As an unperfect actor on the stage,

Who with his fear is put beside his part,

Or some fierce thing replete with too much rage,

Whose strength's abundance weakens his own heart;

So I, for fear of trust, forget to say

The perfect ceremony of love's rite,

And in mine own love's strength seem to decay,

O'ercharged with burthen of mine own love's might:

O let my looks be then the eloquence

And dumb presagers of my speaking breast,

Who plead for love, and look for recompense,

More than that tongue that more hath more expressed.

 O learn to read what silent love hath writ:

 To hear with eyes belongs to love's fine wit.

    Sonnet 23 as quoted above appears not only in Evans's 1996 edition, but also in Helen Vendler's The Art of Shakespeare's Sonnets (1997), as well as in many other editions.(FN2) Their text departs in an important way from the quarto of 1609 in which the poem was first published. In line 9, where Evans, Vendler, and others have "looks," the 1609 quarto has "books." The emendation from "books" to "looks" has a long history, going back all the way to George Sewell who edited the unauthorized seventh volume that was added to Pope's Shakespeare edition of 1725. Sewell has found countless followers in the course of the last three centuries. By the early twentieth century, "looks," as one editor had it, was "an almost certain emendation," and another editor, twenty years later, thought that "looks" was "entirely necessary."(FN3) Reassuringly, others have disagreed, pointing out that "Books alone agrees with line 13," "O! learn to read what silent love hath writ."(FN4) As Stephen Booth writes, "books is the Qreading and makes sense," so there is no need to emend it.(FN5) Some have argued that books can hardly be "presagers of [someone's] speaking breast," but Colin Burrow rightly points out that "The word is a new one in the 1590s, and Shakespeare seems to be using it as a near synonym for 'ambassador,' lather than exploiting its associations with understanding of the future."(FN6)
    In fact, the sonnet carefully constructs an opposition between the oral and the literate: the "actor on the stage" (1), the "ceremony of love's rite" (6), "eloquence" (9), "dumb" (10), "speaking" (10), and "tongue" (12) all contribute to the notion of orality to which the sonnet opposes that of literacy: "books" (9), "read" (13), "writ" (13), and "hear with eyes" (14). Yet even when editors realize how important the word "book" is in establishing one of the sonnet's central oppositions and thus do not emend "books" to "looks," their annotation at times seems tendentious. "Books" refers exclusively to Shakespeare's Venus and Adonis and The Rape of Lucrece, the argument goes. Since both narrative poems are dedicated to the Earl of Southampton, the so-called Southamptonites -- who argue that the famous "Mr. W. H." in the prefatory material to the sonnets corresponds to the Earl of Southampton -- have been particularly keen to argue this case. Others have maintained that "books," which could refer to any kind of text on paper, even to a single handwritten sheet, in fact refers to the sonnets themselves.(FN7)
    Considering the theatrical context of the sonnet's opening lines, however, it seems significant that editors have generally failed to investigate a rather more plausible reading, namely, that "books" refers to printed Shakespearean playbooks. Stephen Booth is the only exception of which I am aware, pointing out that "Shakespeare may intend a play on 'book' meaning the written text of a stage play."(FN8) Indeed, just as the speaker, incapable of adequately expressing his love in speech, asks the addressee to witness his written profession of love, so the speaker, an imperfect actor, says that his true eloquence is apparent in his playbooks. To discount this possibility fails to register a simple but powerful analogy established by the sonnet. If we take seriously recent arguments about the dating of the sonnets,(FN9) then we realize that many of the sonnets addressed to the young man may have been written or revised around the middle of the first decade of the seventeenth century, a time in which a good number of Shakespeare's playbooks had been published. It seems clear that the editorial history of Shakespeare's Sonnet 23 has much to do with the traditional resistance to the view that William Shakespeare wanted his dramatic productions to be read and was aware of the eloquence of his printed playbooks.
    I have sketched elsewhere what seems to have sustained this resistance for so long, namely, the interaction of four misassumptions that, by mutually reinforcing each other, contributed to keeping each other in place:
 

