Configuring the Book

Andrew Murphy
University of St Andrews


  1. As the contributions gathered in this special issue of EMLS indicate, there are many different ways in which we can conceive of Shakespearean configurations. What I would like to focus on in this brief piece is a quite literal form of configuration: namely, the physical makeup of the printed Shakespeare text, how it changes, and the implications of those changes. This is a matter, to some extent, of size, but it is not just about size. The history of changing configurations of the physical book is interwoven with the history of readers’ access to the text. Size matters, in this sense, but often in rather unexpected ways.
  2. Shakespeare begins, of course, in quarto. His first datable surviving printed text — Richard Field’s 1593 Venus and Adonis — appeared in that format, and the following year’s Titus Andronicus was also printed in quarto. But the print histories of the poems and the plays soon diverged. After a second quarto edition, in 1594, Venus and Adonis was downsized to octavo and, with two exceptions — even smaller sextodecimos published in 1636 and 1675 — the poem continued to be reproduced in octavo through to the end of the seventeenth century.[2] Strikingly, the number of text pages in the octavo editions is uniform with that of the quarto editions. Indeed, the octavos reproduce the page divisions of the quarto editions (an indication that the compositors of the first octavo edition must have used one of the quartos as copy).[3] One implication of this reproduction process is that while the octavos would have required the same amount of compositorial time, they would have taken half the time of the quartos to print (since each octavo sheet included twice as many pages as the quarto sheet) and, crucially, the octavos would have required half as many sheets as the quarto for the same size of press run. The octavos were, then, significantly cheaper to produce. It is hard to know whether the savings involved were passed on to purchasers, but it is possible that they were and we can certainly say that the small-sized editions were popular: Venus and Adonis ran to a total of fourteen smaller-format editions between 1595 and 1675.[4]

  3. The Rape of Lucrece follows a roughly similar track. First published in quarto in 1594, it reappeared in octavo four years later. By 1655, a total of eight octavos had been published. In this instance, the poem, which had taken up 96 text pages in quarto, was reduced to 72 pages (in an edition of 1598, and two editions in 1600) and then to 64 pages.[5] The savings on paper and press work were, in this instance, not quite so great as those achieved in the case of Venus and Adonis, but still, there would have been some significant savings and potentially these could have been used to reduce the price, should the publishers have chosen to pass the savings on to their customers.[6]

  4. When we turn to the plays, we find just one instance of a text moving downward in format from quarto to octavo. This is the variant text of 3 Henry VI, titled The True Tragedie of Richard Duke of Yorke . . . ,  printed by Peter Short for Thomas Millington in 1595. Short had printed one Shakespeare-related text previously: the 1594 quarto of The Taming of A Shrew. He was involved in producing two further Shakespeare plays after the above-mentioned octavo. In 1597, he would appear to have taken over the printing of the variant text of Richard III, after Valentine Simmes had completed sheets A to G; in this case, he would, in all likelihood, have been following a formatting decision set by another printer.[7] In the following year, Short printed  the last Shakespeare play he would print, a quarto of 1 Henry IV for Andrew Wise. Taking the run of texts as a whole—Q A Shrew; O True Tragedie; final sheets Q RIII; Q 1HIV—it is probably a safe bet simply to conclude that the octavo edition of The True Tragedie is an anomaly. It is interesting to note, however, that Short’s work printing Shakespeare editions did not end with the Wise 1 Henry IV quarto. In 1598, he produced a Lucrece for John Harrington and, in the following year, a Venus and Adonis for William Leake.[8] Both these texts were octavos, and thus part of the run of texts that signalled the poems’ popularity. So a possible, alternative way of looking at Short might thus be to say that he sits at the point of intersection between different formatting choices, with his 3 Henry VI representing an experiment in shifting the playtext down to a format that was cheaper to produce and which could, potentially at least, have been offered to bookbuyers at a lower cost than the standard quarto.[9] The experiment was not pursued beyond this one edition and it is difficult to say why.[10] One wonders what might have happened if it had been tried with a play that might likely have been more appealing to the reading public than 3 Henry VI.[11]

  5. In fact, however, what we get with the plays is not a shift downward in format, but rather a shift upward: to the monumental—and expensive—folios of 1623, 1632, 1663/4 and 1685. The decisiveness of this shift is striking. If we take the period between the appearance of Titus in 1594, and of the First Folio in 1623, a span of 29 years, we find that a total of 58 editions of individual Shakespeare plays appeared in print: an average of exactly 2 per year.[12] By contrast, in the entire seventy-seven year period between the publication of F1 and the end of the seventeenth century only 33 individual play editions were published. After 1623, then, the annual rate of quarto play publication fell to just 21% of the pre-F1 rate. It is important not to take these raw numbers wholly at face value. By 1623, Shakespeare had been dead for seven years and there were no new titles to be brought to publication, so the publication of old titles might well have tailed off anyway, decade by decade, even without the appearance of the folios. Also, of course, the eighteen-year period during which the theatres were closed is likely to have had an impact on playtext publishing, as there would have been no contemporary performances to serve as a driver prompting new editions on the backs of popular productions.[13]

