1We have three early texts for Hamlet. The earliest is an apparently unauthorized quarto of 1603, Q1, a text that was unknown until a copy was discovered in 1823 or possibly a little earlier in the library of Sir Henry Bunbury, in his manor house of Great Barton, Suffolk. (See Zachary Lesser's Hamlet after Q1, Universlty of Pennsylvania Press, 2015). Its title page reads as follows:

2Theories vary as to how this printed version of the play came into being. It differs substantially from the second quarto (Q2) of 1604 and from the Folio text of 1623 (F) in many ways. It is about half the length of those other two. Many readings diverge, as in Hamlet's "Why what a dunghill idiote slaue am I" in place of Q2's "O what a rogue and pesant slaue am I" (2.2.550). Hamlet's famous "To be, or not to be" soliloquy, along with the ensuing dialogue between Hamlet and Ophelia, takes place in Q1 considerably earlier than in the other texts, before Hamlet's encounter with the traveling players and his "O what a rogue and pesant slaue am I" soliloquy. The lines in Q1's "To be, or not to be" soliloquy are substantially rearranged. The soliloquy begins "To be, or not to be, I, there's the point," in place of Q2's "To be, or not to be, that is the question." The Queen in Q1, named Gertred, denies any knowledge of the murder of her dead husband, and pledges herself to assist in revenging that murder. Polonius is named Corambis, and Reynaldo is named Montano. Hamlet's erstwhile boyhood companions are "Gilderstone, and Rossencraft" when they first enter.

3Early investigators of Q1, including Alfred W. Pollard, in his Shakespeare's Fight with the Pirates and the Problems of the Transmission of His Text, 1917, have argued that Q1 was a pirated text, stealing the march on rival publishers by putting out a copy of this already-famous play with the help of a minor actor or actors who recalled their parts as best they could. The most plausible culprit, according to this hypothesis, was the actor who played Marcellus and may have doubled as Voltimand and Lucianus, since the printed text is notably more accurate when these characters are on stage. More recent editors, including the general editors of the Oxford Shakespeare of 1987, allow that the Q1 text is indeed technically a pirated one in the technical sense that it appeared in print when the play had been "entered" in the Stationers' Register to another man and before the owner to the publishing rights had been able to produce his own edition. A printer named James Roberts had in fact entered his claim in the Stationers' Register, on 26 July 1602, "for his Copie vnder the handes of mr. Passeild & mr. waterson A booke called the Revenge of Hamlett Prince Denmarke as yt was latelie Acted by the Lo: Chamberleyn his servants." Pollard speculated in 1909 that this entry was Roberts's unsuccessful attempt on behalf of the Chamberlain's Men to block publication by another publisher. Perhaps, as E. K. Chambers has argued (William Shakespeare: A Study of Facts and Problems, 1930, 1.146), Roberts hoped to establish a claim to a popular play that he could then sell or trade to some stationer. Whether Q1 was "pirated" by means of memorial reconstruction or other means is, however, a matter of debate and conjecture. A more neutral name is to refer to Q1 as unauthorized.

4A second and authorized quarto appeared in 1604. It made some use of Q1, in fact, especially in the first act, but it seems to have based its text primarily on Shakespeare's own papers, along with some annotations by the bookkeeper. Q1 may have proved useful to the Q2 compositors when those papers were illegible. This second quarto served as copy for a third in 1611 (Q3), by which time Nicholas Ling had transferred his rights to John Smethick. A fourth quarto (Q4), based on the third, appeared some time prior to 1623 but undated. Still other quartos appeared in 1637 and then as "Players' quarto" after the Restoration in 1660. None of these quartos after Q2 can lay claim to any authority.

5Q2's title page reads as follows:

6Q2 contains a substantial number of misprints, and in some few places it must yield to Q1 as an authority when it used portions of that earlier unauthorized edition as its copy text, but on the whole Q2 has the attested authority of having been printed from Shakespeare's own manuscript as occasionally annotated by the bookkeeper.