Firstly, printed playbooks, mostly in quarto format, roughly the equivalent of the modern paperback, allegedly represented mere ephemera.... Secondly, Shakespeare had no interest in the publication of his plays.... Thirdly, the Shakespeare playbooks that have come down to us supposedly give us the texts as they would have been performed, even in the case of very long plays such as Richard III or Antony and Cleopatra. And fourthly, when we have plays that survive in both long and short versions -- like Romeo and Juliet, Henry V, and Hamlet -- the long texts are thought to represent the "normal" stage version whereas the short ones, which scholars sometimes refer to as "bad quartos," represent anomalies of some kind.(FN10)

    It is easy to see how these misassumptions supported each other: the ephemeral nature of playbooks allegedly explained why Shakespeare was indifferent to their publication. The supposition that Shakespeare only wrote for the stage meant that even Shakespeare's longest plays were thought to have been designed for performance in their entirety. And the idea that these long play texts were performed meant that shorter versions of the same plays had to reflect something different and inferior. This mechanism of mutual reinforcement may well be an important reason why belief in Shakespeare's indifference to his literary reputation, his authorial standing, and his works' survival long remained in place.
    The present forum in the pages of Shakespeare Studies is one among a number of signs testifying to the fact that our view of Shakespearean authorship has recently received an overdue revaluation. Several publications since the turn of the century have made plausible a Shakespeare who cared about his standing as published dramatic author. In a monograph of 2001, James Bednarz sees the Shakespeare who was involved in the poets' war around the turn of the century as self-consciously literary, aware of his authorial standing, competitive.(FN11) Even slightly earlier, Richard Dutton was arguing that "in writing plays which were in some respects unplayahle... [Shakespeare] was effectively writing for a readership no different in essence from that of his sonnets and epyllia," and that "Shakespeare had leaders in mind too, however much practical theatrical applications must also have shaped his thoughts."(FN12) Building on the work of Dutton as well as on that of Peter Blayney, Andrew Gurr, and Stephen Orgel,(FN13) my Shakespeare as Literary Dramatist (2003) similarly argues for an alternative to the exclusively theatrical Shakespeare in whom many have long believed, opposing the beliefs that he was indifferent to the publication and afterlife of his plays and that the only form of publication he ever sought for his plays was the stage. I suggest instead that Shakespeare was acutely aware of and cared about his rise to prominence as a print-published dramatic author, that he and his fellow actors of the Lord Chamberlain's Men had a policy of having his plays published, and that he anticipated and catered to a readership for his plays. Shakespeare's long play texts, I argue, thus give us access to literary versions of his plays that would have been significantly abridged -- to something like the length of short quartos such as Ql Romeo and Juliet or Hamlet -- before reaching the stage.
    Scholarship published since 2003 may suggest that the time was right for the view presented in Shakespeare as Literary Dramatist. In his British Academy Shakespeare Lecture of 2004, Henry Woudhuysen -- one of the general editors of the Arden Shakespeare -- holds that printed playbooks were not the "ephemeral items" they have often been taken to be.(FN14) He argues instead that "Shakespeare must have been aware that his plays had reached print, and this may have influenced the ways in which he wrote," adding that "It is possible to argue, on textual as well as aesthetic or historical grounds, that distinct authorial versions of [Shakespeare's] plays were produced for reading rather than performance."(FN15) Woudhuysen's argument that Shakespeare "cannot have been entirely indifferent to the phenomenon of seeing plays, including his own, printed" is corroborated by an article by MacD. P. Jackson published in 2005, "Francis Meres and the Cultural Contexts of Shakespeare's Rival Poet Sonnets," which establishes a wealth of telling connections between Meres's "Comparative Discourse" in Palladis Tamia (1598) and the Rival Poet sonnets (78-86), for which recent research strongly suggests a composition date of ca. 1598-1600.(FN16) Meres compares English and ancient writers, singling out for repeated praise a number of contemporary or recent English authors, including Shakespeare (whose name, it may be well to recall, had not appeared on a single title page prior to 1598). Jackson convincingly demonstrates that "Shakespeare read Meres's 'Comparative Discourse' attentively" and argues that "[t]he Rival Poet sonnets originated... in a general sense of rivalry fuelled by Francis Meres's glib inventory of England's top poets and playwrights."(FN17) Jackson's article considerably strengthens our sense of Shakespeare's self-awareness as a print-published poet and dramatist who profoundly cared about seeing his name and his works in print.
