The Progeny of Print: Manuscript Adaptations of John Speed’s Chaucer Engraving

4John Speed’s engraved portrait of Chaucer, made for the 1598 edition of the Workes, relies rhetorically upon a manuscript tradition of Chaucerian portraiture to establish its authenticity. During the seventeenth century and onward, Speed’s printed plate exhibited a high degree of mobility, being removed from the editions and reappearing in other Chaucerian books and in later manuscript replicas. This essay tracks the movement of the portrait across the permeable boundaries of print and manuscript, arguing for the role of print culture in its dissemination and as the cause of its eventual reappropriation into hand-drawn and painted forms.

turn of the century, engravings were a desirable print commodity among the book-buying public, as much for their beauty as for their curiosity. It is unsurprising that Speed's copperplate Chaucer portrait, made for Thomas Speght's first edition of the poet's Workes (1598), was advertised prominently on that book's title page, at the head of a list of the new edition's vendible features: "His Portraiture and Progenie shewed" ([a]2r).
Yet for all its technological innovation, Speed's image is everywhere marked by iconographic and textual cues that assure viewers of its own authenticity. The portrait depicts the poet standing in the central panel of an image titled "The Progenie of Geffrey Chaucer." That heading is misleading, for Chaucer is flanked here by a series of medallions that not only contain the names of his descendants but also trace his links back to England's noble and royal families via his marriage to Philippa Roet ( fig. 1).
It is her father, "Payne Roet Knight," who appears atop the genealogy as its symbolic figurehead. The base of the image depicts the tomb of Thomas Chaucer and his wife, Maud Burghersh, in the parish church at Ewelme. Speed's engraving of the tomb reproduces its twenty-four shields representing the family's illustrious pedigree. In framing Chaucer, claimed as the first and "famous" national poet, this heraldic iconography celebrates incipient Englishness itself.
To this work, as to his magnum opus, The History of Great Britaine under the Conquests of the Romans, Saxons, Danes, and Normans, and its accompanying maps, The Theatre of the Empire of Great Britaine (1611-1612), Speed brought the genealogist's enthusiasm for order and the antiquary's diligence . He took pains to endow the picture with credibility, as is evident on the printed page. The medallions that authoritatively cluster around the figure of Chaucer confer historicity and visually sidestep the fact that all of the poet's noble relations were acquired by marriage rather than by a distinguished lineage that was his own. The finely wrought depiction of the tomb is likewise presented as a faithful representation of the monument at St Mary's Church in Ewelme, Oxfordshire. Elsewhere in the Workes, Speght writes of the portrait that "M. Spede . . . hath annexed thereto all such cotes of Armes, as any way concerne the Chaucers, as hee found them (travailing for that purpose) at Ewelme and at Wickham" (C1r). Most telling, though, are Speght's and Speed's efforts to convey the verisimilitude of Chaucer's printed likeness itself.
The central panel of Speed's engraving features a full-length depiction of Chaucer, standing and holding a rosary. 2 An object that is per- haps a penner, or pencase, hangs from his neck, signifying his status as a man of letters and connecting the text printed in Speght's edition to its written manifestation as a product of Chaucer's hand. 3 A panel of text positioned underneath the figure of Chaucer announces its provenance: The true portraiture of GEFFREY CHAUCER the famous English poet, as by THOMAS OCCLEVE is described who liued in his time, and was his Scholar.
The caption is unambiguous in its staging of the image's authenticity: this is a "true" representation of Chaucer's likeness, as reported by the poet and clerk Thomas Hoccleve, who knew him well. Speght confirms the image's Hocclevean origins when he notes elsewhere in the edition that "Occleve for the love he bare to his maister, caused his pic-ture to bee truly drawne in his booke De Regimine Principis, dedicated to Henry the fift: the which I have seene, and according to the which this in the beginning of this booke was done by M. Spede" (C1r).
