This did not discriminate between function and content words. Figures 5 and 6 show the similarity in delta values for the eleven documents using cluster analysis. Cluster analysis organizes the texts into “most similar” groupings.
Figure 5 uses the 300 MFW without any modification. It reveals that the morphosyntactic patterns discussed above appear to operate across the texts more generally and that genre style overrides authorship in terms of greatest likeness. Elizabeth and Parr’s letters group together, with their respective Mirror and Prayers also comparable. The greater interactivity of these texts, based on their vocative and pronominal features, could partly explain this similarity. Interestingly, the cluster analysis suggests that Parr’s Lamentation is most comparable to Coverdale, with the sermons of Latimer and Fisher also placed on the same branch. Elizabeth’s God is located mid-way between, suggesting that the vernacular biblical style, and contemporary oratorical expression, may well have shaped the women’s style in these texts. Notably, however, Cranmer’s works are greatly removed from all other texts in the sample.
Figure 6 uses the same measure, but culls all pronouns from the MFW list. Although it appears that pronouns have little effect on the accuracy of delta in poetic works (Burrows 2002; Smith and Aldridge 2011), they have been found to be influential in prose texts (Hoover 2004). As the preceding analysis showed, pronouns are an especially distinctive, and differentiating, feature in these particular letters and religious prose works. Their significance is clearly picked up by the delta test. Once removed, authorship becomes a stronger, although not absolute, organizing factor. Again, the test identifies greatest similarity between Parr’s Lamentation and Coverdale, with these works also similar to Elizabeth’s prose (now grouped together) and the sermons. Cranmer’s writings remain on a separate branch, suggesting little stylistic influence on Parr or Elizabeth; or, from the opposite perspective, an exceptionally strong authorship signal in his writing.
Figures 7 and 8 run the same delta analysis, but calculate the findings using Principal Component Analysis (PCA). PCA can capture more nuance than cluster analysis, which may over-sort the data (Antonia and Jordan 2014: 312). Figure 7 retains the pronouns in the MFW list. The letters, Mirror and Prayers show similar properties on both the first and second principal components identified in the analyss (PC1 and PC2, respectively). This situates these texts in the bottom-right quadrant, a distribution which could again be attributed to the greater interactive and emphatic properties of these texts. The other prose works, Lamentation and God, are located amidst the religious writings of Coverdale and the sermons, with a much higher score on PC2 than the other writings. The contrast with Cranmer reaffirms the cluster analysis results. Figure 8, with pronouns excluded, shows only slight differences, with Coverdale, Latimer’s sermon, God and Lamentation scoring less highly on the second factor than Fisher (which could indicate that pronouns are less central to Fisher’s text than these other works).
As one statistical test, this is not conclusive proof of the stylistic influences relevant to Parr and Elizabeth’s prose compositions. However, it does support the stylistic contrast identified for particular morphosyntactic features in the writings discussed above. As an indiscriminate, quantitative measure, it also confirms that style extends from culturally-salient choices of language, address forms, and self-reference, through to the structural components that are far less noticeable but nevertheless linguistically integral, such as morphosyntax. In drawing on the different stylistic variants available to them in English, Parr and Elizabeth are able to construct genre-specific voices for themselves, which conform to conventions of letter-writing or religious writing, whilst at the same time offering space for autonomy of expression, and serving their own, specific ambitions as Tudor women writers.
Like any communicative act, Parr’s and Elizabeth’s style choices are informed by the context of production. The analysis has shown how such choices in material and linguistic form are informed by cultural conventions and local goals of communication, which cohere into a sophisticated matrix that articulates their education and literacy within a vernacular frame that is both feminine and humanist. In some ways, the findings are unremarkable: each woman uses material and linguistic cues that are conventional for their purposes. Yet, for precisely the same reasons, these texts signal the exceptional social experience of the two women, who work within the newly elevated vernacular to showcase their learning and literacy, and configure their identities within contemporary boundaries permitted for their gender and social status. In exploring the style of their writings, the study attempts to respond specifically to the works as written texts, considering material features alongside linguistic ones. This arguably makes for a more complex analysis, but also, I hope, gets closer to the processes of identity construction informing, and arising from, each written, communicative act. As Elizabeth advised her stepmother, her textual accomplishments need both to be read, and to be seen. It is through the appreciation of the stylization of both material and linguistic forms that we can better understand these Tudor women writing.
