2Such a tension between constraint and freedom is an invitation to read paratexts in generic terms. If genre theory is “a means of accounting for connection between topic and treatment within the literary system” (Colie 1973, 29), then the early modern paratextual material should be read through that frame, especially as one of the main functions of paratexts is to connect social situations with linguistic and material codes (Belle and Hosington 2018, 3). Even informational types of paratexts which do not allow for much variation, such as imprints, can be read through that lens, as Helen Smith’s study has shown: “these paratexts […] operate within generic conventions which, read carefully, reveal imprints to be fictive engagements with a surprising range of literary and cultural concerns” (Smith 2011, 17). These generic conventions vary depending on the specific section or category of paratext (a preface does not follow the same rules as a title-page, whether it be formally or in terms of content), though they can share the same goals, so that paratext should probably be less defined as a single genre than as a category that combines items with various autonomous generic characteristics, as a sequence of “genres of textual liminality” (Saenger 2006, 7). In other words, various kinds of paratextual writings (dedications, addresses to readers, commendatory poems, etc.) engage with various genres (which would have been readily identifiable by early modern readers). This article is precisely about understanding what it means to read paratexts generically.

3That paratexts are generic objects is sometimes manifested within paratexts themselves. A notable example of generic consciousness in paratextual material appears in the 1622 edition of Shakespeare’s Othello. The play is introduced by a short preface written by the stationer Thomas Walkley, where he justifies his direct address to the reader:

• 2 Shakespeare 1622, A2r. Quotations are from the original editions available in EEBO, with modernised (...)

To set forth a booke without an Epistle, were like to the old English proverbe, A blew coat without a badge, & the Author being dead, I thought good to take that piece of worke upon me. To commend it, I will not, for that which is good, I hope every man will commend, without intreaty.2

4Walkley legitimizes his epistle to the reader by the absence of any prefatory material written by the author for this work before his death. A book without any preceding front matter would be like “a blue coat without a badge”, meaning an incomplete livery, a servant’s livery lacking his master’s crest and motto stamped on it. This suggests that a London stationer in the 1620s thought that the reading public was expecting some kind of front matter to introduce a book and that it would be an odd singularity not to have any. Walkley’s elliptic definition also implies that he supposed that the readers would be able to recognize a specific type of paratext, “an epistle”, which is not any epistle but a standard form with identifiable characteristics: a formal structure (the “epistle” itself), with a clear addressee (“the reader”) and a subject matter (in Walkley’s terms, “commending” the book). Walkley therefore seems to characterize liminary epistles as a genre in itself. Indeed, if we refer to Tzvetan Todorov’s definition of a genre (himself quoting from Hans Robert Jauss’ reception theory): “[genres] function as ‘horizons of expectations’ for readers and as ‘models of writing’ for authors” (Todorov 1976, 163), which is exactly what Walkley is describing here. In the first part of this article, I shall explore these generic characterizations in dedications and readerly addresses, through analysing the role of the epistle as “horizon” and “model.”

“I must needs discharge two Epistles”: the epistolary genre in dedications and readerly addresses

6Because dedications and addresses to the reader are essential in publicising social relationships (between an author and a patron, between an author and the readers or between a stationer and the readers), they are often read through a sociological lens, which concentrates on the type of socio-economic situation that these paratexts describe. But Walkley’s rejection of “a book without an epistle” also reminds us that these paratexts were first identified by readers because of their generic nature, that of being precisely “an epistle”. Following the evidence provided by many scholars such as Rosalie Colie (1973), Alastair Fowler (1982) and Michael Saenger that “people in the Renaissance thought generically” (Saenger 2006, 8 n. 21), we are invited to look in more detail into the kind of epistolarity involved in dedications and readerly addresses.