  6. The extent to which the quarto segment of the publishing market contracted in the wake of F1 is still striking. Lukas Erne has argued that Shakespeare imagined himself as a “dual mode” writer – producing “literary” texts with an eye to publication, while accepting that they would have to be configured into alternative “scaled down” versions for performance (passim). What we can say in relation to the publication history of the plays is that whatever Shakespeare’s intentions as a writer were (and these remain, of course, ultimately unknowable), the readerly audience that he actually achieved narrowed considerably from 1623, when the price of access to the text of the plays increased very considerably. The cost of F1 was between 30 and 40 times that of a typical quarto, pricing it well outside the casual purchasing power of all but the better-off segment of readers (taking the typical price of a quarto to be sixpence and the typical price range of F1 to be from 15s to £1).[14] By contrast with the history of the poems, where a downsizing of format may have facilitated and garnered an expanded readership, the effect of configuring Shakespeare into a collected folio volume was the contraction of the playwright’s readership to a small, relatively elite core. There is a further qualification that might be registered here. There are far more plays than there are poems and, had all of the plays been available in quarto, assembling a full collection would, of course, have cost about as much as purchasing a copy of F1. My point, however, is that F1 effectively creates an all-or-nothing access point: if you do not have the minimum entrance payment of 15s, you do not get to cross the access threshold.

  7. It is not uncommon for scholars to speculate as to how our understanding of Shakespeare would be different if F1 had never been published. The default conclusion of such imaginings is to note that, without F1, we might well have a much smaller canon of plays, eighteen titles having appeared for the first time in the Folio. But one could wonder also what might have happened if the Folio had not locked the playwright’s work into an elite—and expensive—format. (One thinks here of William Prynne’s complaint that “Shackspeers Plaies are printed in the best Crowne paper, far better than most Bibles” [**6v]). What if the octavo 3 Henry VI had proved to be a successful experiment, prompting a general downsizing of the playtexts, so that they could have reached, and could have continued to reach, a general audience over the course of the seventeenth century, and beyond? And how might the history of the greater cultural configurations of Shakespeare have been different, if smaller had triumphed rather than bigger? Might we, for instance, have inherited a more dispersed and less monolithic Shakespeare? We certainly would have inherited a far less uniform Shakespeare, in that different versions of individual plays might have continued to be circulated, rather than being – as they are in F1 – reduced to singularity (one King Lear, one Hamlet, one Henry V . . . ). Might the relationship between the poems and the plays be different if what might be styled the “perspective lines” of textual formatting had created more of a sense of equivalence between them? As it was, the poems were excluded from F1 and, once the run of octavo editions of the poems came to an end, they entered into a state of supplementarity in relation to the collected plays, at least until the end of the eighteenth century (if not beyond).

  8. We can only speculate with regard to most of these matters. What we can confidently say, however, is that the model established by F1—of the expensively produced edition (albeit an edition that included the collected plays) sold at high cost to a relatively narrow readership—was remarkably persistent. There were challenges along the way, of course. In the eighteenth century, for example, Scottish and Irish publishers (and some English publishers, such as Robert Walker) sought to undercut the dominant Shakespeare publishing cartel by producing cheap texts, generally in octavo or duodecimo formats. And, in the wake of the 1774 Donaldson v. Beckett House of Lords ruling, which effectively confirmed Shakespeare as a “public domain” author, a number of publishers began to offer cheaper, smaller-scale editions, both of individual plays and collected works. But the expensive, exclusive text persisted. As Simon Eliot has noted of the English book trade more generally, it was not until the mid-nineteenth century that the default book trade model of seeking high profit margins on low volume sales was finally broken, with low-priced, rather than high-priced books finally coming to account for the largest percentage share of publishers’ outputs.[15] This shift in publishing priorities coincided with a significant expansion of access to education in the UK, with the result that Shakespeare gained a readership much greater than he had ever achieved before.

  9. I have tracked the emergence of this readership in Shakespeare for the People, and have noted there the central importance of John Dicks’ edition to the wide dispersal of Shakespeare’s text among a very broad audience. The key to Dicks’ success was the manner in which he exploited a variety of formats in selling his Shakespearean wares.[16] Initially, the plays were issued in paper-covered numbers, including two plays, at a penny per number. Subsequently, the plays were gathered into a single cloth bound volume, priced at 2s., quickly followed by a 1s paper-covered collected volume. The variety of formats ensured a broad readership, with Dicks ultimately achieving sales in excess of one million (Dicks 451). Finally, we might say that in the Dicks edition, the potential that had been suggested by the early modern octavo editions was fully realised: by configuring Shakespeare in multiple, inexpensive, formats the publisher brought the writer to a wide audience. But Shakespeare had, of course, already by then become . . . “Shakespeare.” And there is something ineradicable about that process of monumentalisation – a process that had its roots, at least partly, I am suggesting, in the triumph of folio over quarto. There are configurations and there are reconfigurations; but some configurations cannot ever quite be undone.



