of the early modern printing house dispensed with “the erroneous assumption that a book was normally put into production as an independent unit.”6 Works were printed in parts, frequently across multiple presses, and put together in nonuniform ways that vex attempts to describe a standardized practice. On the consumption end of early book culture, however, the primacy of the independent unit of reading and interpretation is more often than not upheld. Modern collectors standardized early printed texts of great value as a matter of conservation, and in each multibook volume like Caldecott’s Shakespeare, broken up and rebound in individual units in modernity, taxonomies of text, work, and author from the modern period were made to organize the reception of premodern literature. In the earliest domain of printed literary materials in English, the works of the later Middle Ages, scholars have made much of the “clusters of literary writing” discovered in Sammelbände, the “fluid canonicity” reminiscent of the manuscript miscellany that seems to have been generated in everyday acts of anthologization by readers and printers.7 Yet by the time we reach the age of Shakespeare in archives, we find books and collections that are seemingly indistinguishable from modern ones—that is, neat rows of independent, leather-bound plays and books of poetry. In nearly every case, as we will see, the bindings, labels, and catalog identifiers are products of the last two centuries.
The works of the English Renaissance, in fact, offer a particularly promising field of primary materials within which to pose the question of how early books were made (or unmade) in this way. As common sense suggests, the likelihood that a text has undergone modernizing structural renovations such as those sketched out in the last chapter is directly proportionate to how valuable it was in the eyes of modern owners and collectors. Texts now considered literary, therefore, often reach us as the most heavily processed of all early printed materials, a fact obvious to researchers of the period’s lowerprestige, nonliterary books, which are far more frequently found in original bindings and seemingly unkempt Sammelbände. Within that literary subset, the dramatic works by Shakespeare—which Thomas Bodley famously ranked among the “riffe-raffes” and “baggage books” to be excluded from his library8—were eventually of the utmost value and thus subject to all forms of bibliographical intervention that may have come into fashion. As a result, the surviving archive of Shakespearean texts has a particularly varied morphology, though it is one that has been oversimplified, or suppressed, by modern collecting practices.
A representative example is the sole copy of the “sixth quarto” of Pericles (1635) now held at the British Library.9 Like many extant Shakespearean plays, the text is trimly bound in luxury leather as it might have adorned a gentleman’s shelf in the nineteenth century, though this was not the case.10 The book was one of many bequeathed to the library by David Garrick, the actor and playwright, at his death in 1779. It owes its neat, modern appearance to the British Museum bindery, where the book was given new covers (tooled in gold with Garrick’s coat of arms) in the century after its donation. Though the library’s integrated catalog lists the text as an individual item—with no notes in the entry suggesting anything to the contrary—earlier catalogs from Garrick’s collection give us a different picture.11 Before it was rebound in the nineteenth century, Pericles was one part of a larger compilation, and this volume, left mostly intact, retains an eighteenth-century table of contents originally written in the flyleaves that confirms the arrangement of texts in Garrick’s time (Fig. 9).12 To say the least, these texts are strange bedfellows: a morality play, Conflict of Conscience (1581); an interlude called New Custome (1573); the sometime Shakespearean history play Edward the Third (1599); John Marston’s tragedy Antonio’s Revenge (1602); Pericles; the early tragedy Gorboduc (1590); and the comedy Albumazar (1634). The grouping is not arbitrary, though it may seem so to us. Garrick was an avid collector who assembled a wide-ranging library of dramatic texts, most of them in composite volumes, for his own use and others’.13 This volume, one of the few to survive the nineteenth-century rebinding campaigns at the British Museum,14 bears the traces of its shifting shapes and uses. The contents list indicates that Garrick at one point moved Albumazar to another volume (likely as he adapted it for the stage).15 Moreover, a second contents list (Fig. 10), written on the leaf preceding Gorboduc,16 indicates that this composite book has origins in an even earlier composite volume whose texts seem to have been reshaped and redistributed throughout Garrick’s collection as they were acquired. The earlier hand is that of the seventeenth-century collector and former owner Richard Smith,17 and the superseded arrangement of texts is even more peculiar: sixteenth-century interludes mixed with Stuart masques and Restoration comedies; works from authors as diverse as John Bale, Ben Jonson, George Chapman, and Matthew Medbourne.18 These composite books stand in stark contrast to the slim, modern-looking Pericles, whose status relative to the other texts is now encoded in its fine binding. That it once formed part of an eighteenth-century assemblage of texts, which itself once formed part of a seventeenth-century assemblage of texts, is made all but imperceptible by an imposed nineteenth-century notion of its fixity, autonomy, and canonicity.
Figure 9. An eighteenth-century contents list from British Library C.21.b.40, listing seven plays in a book. © British Library Board.
Figure 10. An earlier arrangement of texts written on the verso of a back page of a different item in British Library C.21.b.40. © British Library Board.
For generations of collectors and owners whose legacy is still visible in archives, the relatively flexible composite volume was the most conventional, practical means of storing and using most kinds of literary texts. Sammelbände, binding experts and rare-book curators tell us, were staples of early book culture.19 But as artifacts of literary history—artifacts conveying a range of possibilities for intertextual reading and canon formation that are perhaps not obvious to us today—these composite volumes have not been closely examined by critics. One reason for this neglect is a tendency to see intellectual activities independently of knowledge organization, considered merely practical—a tendency that is especially evident in a figure like Garrick, whose revivals and adaptations have long proven resonant in modern interpretations of Shakespeare’s plays, but whose methods of reading and organizing the texts that presumably facilitated those revivals and adaptations have hardly been explored at all. The most fundamental reason for this neglect, however, is clear in the fate of Garrick’s copy of Pericles: despite the ubiquity of composite volumes in the handpress era, Shakespeare’s books are rarely found today in these configurations. In the modern era, the most prestigious literary works—the works that attract the most critical attention—were systematically extracted, decontextualized, and clothed anew in material configurations that reflect little history of ownership or use. Where a Shakespearean text can be found in an undisturbed composite volume, it is most often one of the apocryphal or otherwise noncanonical texts. At St. John’s College, Cambridge, for example, there is a mid-seventeenth-century volume combining eight books of controversial religious and political prose with a copy of The Birth of Merlin (1662), a play attributed to Shakespeare and Rowley (Fig. 11).20 St. John’s College, Oxford, preserves a similar example: a collection from the eighteenth century (with an original handwritten table of contents) bringing together a diverse array of plays, masques, and pageants, including the 1662 Birth of Merlin text and the second quarto of The Merry Devil of Edmonton (1612), a play also attributed on its title page to Shakespeare.21 In fact, a cursory survey of the extant copies of these two noncanonical plays at the British Library, Oxford, and Cambridge shows that over half occur in composite configurations. Plays with less dubious canonicity almost never occur in composites.22 The implication is something of a bibliographic corollary to the point made some time ago by Stephen Orgel: the “authentic