8Moreover, even dedications that were not initially presented as “epistles” could be contaminated by the logic of generic categorization at an editorial level. This is the case for instance of the dedication in A hundreth good pointes of husbandrie, by Thomas Tusser (1557). When the book was first published by Richard Tottel, it included a verse dedication to the author’s patron, “to the ryght honorable and my speciall good & master the Lord Paget” (Tusser 1557, A2r). The poem, signed by “your servant / Thomas Tusser”, reads as a verse epistle but is not described as such in the title and there is no running title. This book is published again by Tottel in 1573, in an enlarged version, Five hundreth points of good husbandry. In this new version, the dedicatory pages have changed. This time, the dedicatory poem is entitled “The authors epistle to the late Lorde william Paget” (Tusser 1573, A3r), with “the epistle” as running title and is preceded by another verse dedication, addressed to Lord Paget’s son, also with “the epistle” as running title. This example suggests that the practice of dedicating a book had become codified as a liminary genre, “the author’s epistle”, which was integrated into the economy of the book not only by authors but also by stationers. The dedicatory epistle is not only defined through its form, given the flexibility of early modern epistolary practices (Guillén 1986, 71-73), nor just by a rhetoric of duty and affiliation. It is first identified through the editorial conventions that give it its recognisable structure.

10The epistolary nature of the address to the reader is more difficult to assess formally. In terms of taxonomy, the terms used are more varied than for the dedication: addresses to the reader are called interchangeably “preface”, “epistle”, simply “to the reader” and even sometimes “prologue”. According to Philippe Schuwer, these different denominations probably allude to different generic precedents, which influenced the later uses: the term “prologue” has dramatic origins and is more likely to be used in literary works, the term “preface” is used first in Greek and Latin historical writing and in theology and therefore has a more authoritative ring (Schuwer 2011, 342). This variety of denominations and generic connotations may be a reason why generally speaking the readerly address adopts the material signs of epistolarity (salutation formula, subscription, signature, date and place of composition) more rarely than the dedication. Yet it is also sometimes called “an epistle”, as is the case in Thomas Walkley’s expression “a book without an epistle”, which is probably not alluding to a dedication since his address is entitled “the Stationer to the reader”. More than formal characteristics, what is at stake with epistles to the reader seems to be the scope of the address.

11For example, in a similar vein to Walkley’s description, Michael Drayton also defines readerly addresses as epistles that are expected to perfect a book and make it complete in The second part, or a continuance of Poly-Olbion: “Stationers […] have either despightfully left out, or at least carelessely neglected the Epistles to the Readers, and so have cousoned the Buyers with unperfected Bookes” (Drayton 1622, A2r). Drayton’s own address is in fact ambiguous in terms of epistolarity. It uses the epistolary codes such as the subscription and the signature but at the same time it embraces the fact that the fiction of an exclusive relationship with an individualised and identifiable reader is impossible in the printed book. This epistle to the reader is thus addressed “to any that will read it”, implying that the interpersonal relationship usually required by the dedicatory epistle – whether it be real or fictional – has disappeared in this readerly epistle, which anyone can read. In Drayton’s case, the epistle to the reader complements a dedicatory epistle addressed to Prince Charles. The combination of a dedication to a privileged reader (the royal patron) and an epistle to “any that will read it” suggests that the exclusive relationship between the author and his patron has not been completely replaced by the community of anonymous readers in the context of the expanding market of books, but this example still testifies to the difference of perspective in the kinds of epistolarity entailed in dedications and in addresses to the reader. If books often include both, and if both are “epistles”, they are nonetheless distinctive as they do not rely on the same type of exchange.