Before I assess the implications of this claim to Hocclevean provenance, it is worth noting that there is no robust surviving evidence of the engraving's origins, and Speght's editorial assurance that the portrait was a copy or adaptation from a Regement manuscript therefore remains unproven. Portrait miniatures of Chaucer survive in at least three manuscripts of Hoccleve's Regement, of which the earliest, and the best candidate of the three for having been Speed's source, is British Library,Harley MS 4866. 4 Two additional Regement manuscripts, British Library, Harley MS 4826 and Arundel MS 38, each contained a painted Chaucer portrait, both now excised (Pearsall,"Portraits" 289).
If we disregard Speght's pledge that Speed's source was a Regement portrait, other candidates for its exemplar emerge. 5 British Library, Cotton MS Otho A.XVIII-a fifteenth-century miscellany that was nearly destroyed in the Cotton fire of 1731-formerly contained a Chaucer portrait seen and described by leading antiquaries of the day. 6 The prolific George Vertue, who became official engraver to the new Society of Antiquaries in 1717 and who himself produced four engraved Chaucer portraits, noted that the Cotton image was painted on a single vellum leaf pasted into the end of the book and featured "his right hand holding the penknife, & his left the beads," as in Speed (British Library, Additional MS 23070, fol. 43v). In surviving Regement manuscripts, by contrast, Chaucer's right hand points from the margin toward the text of Hoccleve's poem. 7 Vertue thought that the Cotton image might be the model for the portrait added to Chaucer's tomb at Westminster Abbey in 1556 and for the picture of Chaucer made by Speed for Speght's edition. 8 But between Harley 4866 and the Cotton image, and even "after several years enquiry," he, too, was unsure about which portrait had priority. 9 The unsolved case of Speed's exemplar could have been shelved at this point, if not for the survival of another early manuscript portrait, British Library, Additional MS 5141. This single vellum leaf features a portrait of Chaucer similar to Speed's, with the same stance and accessories on one side and a vita of the poet copied in a sixteenth-century secretary hand on the reverse. 10 Because of its relatively early date and its likeness to Speed's portrait, scholars have considered it to be his model (Pearsall,"Portraits" 295;Driver 247). Moreover, Pearsall believes that it is "entirely possible" (295) that Additional could be the very leaf that Vertue saw in the Cotton manuscript, if it had somehow been removed from the book before the fire. This possibility is enticing but unlikely. There are good reasons to doubt the suggestion that Additional MS 5141 is the presumed-lost portrait formerly in the Cotton manuscript and described by Vertue. First, Vertue makes no mention of the vita on the reverse of the portrait, a substantial text of forty-two lines in which he would have had some interest as an antiquary and a scholar of Chaucer. Second, Vertue describes the material condition of the portrait he saw as "a leaf of Vellum pasted at the end of that book [Cotton Otho A.XVIII]," which might explain why he didn't notice the vita on the leaf's verso (British Library, Additional MS 23070, fol. 43v). But there are no visible traces of adhesive on the verso of the portrait, nor any signs of damage caused by its having been pasted onto another leaf. Finally, it is unlikely that the Cotton portrait that Vertue described is the same as the still-extant Additional MS 5141 because of a detail of dating: Vertue notes that the Cotton portrait was probably "done immediately after his death, being dated 1400 in this manner, the very year (at least) that he died" (British Library, Additional MS 23070, fol. 43v). Crucially, the date that the Additional portrait bears on its upper right corner is 1402. Beyond the Additional MS, and as Pearsall notes, the role of the Westminster tomb portrait as a progenitor of the portraiture tradition followed by Speed is also undetermined ("Portraits" 296).