    Two recent monographs by Patrick Cheney, Shakespeare, National Poet-Playwright (2004) and Shakespeare's Literary Authorship (2008), lend further support to this view by demonstrating how the works themselves are preoccupied with an inscription of Shakespearean authorship.(FN18) Of further significance is The Cambridge Companion to Shakespeare's Poetry, for which Cheney served as editor, in which several contributions pay attention not only to the freestanding poems but also to the poetry in the plays.(FN19) Cheney's work convincingly establishes that Shakespeare was not simply a playwright who occasionally happened to write poems, but that he was a poet and a dramatist throughout his career, writing poetry that could be lyric, narrative, or dramatic, and drama that could function as poems on the page or be adapted and abridged to function as plays onstage.
    Shakespeare's standing as a literary author has also been rendered more plausible by Alan H. Nelson who, in an article of 2005, surveys individuals who, during Shakespeare's lifetime, owned libraries that contained at least one book by Shakespeare. He concludes, "against the grain of much modern criticism, that Shakespeare's poems and plays ought to be approached, if we are to respect history... as verbal and dramatic art, as -- dare I think it? -- English Literature."(FN20) Whether in terms of literary production or of its contemporary reception, the case for a significant Shakespearean literary authorship has been considerably strengthened in the last few years.
    Publications have done most to establish Shakespeare's new authorial standing, but work presented at several conferences has duly followed in their wake. In July 2004, Richard Wilson orga-nized a conference at the University of Lancaster with the title "The New Shakespeare: A Writer and His Readers; The Return of the Author in Shakespeare Studies."(FN21) In March of the following year, a paper session at the conference of the Shakespeare Association of America (SAA), 2005, in Bermuda, was devoted to "Shakespeare's Literary Aspirations," while a seminar at the conference of the British Shakespeare Association (BSA) in Newcastle, in September of the same year, investigated to what extent Shakespearean drama constitutes "an almost oral art" or, conversely, "is already a fully literary art."(FN22) In 2006, at the World Shakespeare Congress in Queensland, Australia, a panel session addressed the topic of "Shakespeare for Reading," which invited a reconsideration of "the received wisdom of the last fifty years that Shakespeare's texts are, without question, play-texts intended primarily for the stage."(FN23) The scholarly dissemination of Shakespeare's new authorial standing at conferences and in publications has become such that Catherine Belsey has diagnosed "a quiet revolution in Shakespeare studies": "More than two decades after New Historicism turned our attention away from close reading and toward locating Shakespeare more firmly in his own culture, scholarship is shifting our focus onto Shakespeare's own place in that culture itself, and the case is founded firmly on the texts."(FN24)
    If the beginning of this century is witnessing "a quiet revolution" with the advent of a "New Shakespeare," then it may be asked how this advent ought to affect, or is already affecting, Shakespeare as he is commonly mediated to us in modern editions. As the unwarranted emendation of "books" to "looks" in Sonnet 23 illustrates, we do well not to underestimate the power of editors in shaping our conception of Shakespeare. John Jowett has recently commented on "a new emphasis at the beginning of the twenty-first century, one that pares hack the theatrical dimension and asserts on new grounds the presence of Shakespeare the author in the field of textual study." Jowett goes on to evoke the possibility of an immanent "restatement of an authorial orientation" in editorial policy.(FN25) Such a restatement would need to counter in the first place the policy of the influential Oxford Complete Works (of which Jowett himself was an editor), the aim of which was "recovering and presenting texts of Shakespeare's plays as they were acted in the London playhouses." Being given the choice between "a text which is as close as possible to what Shakespeare originally wrote" and "a text presenting the play as it appeared when performed," the Oxford editors emphatically opted for the latter.(FN26) A number of reputed scholars have recently argued that this policy is in fact profoundly flawed, not only because it relies on an increasingly dated view that threatens to reduce Shakespeare to "a man of the theatre," but also because the recovery of "Shakespeare's plays as they were acted in the London playhouses" is simply impossible or, to use Paul Werstine's word, "quixotic."(FN27) As David Scott Kastan has rightly asked, "how are we to know how '[the plays] were acted'? How can this information be recovered from the witness of the early printed play texts?"(FN28) The answer is that we can't. All we have access to is "the early printed play texts," not early modern performances. Of the two options outlined above, the only one that can thus be undertaken is editing the text that takes us as close as possible to Shakespeare's authorial composition, not his and his fellow actors' theatrical adaptation. In other words, an editorial policy in line with the recent revaluation of Shakespearean authorship attempts to recover the full, unabridged authorial play texts, the play texts as Shakespeare wanted them to be read, and as he repeatedly saw them published and republished during his own lifetime.