12It could be objected that readerly epistles sometimes mimic the kind of privileged relationship between author and patron. When addresses to the reader invoke “the gentle reader” or “the courteous reader”, both appellations implying strong class associations, they replicate the hierarchical and individualised bond with a patron (Brayman Hackel 2005, 116-117). But this symbolic extension has generic consequences in terms of epistolary writing, as made visible in Edward Sharpham’s dedication of Cupid Whirligig (1607):

Sir, I must needs discharge two Epistles upon you the one the Readers, that should be like haile shot that scatters and strikes a multitude, the other dedicatory, like a bullet, that aimes onely at your selfe […] yet I must confesse, I had rather expresse my love out of the flint, then my meaning in any part of the shot. I aime at you rather then the Reader. (Sharpham 1607, [A2r])

13As observed by Devani Singh, this dedication explicitely and boldly marks the precedence of the patron over the general reader and therefore takes an opposite stance to addresses that value the presence of a readerly epistle (Singh 2021, 296), as with the epistles by Walkley or Drayton and quoted above. But this passage also unambiguously sets into relief the distance between the dedicatory and the readerly addresses in their relationship to the logic of the epistolary genre. Though both conserve some of the codes of the epistolary exchange, their functions are different (in the first instance, expressing the author’s “love”, in the second, expressing the author’s “meaning”) and their scopes are opposite (“scatters and strikes a multitude” / “aimes onely at your selfe”). Sharpham makes it clear that the economy of the epistle as a genre is disturbed by the logic of the printed book and by the parasitical presence of an anonymous mass of readers.

14The two epistolary pieces usually placed at the front of the book are thus both shaped by the conventions of the epistle but they also display specific tendencies, whether it be in terms of form or in terms of perception, so that these two kinds of letters illustrate different relationships to the main text itself and to its readers. The liminary epistles are not just any epistle, they are delineated by the constraints of the book and not only of the epistolary genre. This also implies that we should not only look at the generic nature of the paratext itself but that we should also take into account the influence of the main text on the paratextual construction.

The imprint of genre: paratexts and the sonnet collection

15By definition, paratexts do not exist for themselves. They are fundamentally designed to introduce or bracket the main text. The deep connections between main text and paratext leads to believe that the generic nature of the main text is not neutral in the shaping of the paratextual material. This can be seen as early as in the title. According to Ceri Sullivan’s study, even when the generic nature of the main text is not explicitly indicated in the title (with a phrase such as “a ballad”, “a report” or “a sermon”), specific tags (such as “a conference”, “a relation”, “an exhortation”) are correlated to specific types of content and therefore read as generic labels (Sullivan 2007, 647-49). In other words, the genre of the main text informs the first paratextual item in a printed book, the title, even with terms that at first sight do not seem to indicate a precise genre. That the generic nature of the main text was made clear to the readers from the title-page is not surprising. How other kinds of paratexts such as dedications, readerly addresses or commendatory poems could be influenced by the genre of the main text is another issue.

16I mentioned earlier the case of drama, a genre whose paratextual evolution is well documented and which often reads as a symptom of the legitimation of the genre as a literary work. In Devani Singh’s recent article that compares the presence of dedications to a patron and readerly epistles in professional and non-professional drama, this narrative is complexified as more nuanced patterns of evolutions emerge from paratexts (Singh 2021, 287-290). Not only did dedications and addresses to the reader have different rates of occurrence in professional and non-professional plays, but these two kinds of paratexts were ruled by distinct logics that complicate the thesis according to which addresses to the reader progressively supplanted dedications, and that the figure of the reader became a more central socio-economic force in the book trade than the patron (Singh 2021, 294). This example invites us to further investigate paratextual variations for other genres.

18Because of the fluidity of the term “sonnet” in the early modern period, which could be used to label poems of various lengths, I have also included some collections which are not listed in Lyne and Shrank’s bibliography and in which the poems called “sonnets” do not follow the 14-line structure. These sonnets, that are not technically sonnets as we now define them, are taken into account in this study. They are part of a long poetic tradition that connects early printed miscellanies, in which the tremendous publication of these looser sonnets happened, with the later more standardized sonnet sequences (Shrank 2008, 31). These sonnet collections can particularly be found before the 1590s, when the vogue for the English sonnet sequence reached its climax. Our corpus will therefore include volumes like Songes and Sonettes (or Tottel’s Miscellany, 1557), Barnabe Googe’s Eglogs epytaphes and sonettes (1563), George Tuberville’s Epitaphes, epigrams, songs and sonets (1567) or Newe sonets and pretie pamphlets (1570) by Thomas Howell. Our corpus thus comprises 41 published volumes from 1557 to 1634, starting with Tottel’s Miscellany and ending with William Habington’s Castara. It excludes multiple editions when there is no change in the paratext, which is the case for almost all the collections that were printed again, except Samuel Daniel’s Delia (1592/1594) and Henry Constable’s Diana (1592/1595).