No good, still-extant candidate for the source of Speed's engraving emerges from this confusion of exemplars and copies. The lost Cotton portrait might appear to be the best contender, and Speed, familiar to the men who made up the Elizabethan Society of Antiquaries, is known to have borrowed other manuscripts and coins from Sir Robert Cotton as models for the engravings in his History (Driver 238). Ultimately, however, any affinity between the portrait made by Speed and that once owned by Cotton can be judged only through surviving reports such as Vertue's. Meanwhile, the avowal that Speed used a Regement exemplar for his Chaucer engraving is unverified and unverifiable, given the current evidence. But I do not think there are adequate grounds to distrust the Hocclevean provenance claimed by Speght, who had the fastidious John Stow and, later, Francis Thynne looking over his shoulder as he produced the editions (Pearsall,"Stow";Cook,. Importantly, whatever the source of the 1598 Chaucer engraving, it is clear that Speed and Speght had good reason to align their project with Hoccleve's. In the Regement, a literary petition for the patronage of Prince Henry of Monmouth (and later Henry V) written in 1411, Hoccleve proves his close relationship with the now-deceased Chaucer in pictorial form: That, to putte other men in remembraunce Of his persone, I have heere his lyknesse Do make, to this ende, in sothfastnesse, That thei that have of him lest thought and mynde By this peynture may ageyn him fynde.  As David Carlson has suggested, Hoccleve supervised the production of presentation copies of the work, and the success of his bid to Henry relied on the portrait's "lyknesse" to Chaucer (287). Hoccleve's desire was to make not simply an effigial mnemonic aid but also a realistic mimetic portrait of Chaucer's "lyknesse." This was novel because individualized faces were rarely employed in medieval portaiture when iconography or arms alone could identify a figure. Alongside a few continental examples, Chaucer is therefore regarded as one of the first European vernacular authors to have a portrait attested in copies of his works. 11 As a visual invocation of the poet's near-forgotten likeness, the Harley image has been most frequently interpreted as an attempt to produce an authentic, individualized portrait of Chaucer (Carlson 294;Pearsall,"Portraits" 288).
In this context, Hoccleve's manuscript image of Chaucer recollected "in sothfastnesse" was the ideal exemplar for a new mode of depicting the poet's "true portraiture" in print. Whereas Hoccleve may have reasoned that close affiliation with and instruction under Chaucer would aid his plea for Henry's patronage, Speed relied on the putative intimacy between Chaucer and the clerk to authorize his engraving. The editor, too, assured the reader that the portrait appears in a book by "Chaucers Scoller" Hoccleve, testifying to "hav[ing] seene" it before (C1r). Speed and Speght thus vouched for the accuracy of their representation of the Chaucerian "cotes of Armes" and portrait respectively; like that of Hoccleve, these claims were supported by eyewitness accounts that served as authenticating credentials for the artifacts they described. In its printed incarnation, the image echoes Hoccleve's pledge of the portrait's authenticity-and deftly manages to appropriate it. The antiquaries' claim that the printed image is Chaucer's "true portraiture" is conveniently tethered to the authority of Hoccleve and his book, even as it ventures forth in the fashionable form of metal engraving.
As it appeared in 1598 (and in the later edition of 1602 and its 1687 reprint), Speed's portrait of Chaucer was a printed surrogate of a manuscript original-a representation of another, older image that was itself ultimately a "remembraunce" of Chaucer the man. With each new iteration of his likeness, the poet receded further from both historical view and living memory, but those who reproduced it took care to trans-fer its authenticating hallmarks and to emphasize their contribution to its continued transmission. This narrative is a familiar one in studies of Chaucer's early modern reception. As James Simpson has argued, Chaucer's perceived absence provided the linchpin upon which turned the machinery of his early modern prominence, as the dead poet's corpus was recast as a textual monument to be recovered through archaeological and philological work. 12 What was true for the early philological investigations into Chaucer's works could also obtain for his first engraved portrait, as the recuperation of his physical likeness became a worthwhile antiquarian mission akin to the unearthing and assembly of his Life. 13 In printed form, Speed's Chaucer portrait vastly exceeded the reach initially anticipated by Hoccleve when he commissioned multiple manuscripts containing the poet's likeness. With this wider dissemination and the ability to achieve new levels of realism in portraiture, Speed's engraved portrait could eventually unseat Hoccleve's as the definitive representation of how Chaucer looked. Although Speed's portrait itself ingeniously claims a Hocclevean provenance, the printed image also takes on the authority of the older manuscript tradition and summons the hallmarks of manuscript authenticity-what Siân Echard has called "the mark of the medieval" (vii-xvi)-to do so. As the following discussion illustrates, later generations responded enthusiastically to this printed image of Chaucer, which, alongside its technical novelty, could nonetheless claim to be "true."