7Editors once assumed that the 1623 Folio text was set from a heavily annotated exemplar of Q2, but this assumption has turned out to be not the case. F's numerous departures from Q2 bespeak another manuscript source or sources. The compositors of F in the printing shop of William Jaggard and his son Isaac, identified by textual scholars as B, E, and I, may well have consulted one of the earlier quartos at times, although contamination from Q3 and Q4 appears to have been minor. More important for today's editor is to decide how to handle the many instances in which F and Q2 differ.

8The Folio text of Hamlet omits more than two hundred lines in Q2. Conversely, F adds some shorter materials not in Q2, notably a discussion of some 26 lines between Hamlet and his friends, Rosencrantz and Guildenstern, about a theatrical rivalry between adult actors and "an aerie of children" who have captured the attention of audiences (2.2.337-62). If, as seems plausible, the cuts in F were derived from a text designed for a shorter performance than that embodied in Q2, why then would F also contain new material? Are the cuts substantial enough to have sped performance along to a significant extent? To be sure, the acting company, performing in the afternoons and at times during colder months when the days are short in London's northerly latitude, prompting a concern about spectators getting back to London in the dark, might well have asked for a shortened text. One can imagine Shakespeare writing at considerable length with his well-known facility and then needing to provide his colleagues in the acting company with a shortened practical version. One can also imagine a passage about a theatrical rivalry that must have seen compellingly fascinating to audiences in the early seventeenth century having such immediate appeal as to find its way into a theatrical stage revision even at the expense of a little added material. (The hypothesis that this passage about theatrical rivalry was in the Q2 manuscript but was somehow overlooked or deleted seems less likely.)

9Speculative though the matter remains, this present edition regards the F excisions, especially the substantial ones in toward the end of the play, as a way of coping with length of performance. The ones in Acts 3 to 5 especially are just the sorts of cuts one might want in order to speed the play along. Some 19 or 20 lines are taken out of 3.4, when Hamlet confronts his mother in her chambers, in such a way that the substance of the scene is relatively unimpaired. The King's conspiratorial conversations with Laertes in Act 4 move the action towards its climax and conclusion. A substantial cut in Osric's conversation with Hamlet in 5.1 (106-42) trims down this satirical encounter in such a way as to proceed on to Hamlet's duel with Laertes and the series of violent deaths that follow. A substantial cut in scene one (1.1.112-29) makes for a more direct exposition. So too when Hamlet, on the battlements with Horatio and the guard, nervously awaits the expected arrival of the Ghost (1.4.17-38). We cannot rule out the possibility that Shakespeare, wanting for some reason to revise his great tragedy, trimmed some passages that he felt repeated things said elsewhere or slowed down the action at a critical point. The argument in favor of retaining such passages in a modern edition of Hamlet is that the lines are persuasively and even indisputably Shakespearean. Since they may have been cut in F for reasons of overall length, should they not be allowed to stand?

10Of course this present online edition is specifically designed to let readers and actors compare the early texts, including Q1, in specific detail, so that in the F text the Q2 omissions are indeed omitted, and conversely with F materials not in Q2; they are omitted in this edition's Q2 text. The decision to keep both applies to the "editor's choice" text, which may well not represent Hamlet as it existed at any particular point in time. Importantly, we need to see that Hamlet evolved, in the theatre especially, and that the various early texts give us snapshots of that evolving text at various stages of its existence. The various points in which F agrees with Q1 rather than Q2 supports the view that both Q1 and F are theatrical texts, whereas Q2 is close to an authorial manuscript.

11Q1, for all its odd departures from the other two texts, is invaluable at times in informing us about performance history. Especially in stage directions, the person or persons who helped assemble the Q1 text seem to have relied on visual memory of actual performances. Thus it is that when the Ghost of Hamlet's father appears to Hamlet and the Queen in 3.4, he does so "in his night gowne." Ophelia enters to her mad scene in 4.5 "playing on a Lute, and her haire down singing." Laertes "leapes into the graue" in 5.1, whereupon "Hamlet leapes in after Laertes." In Hamlet's duel with Laertes, "They catch one anothers Rapiers, and both are wounded."