19The paratext I study here corresponds to the prefatory part of the book, i.e. dedications to patrons, addresses to the reader and commendatory verses praising the poet’s work. To establish the generic influence of the sonnets on these paratexts, I will first discuss their formal characteristics and I will then focus on the impact of thematic variations within the sonnets themselves on their paratexts.

Permeable forms

20While the great majority of prefatory paratexts in the early modern period is generally in prose (prose dedication and/or prose address to the reader) or, at least, in a mixture of prose and verse (when commendatory poems are included for instance), the first notable specificity in our corpus is the rather large number of paratexts that are in verse. In 10 volumes out of 41, the paratextual material is strictly in verse. This is less than the number of volumes that have paratexts in a mixture of prose and verse (18 items), but it is far from insignificant compared with the number of volumes with paratexts that are in prose only (13 items). Even if the number of collections with a paratextual material solely in verse is smaller, these volumes suggest that the formal nature of the paratext is not totally independent from the nature of the main text. The tendency here is that poetic forms in the main text call for poetic forms in the paratext.

21An example of this influence of the poetic formal characteristics of the main text to the paratext is particularly visible in John Taylor’s collection The sculler rowing from Tiber to Thames (1612). The poems are introduced by an abundant prefatory material, which starts with a sonnet to Taylor’s protector, followed by two other sonnets addressed to friends. These introductory sonnets are followed in turn by a series of commendatory poems of various lengths, written by friends praising Taylor’s art. This series ends with a return to Taylor’s voice, who addresses the reader in a verse prologue. This consistent use of verse throughout the prefatory material paradoxically gives a homogeneous nature to a book that is supposed to be defined by its variety, as the full title of the collection calls it: “a hotch-potch, or gallimawfry of sonnets, satyres, and epigrams.” The prefatory material, mixing many voices and poetic forms, also turns into a kind of poetic “hotch-potch.” But despite this variety, the strictly poetic nature of these paratexts contributes to creating a form of harmony and gives a formal coherence to the collection as a whole, uniting main text and paratext.

22Another notable aspect in Taylor’s collection is that these liminary pieces are not only poems but actual sonnets for a number of them. In our corpus, sonnets are regularly used in the paratext, either as commendatory poems or as dedications. In Richard Nugent’s Cynthia (1604), the paratext is even reduced to a single dedicatory sonnet to “the right honourable the Ladie of Trymlestowne.” The sonnet sequence that follows thus seems to derive naturally from this dedicatory sonnet. This is not just an authorial practice. For instance, the 1595 edition of Henry Constable’s Diana opens with a prose address by the printer to the reader, followed by a sonnet signed by the publisher, Richard Smith, addressed “unto her majesties sacred honorable maydes,” which faces the first sonnet of the collection. This dedicatory sonnet seems to help the reader transition from the printer’s address in prose to the sonnet sequence itself, through formal continuity. In a more fluid way, Michael Drayton’s Ideas Mirrour (1594) opens with a regular sonnet (with two quatrains and one sestet) to his patron Anthony Cooke, and is followed by a poem close enough to a sonnet although it includes extra lines (three quatrains and one sestet), which faces the beginning of a sequence of quatorzains. In this example, the sonnet form is dispersed and expanded outside the sequence itself.