Chaucer on the Move
The starting point for my work on the Chaucer portrait was the observation that several of the copies I have examined are missing their Progenie leaves. 14 On the other hand, as Hope Johnston has demonstrated, the plate intended for Speght's edition was belatedly added to other books and survives in copies of John Stow's 1561 Chaucer edition in at least three cases. 15 Outside of printed Chaucer editions, CUL MS Gg.4.27, the remarkable fifteenth-century manuscript containing many of Chaucer's collected works, was repaired and supplemented in about 1600 by the antiquary and amateur herald Joseph Holland, who added Speed's plate, with its arms gilded and tinctured (Cook, "Holland"). The portrait leaf also appears as a frontispiece to a seventeenth-century manuscript of Sir Francis Kynaston's complete Latin translation of the five books of Troilus and Criseyde, which survives in the Bodleian Library. 16 Like the holes left in places where illuminated initials have been excised from manuscripts, the absence of the Progenie leaf in some cop-ies of Speght might signal its high cultural value for enthusiasts and collectors who envisaged other uses for it. What is certain is that these repurposings of the plate in books made by Holland and Kynaston in the seventeenth century-outside of the 1598, 1602, or 1687 Workesdemonstrate the portrait's status as a vendible and prized accessory for other books containing Chaucer's or Chaucerian works.
Even when it remains intact within copies of Speght, the plate has survived in a range of positions. In copies I have seen, it is most frequently positioned facing the poetic dialogue "The Reader to Geffrey Chaucer" by the anonymous "H.B." 17 This positioning is especially apt in the 1602 edition, where the portrait directly precedes the verses titled "Vpon the picture of Chaucer" composed by Francis Thynne for the updated publication. 18 But the plate often appears elsewhere within Speght, too, even in copies with early bindings. In a copy of the 1598 edition at Balliol College, it serves as a frontispiece to the book and faces a Canterbury Tales title-page border (normally found later in the same edition) that has been repurposed as the volume's main title leaf, where the original is missing. 19 In copies of the 1602 edition at the Queen's College, Oxford, and the Bodleian Library, the Progenie leaf appears between the title page and the dedication to Sir Robert Cecil. 20 Another copy (1602) at Caius College, Cambridge, is in a contemporary binding and faces the page titled "The Life of our Learned English Poet, Geffrey Chaucer." 21 Meanwhile, a copy (1602) at the Huntington Library contains an inlaid plate, apparently a later imitation of Speed's original, facing Francis Thynne's verses on Chaucer's picture. 22 Inserted plates generally seem to have had a standard position within books , and in his editions Speght referred to Speed's plate as being in the "beginning of this booke" (C1r). Both within and beyond copies of Speght, these survivals of the portrait in atypical positions prove that it was a highly mobile artifact whose success as a print commodity is amply attested by its reception at the hands of readers.
In this respect, the Speed plate exemplifies some of the characteristics of what Remmert has called the "itinerant frontispiece," a term that demonstrates the separateness of the frontispiece's materiality (268; . Far from being confined to their original bibliographical contexts, such plates traveled from book to book and out of books and into new contexts. Where Chaucer's engraving is concerned, this travel radiated outward in several directions: movement of the plate to different locations within individual copies of Speght's works; back in time, into prior editions like Stow's; forward in time as they were rendered again by later collectors in the older medium of manuscript.