    The second edition of the Oxford Complete Works (2005) shows no change in editorial policy, which has prompted Brian Vickers to lament "The Oxford editors' stubborn adherence to their theatrical paradigm."(FN29) Elsewhere, however, it is apparent that the paradigm has started to shift. For instance, whereas the Oxford editors based their Hamlet on the supposedly theatrical Folio text, Ann Thompson and Neil Taylor, in the Arden 3 (2006) edition of the same play, rightly point out that only the much-abridged First Quarto is of a length that could have been performed, therefore choosing to base the main volume not on the Folio but on the Second Quarto, the version reflecting the play as Shakespeare originally conceived it.(FN30)
    Revising the Oxford editors' theatrical paradigm is not only a matter of copy text, however, but also requires a reconsideration of the editorial mediation of dramatic action. Early modern play texts usually contain few stage directions, which has led modern editors to add their own. As Wells put it, "the editor needs to identify points at which additional directions, or changes to those of the early texts, are necessary to make the staging intelligible."(FN31) As a result, modern Shakespeare editions suggest that the printed play texts are surrogate performances, performances of the mind in which (added) stage directions encourage readers to imagine a theatrical performance. Yet as far as we know, printed play texts were conceived, of rather differently not only by Shakespeare (who could have, hut did not, provide additional stage directions) hut also by his readers: what we know about their reading habits suggests that they were particularly interested in poetic, "purple" passages or in sententiae, which they highlighted or excerpted, but showed little interest in inferring stage action from the play text.(FN32) Modern editions with stage directions that are added to the main text (rather than suggested in the commentary) thus arguably fabricate a more theatrical and less literary play text than Shakespeare ever intended.(FN33)
    Even if we grant the usefulness of added stage directions for certain kinds of editions, the exact makeup of these stage directions may still deserve reconsideration. Wells argues that an editor, when adding stage directions, "has to think in terms of the Elizabethan stage," and takes it "as axiomatic that the plays take place, not on heaths, in forests, in castles, in palaces, in ante-rooms, or bedrooms, or throne-rooms, but on a stage."(FN34) Much editorial practice since the Oxford Complete Works conforms to these ideas. Yet if we want to do editorial justice to a Shakespeare who not only wrote play scripts for the stage but also dramatic texts for the page, there is no reason why added directions should be conceived of exclusively in terms of the theatrical representation (i.e., the stage), and not the represented dramatic fiction (i.e., the castles, etc.). Pace Wells, what takes place onstage is not simply plays but theatrical performances, whereas a play text that is read (as is the case with modern editions) takes place in the reader's imagination, which can easily picture the dramatically represented instead of the theatrical representation. An editorial practice that encourages readerly engagement with the fictionally represented seems all the more appropriate as Shakespeare's early modern play texts contain not only theatrical but also fictional stage directions: in Coriolanus, characters "enter the City" (TLN 568); in Timon of Athens, the protagonist enters "out of his Caue" (TLN 2360); in Julius Caesar, Brutus enters "in his Orchard" (TLN 615); and a stage direction in 2 Henry VI records a "Fight at Sea" (TLN 2168).(FN35) Examples could be multiplied at will.(FN36) What this means is that an editorial practice that adds not only theatrical but also literary stage directions does better justice to a Shakespeare who was not only a playwright but also a literary dramatist.
    If Shakespeare, to return to Sonnet 23, not only cared about the "actor on the stage" but also about his own dramatic "books" and their "eloquence," then it seems important that modern editions reflect the Shakespeare that a growing body of recent scholarship has proposed to us. By avoiding a policy that unduly privileges the theatrical over the literary, editors thus play a vital role in mediating to us the plays in a form that does justice to Shakespeare's early modern authorial standing.
ADDED MATERIAL
 