23This diffractive nature of the sonnet form (or at least of verse) spreading out from the main text to the paratext confirms the flexible nature of the latter, which can easily adapt to the form of the text that it introduces. This permeability between text and paratext also reads as a sign of a consciousness of genre: the paratext surrounding a collection of sonnets cannot be exactly the same as the paratext surrounding another kind of text. This generic concern can be seen in our corpus in Alexander Craig’s Amorose Songes, Sonets, and Elegies (1606). In his short prose address to one of his fictional mistresses, Lithocardia, the poet writes: “I Feare to prefixe (Hono. Lady) to these few Poyems, a long Epistle” (Craig 1606, A8r). In suggesting that introducing a collection of sonnets by “a long epistle” could seem off-putting as well as disproportionate, Craig demonstrates a sense of generic decorum. This allusion to the expected length of the standard epistle before a sonnet collection, that is to say a collection of poems characterized by their brevity, is one more sign of the formal influence of the main sequence on the liminary pages. Length and proportion are essential ingredients in the matching process between main text and paratext. Craig eventually finds a formal solution to his generic concerns: whilst the first paratextual addresses to the reader and to his mistresses were in prose, he uses a sonnet in his final dedication to the queen, just before the beginning of the collection per se. Once again, poems in the main text tend to invite poems in the paratext.

Mistresses, patrons and friends: creating communities

24The influence of the main text on the paratext does not only manifest itself in formal terms, however. The sonnet provides a very good example of a genre that evolved thematically over a very short period of time and this thematic evolution seems to be mirrored by paratextual effects. Although many sonnets use a sentimental or erotic discourse as a way of dramatizing more complex social relationships, as Arthur Marotti has shown (Marotti 1982, 398), the sonnet is commonly associated with love. But this focus on love is particularly prevalent in the period when the Petrarchan sonnet came to dominate the field, in the 1590s, and gave a particular structure to the paratextual construction, with an important focus on the mistress figure in the liminary pages. Yet there is also another trend in the sonnet, which is more discursive and reflexive than amorous. Cathy Shrank in particular has described the mid-Tudor sonnet as a conversational poem turned towards horizontal networks of peers and addressing a great variety of themes, rather than concentrating on the exclusive bond between the poet and his mistress (Shrank 2008, 48-49; 2009, 102-103). Using our previous corpus, I have tried here to map out the influence of this thematic variant on the paratext.

25The gender of the dedicatees seems to be a significant factor of variation. 13 sequences are dedicated to women and 21 to men, the remaining 7 collections are just addressed to “the reader”, with no mention of gender. This gender division partly coincides with a thematic division. While collections of love sonnets are interchangeably dedicated to male or female figures, collections of non-amorous sonnets seem more likely to be addressed to men (Table 1). There are two exceptions nonetheless: Henry Lok’s Sundry Christian passions (1593), dedicated to Queen Elizabeth, and Nicholas Breton’s The Souls harmony (1602), dedicated to Lady Hastings. These two exceptions to the overall tendency can be explained though. Both these sequences are collections of religious poems and therefore suitable for a lady dedicatee. The Queen more specifically is not an ordinary female dedicatee, especially in the context of religious writing, in which she can also be addressed as a monarch and as a religious authority. Besides, in Lok’s case in particular, devotional sonnets remain influenced by a Petrarchan love rhetoric (Stull 1982, 79-80) that expresses the “Christian [passion]” between the soul and God. The courtly rhetoric of love sonnets is thus reused in religious writing, so that the presence of a female figure in the dedication is all the more appropriate. Because of their specificities, these religious sonnet collections dedicated to women could therefore read as false exceptions to the tendency in the table, that associates non-amorous collections to male dedicatees for the most part.

Table 1: Dedications, themes and gender.

	Love as main topic (27 collections)	Other topics (14 collections)
Male dedicatee	12	9
Female dedicatee	11	2
Unspecified (“to the reader”)	4	3

26If we consider again the figures in Table 1, it can be argued that in 11 instances out of 13, a collection of poems dedicated to a woman is about love. In terms of paratexts, the result of this distribution is that dedications to women tend to blend the figure of the mistress and the figure of the patroness. The love rhetoric directed to a female figure in a dedication must therefore not only be interpreted as a variation on the patron/client relationship, expressed in erotic terms, but also, in our perspective, as another symptom of the generic contamination of the sonnet, which disseminates its thematic characteristics into the paratext.