The remainder of this essay traces this extraordinary reception, in which Chaucer's printed portrait enjoyed a rich and unforeseen afterlife in manuscript, in tandem with the wider vogue for printed author images that would emerge in the seventeenth and eighteenth centuries. This winding transmission from manuscript, into print, and back again to manuscript suggests the portrait's attractiveness and the critical role of Speght's edition in its dissemination.
Just as Speed's engraved portrait found its way into other printed and written copies of Chaucer's works, so, too, did manuscript representations of the same image. Skilled later replications of Chaucer's portrait, strongly suggestive of Speed's, appear in a number of early printed books. A nearly perfect copy of Caxton's first edition of The Canterbury Tales held at the British Library (De Ricci no. 22:1) now has as its frontispiece an eighteenth-century painted portrait of Chaucer in the same orientation and style as Speed's and surrounded by a colored and gilded foliate border evocative of the illuminations found in fifteenth-century English manuscripts ( fig. 2). 23 A later edition of Chaucer's Workes, published by William Thynne in 1542 and now at Columbia University likewise has a later watercolor rendition of the portrait inserted as a frontispiece. 24 This version, however, also features Chaucer's arms, which are borne on a shield resting on a rock in the image's background. In Takamiya MS 32, formerly known as the Delamere manuscript, appears another modern variant, this time with Chaucer's arms displayed in the top lefthand corner of the leaf, as in Additional MS 5141 ( fig. 3).
A final example of a Speed-style manuscript portrait appearing in a printed copy of Chaucer's works comes an edition of Speght (1602) at Trinity College, Cambridge, where the Progenie leaf is missing but a facsimile tracing has been pasted onto the book's front board, complete with the genealogy, heraldic shields, and familial tomb as originally rendered by Speed ( fig. 4). 25 In all but the last case, it is impossible to prove that these manuscript portraits were based on Speed's Progenie page rather than on another exemplar, such as Additional MS 5141. (And of course, it is possible and even likely that Additional itself may be an early copy of Speed's plate.) What is indisputable is that all of these manuscript imitations cater to a desire to locate the author's image in copies of his works. This is a phenomenon older than print, but in Chaucer's case it took Speght's editions to formalize it. To these Speed-style manuscript portraits in copies of Chaucer may be added two iconographically similar items in contexts outside of Chaucer's books: an undated manuscript   fragment at Stanford University and a drawing of Chaucer used as an example of medieval clothing in John Aubrey's Chronologia Vestiaria. 26 The Stanford fragment is on vellum and features a colored miniature of Chaucer in the pose and configuration that Speed made famous. It is damaged and difficult to date but appears to be older and less skilfully executed than the painted portraits in copies of his works already described, and has been suggested to be a copy of Speed's plate (Jordan).
By contrast, the portrait in Aubrey's history of costume is clearly derived from Speed and dates from the 1670s, when the Chronologia Vestiaria is estimated to have been written (Bennett 464). Here, Chaucer's portrait shares a page with other figures drawn from English church monuments-among them Sir Thomas De Littleton, who, Aubrey carefully notes, "is pourtrayed in this Habit in his Monument in the Cathedral church at Worcester: in brasse: he was a Judge. / Lived in the reigne of Edw. 4" (Bodleian Library, MS Top.Gen.c.25., fol. 202r). Aubrey's Chaucer appears at the foot of the same page but is rotated ninety degrees to the left in order to fit. The image might be a trac-ing from Speed and, in Aubrey's characteristic mode, seems to have been crammed in to save space on the page. Aubrey also transcribed from Speed's plate the caption concerning Hoccleve's status as Chaucer's Scholar. Sometime later, perhaps, he added to it a further short note about Chaucer's dates of birth and death, which appears in a different ink. It is striking that Aubrey treated Speed's portrait with the same reliability as the church monuments he documented elsewhere on the page. Its credibility might have rested on a putative memorial description in an unspecified copy of Hoccleve's Regement rather than on tangible evidence carved in stone or brass, but Aubrey's faithful recording of Speed's portrait and its caption alongside other graven monuments suggest that he took its truth-claim seriously.