12The "editor's choice" text in this present edition makes use of such materials, and retains cuts in both Q2 and F. The editor is still left with a host of difficult choices of wording, especially between Q2 and F. Proposed cuts for performance are only one factor potentially differentiating the two texts. Many differences arise from typographical errors in one or the other texts, or both. Some differences may well be matters of printing house style. The compositors were free to impose their own spellings and punctuation, or those of the printing shop for which they worked. When F changes Q2's "married with my Vncle" to "married with mine Vnkle" at 1.2 .151, should we take this to be an authorial change or simply a matter of house style? The early texts of Hamlet is filled with minor alterations of this sort. "Mine uncle" has a euphonic cadence that may seem preferable to our ears, but may be an editorial sophistication. The present edition is wary of such alterations. When, on the other hand, F changes Q2's "Thou turnst my very eyes into my soule" to "Thou turnst mine eyes into my very soule" at 3.4.91, the movement of "very" may be a change that Shakespeare saw as a genuine improvement, or it could have been a compositor's error in Q2 in the first place. Or else it could be neither of these things, but the compositor's imposing his sense of style, or simply misreading. The compositor's work was tricky: he had to lay out individual letters by hand into a row, reading backwards so that the printed version would be in the expected order, and was sure to make a number of mistakes. Many alternatives that confront the editor are of this sort, indeterminate or nearly so as to authorial intent. It would be a comfort if textual editing were a neatly manageable science, but it is often not.

13F's authority is to some extent compromised by features suggesting that it was set from a transcript rather than an authorial manuscript, as was the case with Q2. The manuscript lying behind F appears to have been a scribal copy, relatively close in date to 1623 when it was published. While Q2 descends directly from authorial "foul" papers, with occasional input from Q1 that can be determined and evaluated accordingly, F may descend via a prompt-book and then a scribal transcript. Even its occasional consultation of the second quarto may have been affected by contamination in Q3. Q2 is thus more likely to preserve Shakespeare's incidentals.

14On the other hand, an editor must ask, in each instance of variation between Q2 and F, if the F reading can be plausibly explained other than by the hypothesis of authorial revision. The instances favoring F in such cases are numerous. As editor, I have found myself drawn to F's reading more than in my earlier editorial work, when I was impressed by Q2's more direct line of descent from an authorial manuscript. How can one explain F's reading as other than authorial when, in the play's first scene, Marcellus says to Horatio, about the Ghost, "Question it Horatio" rather than Q2's "Speake to it Horatio" (1.1.49)? The meaning is similar, but why would a scribe or compositor make such change? Perhaps the alteration was an authorial one, made to avoid repeating Marcellus's "speake to it Horatio" three lines earlier. Even more substantively, when Horatio, in F, refers to young Fortinbras's followers as "Landlesse resolutes" rather than Q2's "lawlesse Resolutes" (102), the shift in meaning is more pronounced and would seem to require authorial intervention. "Landlesse" suggests restless and ambitious younger brothers lacking landed title. The shift from "compulsatory" to "Compulsatiue" a few lines later improves the meter of the blank verse line, "And termes Compulsatiue, those foresaid Lands" (107).

15The famous crux in scene 2, when Hamlet's "O that this too too sallied flesh would melt" in Q2 becomes "Oh, that this too solid Flesh, would melt" in F (129), is perhaps more complex as a choice, since "sallied" could have been a misprint for "sullied" that a scribe or compositor might then have "improved" in his own way. But since the F reading is "solid" in more ways than one, and since Shakespeare does indeed seem to have made numerous revisions of this sort, the F reading takes priority. Polonius's urging his son Laertes in F to be wary of dulling his palm "with entertainment / Of each new-hatched, unfledged comrade" rather than Q2's "unfledged courage" (1.3.45) also poses a complex choice, since "comrade" might conceivably have been some copyist's way of coping with the less familiar "courage," which could have been a poet's imaginative way of suggesting "swashbuckler," but once again the F wording could have appealed to Shakespeare in the act of revision as a means of clarifying what he had said. "Courage" in Q2 could have been a misprint for something else.

16The overall consequence of such delicate issues is that the "editor's choice" text in the present edition is more in line with what recent bibliographical scholarship has persuasively argued, that we should not be unduly wary of the idea that Shakespeare did on some occasions have an opportunity to rewrite, and that he apparently did so freely. The phenomenon is observable in some other major plays as well, especially Othello and King Lear.