27A visible example of this contamination can be seen in the changes brought to Samuel Daniel’s dedication to the Countess of Pembroke in the multiple editions of his sonnet sequence, Delia. In 1592, Daniel dedicated the sequence for the first time to Mary Sidney Herbert. His prose epistle reads as a first humble appeal to a famous and influent protector to defend his poems against Thomas Newman’s parallel unauthorized version, which was included in Syr P.S. His Astrophel and Stella that same year. In 1594, when the sequence was printed again, Daniel changed his dedication. Not only did he turn it into a sonnet – another marker of the formal dissemination of the genre – but he also blurred the distinction between the Countess and the figure of Delia. Mary Sidney is turned into the inspirer of the poems, as if she had been present before the writing, which was not the case in the previous edition: “my humble Rymes / Which thou from out thy greatness doost inspire” (Daniel 1594, A2r). Though this change is also probably due to the evolution of Daniel’s relationship with Mary Sidney Herbert, who turned from a prospective patron in 1592 to an effective protector in 1594, the transformation of the dedicatory epistle illustrates how Daniel adjusted his paratext so as to create a coherent unit, delineated by the constraints of a genre in the making, the love sonnet sequence. This unit is both formal, with the use of a dedicatory sonnet to open a sonnet sequence, and thematic, as the paratext takes part in constructing the unique bond between the poet and the mistress-figure.

28Even when there is no real female protector to be addressed, the figures of the mistress and of the patron tend to overlap. This is the case for instance in the first edition of Diana by Henry Constable in 1592. The collection opens with a sonnet addressed “to his absent Diana” followed by an address “to the gentlemen readers.” This positioning turns the sonnet to Diana into a kind of dedication, as it reads as a poem that belongs to the prefatory material and not to the main sequence of sonnets. Even if Diana is not clearly identified as a female protector, she is the expected dedicatee for a collection of love sonnets, despite her “absence.” This sonnet to Diana thus fills a slot and satisfies a generic expectation that requires a female presence in the dedicatory pages of a love sonnet sequence. A similar example of the lingering presence of the female dedicatee can be seen in the dedication for Emaricdulfe (1595), by E. C. The collection is dedicated to two male friends of the poet’s, but the dedication immediately alludes to the first alleged muse and addressee of the sequence: “I tooke in hande my pen to finish an idle worke I had begun, at the command and service of a faire Dame” (E. C. 1595, A3r). The dedication is transferred to the poet’s friends, who have to protect the name of the lady as much as the sonnets themselves (“not doubting but that you will vouchsafe for my sake, to maintaine the honour of so sweete a Saint”), but the lady is nonetheless present in the background. These allusions to a lady dedicatee may be based on real social relationships, as in Daniel’s case, or on a fiction, as is probably the case for Constable’s Diana. More importantly, these hints at the existence of a female muse/patron are so frequently found in dedications for sonnets collections that they seem to correspond to a generic norm, probably influenced by the Petrarchan model of the sonnet, with Laura as the ever-present and ever-elusive female inspirer.

29If the paratexts in sonnets collections often rely on this generic bond between the poet and a mistress/patron figure, how do they consider the presence of readers other than the dedicatee? The question can be illustrated by looking at addresses to the reader and commendatory poems in our corpus. Indeed, when the paratextual material includes an address to the reader or commendatory poems by friends and peers praising the collection, it openly implies that the poems have had or will have more readers than just the dedicatee. The tendency shown in Table 2 suggests that collections dedicated to a male figure are more likely than collections dedicated to a female figure to allude to the non-patronal readers, whether it be the “general reader” as in readerly addresses or specific readers that have provided encomia. While the dedication to a woman is more frequently exclusive, the dedication to a man tends to be inclusive. Collections dedicated to a man are more frequently accompanied by prefatory elements that place the poet among a community of men and readers. Conversely, in collections with a female dedicatee, the bond between poet and dedicatee is more rarely broken by the mention of a reader who is external to the exclusive relationship between the poet and his muse/patron/dedicatee.