Thus stand two intertwined traditions of Chaucer portraiture, in print and in manuscript. The motivations behind many of the Speedstyle manuscript portraits are shrouded in obscurity. Their hazy origins, together with the uncertainty surrounding Speed's exemplar, make the exact relationships between the engraving and its hand-drawn counterparts speculative. Could images such as the Additional or the Stanford fragments be copies of earlier Chaucer portraits like that belonging to Cotton, but now lost? Or might they be early modern copies of Speed's plate? Could either have served as Speed's exemplar? As I have argued, there is no compelling candidate for Speed's exemplar currently known, yet it is also the case that the models for most of the surviving manuscript renditions described here are equally uncertain. Nonetheless, their existence proves that this particular version of Chaucer's likeness-the full-length portrait, with the poet standing and holding rosary and penner-enjoyed an atypical mobility and multiplicity in the seventeenth century and beyond. Echard argues that the print reception of medieval texts is characterized by an "impulse to facsimile," a desire by later cultures to replicate the physical forms and material details of the medieval book (xi, xv, 6-20, 198-216). As we have seen, medieval images, and author portraits in particular, were attractive candidates for this type of replication, in manuscript as well as in print. In Echard's analysis, such images and their analogues might be regenerative, and in their new incarnations they "participate in a process by which an image comes to stand in for a text, a tradition, and sometimes both" (19). The dizzying range of later lookalikes of Speed's Chaucer, together with the image's accelerated and unprecedented circulation in print, suggests the likelihood that the manuscript portraits were copies of the plate.
Like Speed's Progenie plate of Chaucer, which relies on Hoccleve's having "lived in his time," many of the manuscript images invoke the poet's ancient status, even if all but one of them (Munby a.2) exclude the genealogical tree and the later Chaucers' tombs. Yet these manuscript portraits take care to inscribe Chaucer's historical worth and stature in other ways. To the portrait in Takamiya MS 32, someone added the word "Chaucer" and the date "1400" in black ink on either side of the figure's feet, in a script imitative of black letter. The Stanford miniature, although not securely dated, has text on its verso that reads "Chaucer's portrait-S. xiv," written in faded red pencil. As discussed, the Additional MS 5141 fragment, which dates from the early modern period, bears the year 1402. There is no evidence to suggest that any of these artifacts has medieval origins. Still, each announces its ancient associations with Chaucer and "his time." While they do not explicitly pose as medieval artifacts, their manuscript form and their foregrounding of a storied past allow them to obscure their own histories and embody an authority that Speed's printed portrait could only ever claim to represent imperfectly in the medium of print. 27 Speed's engraved plate provided an enduring blueprint for how Chaucer looked, but its later imitators were impelled to render his likeness anew in manuscript images modelled after the older, full-length portrait even as newer depictions were disseminated in print. 28 Excluding the Aubrey drawing-which dates from the seventeenth century and was motivated by different intentions-the manuscript portraits in copies of Chaucer use their medium to stage a historical authenticity suited to the age of the fifteenth-, sixteenth-, and seventeenth-century works in which they survive and to the fourteenth-century poet who such books commemorate. This desire to enhance copies of Chaucer's works, both manuscript and printed, with the deluxe embellishment of a hand-painted portrait is especially transparent in the case of the portrait in the British Library Caxton (167.c.26), with its elaborate border illumination, but it is also true of the portraits in Additional (British Library, Additional MS 5141), Columbia (Columbia University, Phoenix P017.En1 B64 1542C), and Takamiya (Yale University, Takamiya MS 32), which are finely and perhaps professionally produced. The efforts of Speght and Speed may have invested the plate with a totemic significance for centuries to come, but the later Chaucerians who embellished these books restored some of its perceived authenticity by returning the portrait to its original medium.