 Letters are the genre most commonly used for historical sociolinguistic research, in part due to the more “speech like” language than other genres. [Go back up]
 With thanks to the Mayor of Kendal and Kendal County Council for their help with the images. [Go back up]
 With thanks to the anonymous reviewer for alerting me to this contemporary reference. [Go back up]
 Notably, the depiction of herself as “servant” is constrained to her letters to Henry VIII, appropriate to his standing. She identifies herself as “wife” only in her letters to Thomas Seymour. [Go back up]
 Elizabeth uses red ink in one of her adolescent letters, written to Thomas Seymour in 1548, where it borders the page. The similarity with the red border in Parr’s gift manuscript is, when viewed from a local community perspective, interesting – especially given Elizabeth’s involvement with Parr’s correspondence to Seymour – but red ink borders can be found in other letters of the period (Daybell 2012: 101), which makes it difficult to assign any specific, intrafamilial meaning. [Go back up]
 I have previously discussed some aspects of morphosyntactic variation in Elizabeth's translations and letters, alongside her speeches, although with a more traditional variationist focus on language change. See Evans 2013. [Go back up]
 This letter survives as a seventeenth-century copy (Bodleian Library, Rawlinson D1070, f.4-5r) and thus is given in modern spelling only in Mueller (2011: 139). I have been unable to view the original manuscript. All quotations are given in their original, holograph spelling, where available. [Go back up]
CEEC = Corpus of Early English Correspondence. 1998. Compiled by Terttu Nevalainen, Helena Raumolin-Brunberg, Jukka Keränen, Minna Nevala, Arja Nurmi and Minna Palander-Collin at the Department of Modern Languages, University of Helsinki. http://www.helsinki.fi/varieng/CoRD/corpora/CEEC/
EEBO-TCP = Early English Books Online – Text Creation Partnership; Phase 1. 2015. Available at http://quod.lib.umich.edu/e/eebogroup/
HC = Helsinki Corpus. 1991. Compiled by Matti Rissanen (Project leader), Merja Kytö (Project secretary); Leena Kahlas-Tarkka, Matti Kilpiö (Old English); Saara Nevanlinna, Irma Taavitsainen (Middle English); Terttu Nevalainen, Helena Raumolin-Brunberg (Early Modern English). Department of English, University of Helsinki. http://www.helsinki.fi/varieng/CoRD/corpora/HelsinkiCorpus/
Antonia, Alexis & Ellen Jordan. 2014. “Identifying Anne Mozley’s contributions to the Christian Remembrancer: A computational stylistic approach’. Victorian Literature and Culture 42: 303–323. doi:10.1017/S1060150313000478
Argamon, Shlomo, Moshe Koppel, Jonathan Fine & Anat Rachel Shimoni. 2003. “Gender, genre, and writing style in formal written texts”. Text: Interdisciplinary Journal for the Study of Discourse 23(3): 321–346. doi:10.1515/text.2003.014
Bale, John, ed. 1548. A godly medytacyon of the christen sowle. Wesel: Dirik van der Straten. http://estc.bl.uk/S111990
Beech, Nic. 2008. “On the nature of dialogic identity work”. Organisation 15(1): 51–74. doi:10.1177/1350508407084485
Benkert, Lysbeth. 2001. “Translation as image-making: Elizabeth I’s translation of Boethius’s Consolation of Philosophy”. Early Modern Literary Studies 6.3(2): 1–20. http://purl.oclc.org/emls/06-3/benkboet.htm
Burrows, John. 2002. “‘Delta’: A measure of stylistic difference and a guide to likely authorship”. Literary and Linguistic Computing 17(3): 267–288. doi:10.1093/llc/17.3.267
Connors, Louise. 2006. “An unregulated woman: A computational stylistics analysis of Elizabeth Cary’s The Tragedy of Mariam, The Faire Queene of Jewry”. Literary and Linguistic Computing 21 (Supplementary Issue): 55–66.