Table 2. Presence of readerly addresses and/or commendatory poems according to the gender of the dedicatee

	Male dedicatee = 21	Female dedicatee = 13
Presence of readerly addresses or commendatory poems	16	6

30This construction of a homosocial community of readers through poems (and sonnets in particular) has been charted by Wendy Wall in The Imprint of Gender (Wall 1993, 23-109; 188-191). According to her, “gender is the axis on which writers register a specific conception of the literary text, one which de-emphasizes the author in lieu of his role in a set of social relations and bonds” (ibid., 40). The most obvious paratextual manifestation of these communities of men is commendatory poems. When commendatory poems occur in a collection in our corpus, it is always in a collection dedicated to a male figure (Table 3). This reinforces the image of a coterie or at least a network of peers or friends that escort the poet’s work.

Table 3. Distribution of commendatory poems according to gender of the dedicatee.

	Male dedicatee	Female dedicatee
Collections including commendatory poems	9	0

Table 4. Distribution of commendatory poems according to theme.

	Love	Other topics
Collections including commendatory poems	6	3

32The example of Barnabe Googe’s Eglogs epitaphs and sonnets (1563) is particularly telling. Googe’s poems could at first be labelled as a love collection, tapping the vein of light pastoral poetry and romance. But Googe uses these genres in a strongly didactic and anti-erotic perspective, so as to warn other men against love (Shrank 2008, 36-37). When a female figure appears in his collection, it is rejected by the moral community that surrounds the poet. This community is put to the fore in the opening commendatory poem by Alexander Nevyll: “Submit thy selfe to persons grave, whose Judgement ryght alwayes / By Reason rulde doth ryghtly judge, whom Fancies none can charme” (Googe 1563, A2r). Googe does not write for a mistress/patron, but for a very specific kind of community, that of the Inns of court. It is made visible in the dedication, addressed to “William Lovelace, Esquier, Reader of Grayes Inn”, while the prefatory material also includes two addresses by Laurence Blundeston, another member of Gray’s Inn. As shown by Jessica Winston, these prefatory poems find an echo in other poems in the collection. This system of echo is reminiscent of a logic of answer poetry, where poems respond to each other, and which is:

not competitive or erotic but appears instead as a form of amiable, if morally stringent, conversation. Googe and his friends do not cement a literary community through the medium of love poetry. They negate the presence of women and avoid competition” (Winston 2016, 88-89).

33Googe’s collection thus provides us with an example of type of sonnet collections which is not centred on the figure of the beloved and on romantic pursuits voyeuristically assessed by a community of readers, contrary to the standard model of the sonnet sequence, with a female dedicatee and gentlemen readers (Wall 1993, 188-191). It suggests instead another kind of perspective on the logic of the sonnet, turned towards collective moral elevation. This thematic variation has paratextual effects, which are different from those that can be found in love sonnets.

Here learne affects to rule, and youth in care to spende
Beware thou mayst by others harmes, how youthly toyes do ende:
Here reape, with other frutes, precepts of mortall minde. (Howell 1570, [A4r])

35The prefatory material thus opens a space for conversation between the poet and his readers: not only those who have already read the collection – which includes not only Keper but also Howell’s “faithful and Fellow Maister Henry Lassels” (his dedicatee) and William Howell (his brother) who also adds a short commendatory poem – but also those who should read it. Contrary to the terms used in readerly addresses for love sonnets, the intended reader is not “gentle” or “courteous” but “upright” (Shrank 2008, 38), thus also entering the moral community forged in the preliminary pages. This example shows two things: firstly, that the presence (or not) of commendatory pieces in sonnet collections should impact our reading of the whole sequence. Secondly, that encomia, like dedications and readerly addresses, symptomatically reveal variants within the sonnet genre itself. Over time, sonnet collections adopted various topics and social logics, that implied different paratextual approaches.