FOOTNOTES
1. G. Blakemore Evans, ed., The Sonnets, the New Cambridge Shakespeare (Cambridge: Cambridge University Press, 1996).
2. Helen Vendler, The Art of Shakespeare's Sonnets (Cambridge, MA.: Harvard University Press, 1997). For other editions reading "looks," see Hyder Rollins, ed., The New Variorum Edition of Shakespeare: The Sonnets, 2 vols. (Philadelphia: Lippincott, 1944), 1:66-67.
3. H. C. Beeching, ed., The Sonnets of Shakespeare (Boston: Athenaeum Press, 1904); T. G. Tucker, ed., The Sonnets of Shakespeare (Cambridge: Cambridge University Press, 1924).
4. Sidney Lee, ed., The Complete Works of William Shakespeare, 40 vols. (New York: George D. Sproul, 1907), vol. 38.
5. Stephen Booth, ed., Shakespeare's Sonnets (New Haven: Yale University Press, 1977), 172.
6. See Rollins, ed., The Sonnets, 1:66-67; Colin Burrow, ed., The Complete Sonnets and Poems, the Oxford Shakespeare (Oxford: Oxford University Press, 2002), 426.
7. See Rollins, ed., The Sonnets, 1:66-67.
8. Booth, ed., Shakespeare's Sonnets, 172. For an incisive recent discussion of the sonnet in the light of Shakespearean authorship, see Patrick Cheney, "'O, let my books be... dumb presagers': Poetry and Theater in Shakespeare's Sonnets," Shakespeare Quarterly 52 (2001): 222-54, revised and reprinted in Shakespeare, National Poet-Playwright (Cambridge: Cambridge University Press, 2004), 207-38, in particular 220-25.
9. See Katherine Duncan-Jones, ed., Shakespeare's Sonnets, Arden Shakespeare (Walton-on-Thames, Surrey: Thomas Nelson, 1997), 1-28.
10. Lukas Erne, "For the Stage and the Page," Around the Globe 26 (2004); 36-37.
11. See James P. Bednarz, Shakespeare and the Poets' War (New York: Columbia University Press, 2001). Scholars working on Shakespearean coauthorship have also strengthened our sense of Shakespeare's proprietary sense of his own writings. See, in particular, Brian Vickers, Shakespeare, Co-Author: A Historical Study of Five Collaborative Plays (Oxford: Oxford University Press, 2002); and Jeffrey Knapp, "What Is a Co-Author?" Representations 89 (2005): 1-29.
12. Richard Dutton, "Shakespeare: The Birth of the Author," in Licensing, Censorship and Authorship in Early Modern England: Buggeswords (Basingstoke: Palgrave, 2000), 111-12.
13. See, in particular, Peter Blayney, "The Publication of Playbooks," in A New History of Early English Drama, ed. John D. Cox and David Scott Kastan, 383-422 (New York: Columbia University Press, 1997); Andrew Gurr, "Maximal and Minimal Texts: Shakespeare v. the Globe," Shakespeare Survey 52 (1999): 68-87; and Stephen Orgel, "Acting Scripts, Performing Texts," in Crisis in Editing: Texts of the English Renaissance, ed. Randall McLeod (New York: AMS Press, 1994), 251-94, repr., Orgel, The Authentic Shakespeare and Other Problems of the Early Modern Stage (London: Routledge, 2002), 21-47.
14. H. R. Woudhuysen, "The Foundations of Shakespeare's Text," in Proceedings of the British Academy: 2003 Lectures (Oxford: Oxford University Press, 2004), 74-77 and 88-89.
15. Ibid., 92, 99.
16. Ibid., 84; MacD. P. Jackson, "Francis Meres and the Cultural Contexts of Shakespeare's Rival Poet Sonnets," Review of English Studies 56 (2005): 224-46 and "Vocabulary and Chronology: The Case of Shakespeare's Sonnets," Review of English Studies 52 (2001): 59-75.
17. Jackson, "Francis Meres," 236, 243.
18. Patrick Cheney, Shakespeare, National Poet-Playwright (Cambridge: Cambridge University Press, 2004), and Shakespeare's Literary Authorship (Cambridge: Cambridge University Press, 2008). Of related interest is Charlotte Scott's Shakespeare and the Idea of the Book (Oxford: Oxford University Press, 2007). See also Jason Gleckman, "Shakespeare as Poet or Playwright? The Player's Speech in Hamlet," Early Modern Literary Studies 11.3 (January 2006): 2.1-13 (http://purl.oclc.org/emls/11-3/glechaml.htm), which finds in the fiction of Hamlet, more specifically in the player's speech in act 2, scene 2, an enactment of the tension between the literary drama by Shakespeare, the author, and the theatrical play by Shakespeare, the playwright.