Chaucer in Mixed Media
Early modern print culture in general and Speght's edition in particular created the conditions for the remarkable spread of these images across print and manuscript. The 1598 likeness of Chaucer is an early and important example of the engraved author portrait in an English book. 29 In this period, published works of poetry and prose were unlikely to contain portraits of their authors (Cooper and Hadfield 411). The portraits of most contemporary poets living and writing at the time, including John Donne, Edmund Spenser, and Sir Philip Sidney, would reach print much later-and posthumously (Orgel 9-10; Cooper and Hadfield 2). Before the 1630s, in fact, most poets would receive a portrait in print only if they were dead, a trend that Leah Marcus reads as motivated by an impulse to "preserve the illusion of human presence within a medium that was vastly expanding the physical distance between writers and prospective readers" (199). For long-dead auctores like Chaucer and Homer, whose works predated print itself, that gulf was wider still. In such cases, the presence conjured by a portrait served to recall rather than bridge the temporal chasm between author and reader and made way for the author's philological recovery in print.
Some of the earliest English books to contain printed author portraits are translations: Harington's Ariosto (1591), Florio's Montaigne (1613), and Chapman's Homer (1616). These books bear portraits of their translators instead of (or, in the case of Homer, in addition to) images of their first authors. The translator portrait is a reminder of reading as a mediated experience, one made possible by the translator's efforts. Although Speght's Chaucer is not a translation, the editor is implicitly framed as someone who has "made old words, which were unknown of many, / So plaine, that now they may be known of any" ([a]6v). The visual rhetoric of Speed's Chaucer portrait, like that of contemporary translations, thereby reinscribes a sense of the work's inaccessibility, save for the editor's or translator's intervention. The stylized portrait confers a formality befitting its distant subject and foregrounds the labors of those responsible for its recovery-in this case, Hoccleve, Speed, and Speght. In the early years of the market for engraved portraits, Chaucer was the ideal subject, and Speght's edition was a suitable medium for its transmission.
By the middle of the seventeenth century, advances in technology enabled engravers to produce larger quantities of prints from a single metal sheet and eventually spawned a trade in collecting so-called "portrait heads" intended to be bound with similar images in one volume . (Watt 142;Griffiths 21). Later in the century, a vibrant trade in printing, recycling, and collecting images developed, giving rise to the "itinerant frontispieces" previously discussed: some images in books were reprinted from existing plates made for other volumes; others were produced to be inserted into books that had already been published or were created in anticipation of future editions, some of which might never see publication; still others were printed to serve as the frontispieces to books but might be sold separately as a single print (Alexander 299). In 1700, Samuel Pepys compiled such engravings into a set of three albums, in which Speed's plate also makes an appearance among a group of "Poets, Comedians, & Musicians" (Chamberlain [2980/201]). 30 Like numerous other plates published for printed books, the Chaucer portrait first made by Speed had become a collector's item.
During the eighteenth century, the consumer-driven practice of extra-illustration was increasingly commercialized. Bespoke illustrated copies of Edward Hyde, Earl of Clarendon's History of the Rebellion (1702) and its later reissues were produced by print sellers and publishers; and in 1760, much to the delight of zealous collectors, James Granger published A Biographical History of England, from Egbert the Great to the Revolution, combining images with prose accounts of the lives of notable English figures, spread over four quarto volumes (Peltz 97). Writing of this period, Peltz describes a "demonstrative culture of book consumption" that reached its apex with Granger's contributions (109).