Conclusion

36Early modern paratextual material often requires a generic reading. This is not only because various types of paratexts function according to specific generic conventions that have evolved from pre-existing genres “by inversion, by displacement, by combination” (Todorov 1976, 161), as the epistle for dedications and readerly addresses have. It is also because when focusing on a single genre, it can be argued that generic innovations find counterparts in paratextual variations. As a final example of how a generic reading helps understand the differences and similarities among paratexts, I will compare the dedications in two collections of devotional sonnets, Barnabe Barnes’ A divine centurie of spirituall sonnets (1595) and Henry Lok’s Sundry Christian Passions (1593). Using our double perspective (paratexts follow generic conventions and paratexts are influenced by the genre that they introduce), I will first look at the enduring influence of the epistolary genre in these dedications, and then I will focus on the generic contamination of these paratexts by the main texts.

37Barnabe Barnes’ A divine centurie of spirituall sonnets opens with a prose dedication to the Bishop of Durham, which fully embraces the epistolary model. This “epistle dedicatorie”, according to the running title, starts with a salutation formula and ends with a subscription, a signature, and the date and place of composition. Formally speaking, this dedication thus closely follows the conventions prevailing in prefatory material. In terms of content, the dedication unsurprisingly emphasizes the dedicatee’s virtues, the poet’s good intentions and his authentic devotion. The liminary material in Henry Lok’s Sundry Christian Passions provides an interesting counterpoint. The collection opens with a sonnet dedication to the Queen, but this poem is introduced by a salutation formula, thus tapping the epistolary vein. Besides, the salutation is a chronogram that gives the date of composition: “To the right renowneD VertVoVs VIrgin ELIzabeth, VVorthy QVeene of happIe EngLanD, her hIghnesse faIthfVL subIeCt, Henry Lok, VVIsheth Long Lyfe, VVIth eternaL bLIsse. IVne. VII” (Lok 1593, sig. *iijr). The day and month are explicit (the 7^th of June) but the year is expressed by the addition of the Roman numerals: “DD. C. LL.LL.LL.LL.LL. VV.VV.VV.VV.VV.VV. IIIIIIIIIIIII”, that is to say 1593. This address thus subtly reintegrates the epistolary origins of the dedication, through a typographical play, without slavishly following the editorial model of the “epistle dedicatorie.” Reading the dedication as a genre itself therefore explains the lingering presence of epistolary markers to introduce a poem, while Barnes more overtly adopts the letter model.

38Conversely, if we consider the dedication not only as a genre but also as a response to a genre, then Barnes’ and Lok’s texts also follow two different logics. On the one hand, Barnes chooses a dedicatee whose spiritual authority gives legitimacy to the collection and whose patronage would endorse the book as a valuable religious guide. On the other hand, Lok chooses a dedicatee who can be addressed as a monarch, as a religious authority and as an unreachable female figure. Hence in Lok’s case a dedicatory sonnet with Petrarchan undertones to introduce a collection in which the love rhetoric can be adopted by the devout soul.

39In other words, though genre is certainly not the only ingredient that shapes the front matter, it is a legitimate factor to explain some of the differences in paratextual material from one book to the other. We have seen how the specific kind of epistolarity that happens in dedications and readerly addresses is shaped by the conventions of letter-writing but that it is also fashioned by the constraints of the printed book. The epistolary logics are not the same in dedications and in readerly addresses, because they require a different perception from the readers. Conversely, reading paratexts generically enables to see that the front matter is not impervious to what it introduces. Whether it be formally or thematically, the paratextual material often matches the genre of the main text and should therefore sometimes be less considered as a frame than as an essential component of the text itself.