[1] Thanks to RandyM cLeod for comments on a draft of this paper.

[2] In Shakespeare in Print: A History and Chronology of Shakespeare Publishing, I mistakenly identify the 1675 edition as an octavo (308).

[3] Of the smaller format editions that survive as complete texts, only the 1627 octavo diverges from the text distribution of the quartos. This edition was printed by John Wreittoun in Edinburgh and the text is compressed into a smaller number of pages than in the other editions, thus reducing the production costs still further.

[4] The total includes the two sextodecimos already noted. The 1595 date for the first octavo is speculative: the surviving copy (Folger PR2749 .Y4 n.d.) lacks all before B1. I should note, of course, that “popular” is a relative term here. I use it in the sense that Peter Blayney and Alan Farmer and Zachary Lesser have used it in their seminal exchanges on the “popularity of playbooks,” by basically taking reprint rates as an index of popularity. It must be remembered, however, that literacy rates were low in the period and popularity is measured in sales of thousands, by contrast with the sales of hundreds of thousands that would characterize the successful editions of the nineteenth century. For Blayney, see “The Publication of Playbooks” and “The Alleged Popularity of Playbooks;” for Farmer and Lesser see “The Popularity of Playbooks Revisited” and “Structures of Popularity in the Early Modern Book Trade.”

[5] The 1655 edition was longer, as it included Quarles’ Banishment of Tarquin.

[6] The sonnets, interestingly, fail to match the pattern established by their predecessor poetic texts. Thorpe’s edition of 1609 was a quarto and Benson’s (very different) edition of 1640 followed the trend of downsizing to octavo, but no further edition of any kind—at least that we know of—appeared during the seventeenth century. It may possibly be that sonnet sequences were more subject to the vagaries of literary fashion than the narrative poems. (I am grateful to Faith Acker for raising this latter point; I am also indebted to her in my comments on the poems and “supplementarity,” below).

[7] See Greg 84, n. 4.

[8] There were two editions of Venus and Adonis in 1599. The other edition was printed by Richard Braddock, probably using the Short edition as copy.

[9] I should note, however, that the analogy with the poems is not exact here. Both Venus and Adonis and Lucrece appeared first in quarto then in octavo. Short’s edition of The True Tragedie was the first to be issued, so, strictly speaking, it did not represent a change in format. It is, however, a downsizing from the standard format for plays. And the arguments regarding cost do still, I think, more or less apply here, though not, of course, exactly. Short’s The True Tragedie takes up 80 pages; the quarto edition William White printed for Thomas Millington in 1600 runs to 64 pages. Short would have used fewer sheets to produce his edition, though not half the number of sheets subsequently used by White. Thus, there would have been savings, but they would not have been as great as those made on the poems.

[10] The way I have framed my argument implicitly suggests that the format choice in relation to The True Tragedie was made by Short and, more generally, that such decisions were made by printers. This may well not have been the case: it is quite possible (indeed, perhaps, even quite likely) that the publisher, rather than the printer, made this decision. In the case of The True Tragedie, the publisher was Thomas Millington. My greater point, however, is that someone is making format decisions that have an impact on textual access.

[11] I am, to some extent, imposing a contemporary judgment here: the fact that 3 Henry VI is little performed and little read in our own time does not necessarily mean that it need have been equally unpopular for an early-modern readership.

[12] In counting the plays I have included the variant texts, together with Pericles and Edward III. I have excluded editions of the apocrypha, such as Locrine and Sir John Oldcastle and A Shrew. I have also excluded the undated 1 Henry IV fragment and the possible 1597 edition of Love’s Labour’s Lost. The second quarto of Hamlet, with title pages variously dated “1604” and “1605,” has only been counted once.

[13] Only three editions appeared in the period: Merchant of Venice (1652 – a reissue of Q3, 1637), King Lear, and Othello (both 1655).

[14] For the pricing of F1, see Blayney, The First Folio of Shakespeare.

[15] See Eliot.

[16] For a comprehensive and thoroughly excellent account of Dicks’ editions, see Nelson. Nelson notes that, by 1864, Dicks was running a total of twelve steam presses and employing over a hundred people.


Works Cited





Responses to this piece intended for the Readers' Forum may be sent to the Editors at

© 2013-, Annaliese Connolly and Matthew Steggle (Editors, EMLS).