Davies, Norman. 1967. “Style and stereotype in Early English letters”. Leeds Studies in English 1: 7–17. http://digital.library.leeds.ac.uk/80/
Evans, Mel. 2013. The Language of Queen Elizabeth I: A Sociolinguistic Perspective on Royal Style and Identity (Transactions of the Philological Society Monograph Series). Chichester: Wiley-Blackwell.
Evans, Mel. 2015. “‘The vsual speach of the Court’?: Investigating language change in the Tudor family network (1544–1556)”. Journal of Historical Sociolinguistics 1(2): 153–188. doi:10.1515/jhsl-2015-0011
Evans, Mel. 2016. “From ornament to armament: The epistolary rhetoric of Lady Elizabeth Tudor”. Linguistics and Literary History, ed. by Anita Auer, Victorina Gonzalez-Diaz, Jane Hodson and Violeta Sotirova. Linguistic Approaches to Literature 25: 49–70. Amsterdam: John Benjamins.
Gibson, Jonathan. 2004. “Katherine Parr, Princess Elizabeth and the Crucified Christ”. Early Modern Women’s Manuscript Writing: Selected Papers from the Trinity/Trent Colloquium, ed by Victoria E. Burke & Jonathan Gibson, 33–50. Aldershot: Ashgate.
Gries, Stefan & Martin Hilpert. 2010. “Modelling diachronic change in the third person singular: A multifactorial, verb- and author-specific exploratory approach”. English Language and Linguistics 14(3): 293–320. doi:10.1017/S1360674310000092
Hoover, David. 2004. “Testing Burrows’s Delta” Literary and Linguistic Computing 19(4): 453–475. doi:10.1093/llc/19.4.453
Kaislaniemi, Samuli. Forthcoming. “Visual aspects of code-switching in Early Modern English manuscript letters and printed tracts”. In Reading the Page: Verbal and Visual Communication in Early English Texts. Brepols.
Kallel, Amel. 2007. “The loss of negative concord in Standard English: internal factors”. Language Variation and Change 19: 27–49. doi:10.1017/S0954394507070019
King, John. 1985. “Patronage and piety: The influence of Catherine Parr”. Silent but for the Word: Tudor Women as Patrons, Translators, and Writers of Religious, ed. by Margaret Hannay, 43–60. Ohio: Kent State University Press.
Kytö, Merja. 1993. “Third-person present singular verb inflection in early British and American English”. Language Variation and Change 5: 113–139. doi:10.1017/S0954394500001447
Labov, William. 1990. “The intersection of sex and social class in the course of linguistic change”. Language Variation and Change 2: 205–254. doi:10.1017/S0954394500000338
López-Escobedo, Fernanda, Carlos-Francisco Méndez-Cruz, Gerardo Sierra & Julián Solózano-Soto. 2013. “Analysis of stylometric variables in long and short texts”. Procedia – Social and Behavioural Sciences 95: 604–611. doi:10.1016/j.sbspro.2013.10.688
Morini, Massimiliano. 2013. “Virgil in Tudor dress: In search of a noble vernacular”. Neophilogus 97: 591–610. doi:10.1007/s11061-012-9329-x
Palsgrave, John. 1530. Lesclarcissement de la langue francoyse. London: J. Haukyns. http://estc.bl.uk/S104266
Prescott, Anne Lake. 1985. “The pearl of the valois and Elizabeth I: Marguerite de Navarre’s Miroir and Tudor England”. Silent But For The Word: Tudor Women as Patrons, Translators and Writers of Religious Works, ed. by Margaret Hannay, 61–76. Kent, Ohio: Kent State University Press.
Roberts, Sasha. 2007 “Women’s literary capital in Early Modern England: Formal composition and rhetorical display in manuscript and print”. Women’s Writing 14: 246–269. doi:10.1080/09699080701314741
Smith, Peter & W. Aldridge. 2011. “Improving Authorship Attribution: Optimizing Burrows’ Delta Method” Journal of Quantitative Linguistics, 18(1): 63–88. doi:10.1080/09296174.2011.533591