19. Patrick Cheney, ed., The Cambridge Companion to Shakespeare's Poetry (Cambridge: Cambridge University Press, 2007).
20. Alan H. Nelson, "Shakespeare and the Bibliophiles: From the Earliest Years to 1616," in Owners, Annotators and the Signs of Reading, ed. Robin Myers, Michael Harris, and Giles Mandelbrote, 70 (London: Oak Knoll Press, 2005).
21. Some of the work presented at this conference is published in Richard Meek, Jane Rickard, and Richard Wilson, eds., Shakespeare's Book: Essays in Reading, Writing and Reception (Manchester: Manchester University Press, 2008).
22. I quote from the seminar description posted on the conference Web site at http://www.ncl.ac.uk/niassh/shakespeare. The BSA seminar was led by Neil Rhodes. The SAA paper session was chaired by the present author and featured papers by Patrick Cheney, Katherine Duncan-Jones, and Richard Helgerson.
23. I quote from the panel description posted on the conference's Web site, www.shakespeare2006.net. The panel featured contributions by Sukanta Chaudhuri (chair), Paul Eggert, and Lena Cowen Orlin.
24. Catherine Belsey, review of Shakespeare, National Poet-Playwright, Shakespeare Studies 34 (2006): 170.
25. John Jowett, "Editing Shakespeare in the Twentieth Century," Shakespeare Survey 59 (2006): 18-19.
26. Stanley Wells, introduction to William Shakespeare: The Complete Works, ed. Stanley Wells and Gary Taylor, with John Jowett and William Montgomery (Oxford: Oxford University Press, 1986), xxxvii, xxxiii.
27. Ibid., xxxiv; Paul Werstine, "McKerrow's 'Suggestion' and Twentieth-Century Shakespeare Textual Criticism," Renaissance Drama 19 (1988): 149-73. See also Andrew Gurr, "A New Theatre Historicism," in From Script to Stage in Early Modern England, ed. Peter Holland and Stephen Orgel, 71-72 (Basingstoke: Palgrave Macmillan, 2004).
28. David Scott Kastan, Shakespeare after Theory (London: Routledge, 1999), 65.
29. Brian Vickers, "Are all of them by Shakespeare?" Times Literary Supplement, August 11, 2006,10.
30. Ann Thompson and Neil Taylor, eds., Hamlet, Arden Shakespeare (London: Thomas Learning, 2006), 8-13, 506-9. The companion volume by the same editors -- Hamlet: The Texts of 1603 and 1623 (London: Thomas Learning, 2006) -- contains the First Quarto and the Folio texts.
31. Stanley Wells, Re-Editing Shakespeare for the Modern Reader (Oxford: Clarendon Press, 1984), 68.
32. See Erne, Shakespeare as Literary Dramatist, 227-30.
33. For an essay that argues against added stage directions and advocates instead discussion of staging options in the commentary, see John D. Cox, "Open Stage, Open Page? Editing Stage Directions in Early Dramatic Texts," in Textual Performances: The Modern Reproduction of Shakespeare's Drama, ed. Lukas Erne and M. J. Kidnie, 178-93 (Cambridge: Cambridge University Press, 2004). For the argument that editors should do more to mediate to modern readers the specificities of the early modern printed play text, see Lukas Erne, "Shakespeare for Readers," in Alternative Shakespeares 3, ed. Diane E. Henderson (London: Routledge, 2007), 78-94.
34. Wells, Re-Editing Shakespeare for the Modern Reader, 70, 69.
35. I refer to the through-line numbering (TLN) adopted in Charlton Hinman, ed., The First Folio of Shakespeare: The Norton Facsimile, 2nd ed. (1968; New York: Norton, 1996).
36. For a fuller development of the present argument, see Lukas Erne, Shakespeare's Modern Collaborators (London: Continuum, 2008).