At the Huntington Library a copy of Jack Upland (circa 1540), an anticlerical work at the time presumed to have been written by Chaucer, has been adorned with a copy of Joseph Collyer's 1774 portrait of the poet. 31 This lone act of extra-illustration is symptomatic of the eighteenth-century interest in what would come to be called Grangerization and of the period's craze for printed images. But the Huntington Jack Upland also finds analogues in the other mixed-media objects charted in this essay: the printed copies of Stow's Chaucer containing Speed's plate; Joseph Holland's Chaucer manuscript, also adorned with a colored and pasted-in version of Speed's engraving; the Takamiya manuscript, with its later painted portrait; and the portrait leaf in the British Library Caxton, which visually imitates a medieval manuscript illumination even as it serves as a frontispiece to a landmark publication of early print. To a modern sensibility, these hybrid objects might dangle worryingly between the categories of manuscript and print, but the complexity and depth of their entanglement defies the urge to consider them in these oppositional terms. Rather, we might see them as exemplifying the reach and unpredictable ripples of print culture-where conventions like engraved portraits or phenomena like extra-illustration were remixed with hand-painted or -crafted elements, and where author images became viewed as essential to the idea of the book itself, whatever its medium.
The engraved portrait contained in Speght's edition had a long afterlife, in print and beyond it. Speed's plate furnished an archetypal image of Chaucer, one authenticated by an eyewitness account from his own lifetime, and successfully co-opted Hoccleve's narrative to disseminate it in the printed editions of 1598, 1602, and 1687. Moreover, Speght's first edition introduced new readers to a compellingly simple idea that would later spread through the seventeenth-century English book trade: that books needed pictures of their authors. As Alexandra Gillespie has shown, Chaucer's status as an author was an especially valuable commodity to the early English printers and one actively constructed in the wares they made (104-143). Lists of contents, author attributions, prefaces, title pages, even the titles of individual texts themselves-all of these paratextual features also existed in recognizable forms before the invention of the printing press. Print permitted such conventions to take root fully, and each eventually became an indispensable and uniform part of the architecture of the book. At the same time, the plate made for Speght's Chaucer served as raw material that could be molded into new cultural roles: as a collectible portrait for Pepys or a model for premium eighteenth-century replicas. While they look back to John Speed's Progenie plate and to Hoccleve's medieval portraits behind it, these manuscript imitations, customizations, and mixed-media objects also show how their makers reimagined the possibilities of the image in print.

Notes
1. I am grateful to Helen Cooper and Jason Scott-Warren for their comments on a prior version of this essay and to the Bibliographical Society for granting me the Barry Bloomfield Award, which supported the travel necessary to conduct this research.
2. According to Hind, Speed may be the designer, not the engraver, who remains anonymous (286-89). On the different states of the engraving, see Driver 246, n. 6. Because this essay is concerned with the afterlife of the image first conceived and attributed to Speed, I refer to this visual tradition as "Speed's" throughout, while recognizing that different and anonymous artisans were responsible for its later material instantiations.
3. The pendant has also been proposed to be a penknife or a vial of holy blood .
4. As Pearsall notes, these three manuscripts represent a small fraction of the more than forty manuscripts of the Regement that currently exist. The other two are in British Library, MS Royal 17.D.VI (a less skilled portrait), and in Philadelphia's Rosenbach Museum and Library, MS 1083/10 (possibly a later copy from the eighteenth century) (Appendix 1 289-91).
5. Pearsall suggests this possibility (Appendix 1 302-303). 27. This ambiguity-about whether the dates refer simply to Chaucer's time or to the age of the artifacts-has fueled the speculation that Additional MS 5141 is a medieval leaf removed from the Cotton manuscript; it also contributed to the inflated value of the Stanford fragment in the 1930s, when that university successfully bid $450 for it in the midst of the Great Depression; see Jordan. 28. In the eighteenth century, Chaucer's portrait was engraved by George Vertue four times (one of which was included in Urry's 1721 edition of Chaucer's works) and by Jacobus Houbraken (1741); see Pearsall, "Portraits" 303; Lam and Smith. 29. The early seventeenth century is generally accepted as the point at which author portraits began to more regularly appear in books printed in English; see Alexander 298.
30. The version in Pepys's album is identified in Chamberlain's catalogue as a copy of Speed. In Pepys's copy of Speght's edition ( [1602]

The Canterbury Tales
Beinecke Rare Book and Manuscript Library, New Haven, Takamiya MS 32