Now Shakespeare not only improved a great deal while composing his plays, but also conformed, to some extent at least, to the different tastes of his audience at different periods of his life. Hence, a knowledge of the order in which his plays were written is very valuable, and should form the first step in a careful study of his writings.
Unfortunately, when we attempt to arrange Shakespeare's plays in chronological order, we encounter many practical difficulties in finding just what this order is. We know that Tennyson developed a great deal as a poet between the ages of eighteen and thirty-three; and we can show this by pointing to four successive volumes of his poems, published respectively at the ages of eighteen, twenty-one, twenty-three, and thirty-three, and each rising in merit above the one before it. We know definitely in what order these volumes come, for we find on the title-page of each the date when it was printed. But scarcely half of Shakespeare's plays were printed in this way during his life. The others, some twenty in all, are found only in one big folio volume which gives no hint of their proper order or year of composition, and which bears on its title-page the date of the printing, 1623, seven years after Shakespeare died. Many plays, too, published early, were written some years before publication, so that the date of printing on the flyleaf of the quarto, even where a quarto exists, simply shows that the play was written sometime before that year but does not tell at all how long before. How, then, are we to trace Shakespeare's growth from year to year, through his successive dramas, when the quartos help us so little and when the majority of these dramas are piled before us in one volume by the editors of the First Folio, without a word of explanation as to which plays are early attempts and which mature work?
At first sight the above problem seems almost hopeless. The researches of scholars for over a century, however, have gathered together a mass of evidence which determines pretty accurately the order in which these different plays were written.
This evidence is of two kinds, external and internal. By external evidence we mean that found outside of the play, references to it in other books of the time, and similar material. By internal evidence we mean that found inside of the play itself.
External Evidence.—This is of several kinds. In the first place, every play which was to be printed had to be entered in the Stationers' Register, and all these entries are dated. Hence we know that certain plays were prepared for publication by the time mentioned. For instance, "A Book called Antony and Cleopatra" was entered May 20, 1608; and although apparently the book was not finally printed at that time, and although our only copy of Antony and Cleopatra is that in the Folio of 1623, yet we feel reasonably certain from this entry that this play must have been written either in 1608 or earlier. In addition to the record of the Stationers' Register, we have the dates on the title-pages of such plays as appeared in Quarto. These evidences, it must be remembered, determine only the latest possible date for the play, as many were written long before they were printed, or even entered.
Again, other men sometimes used in their books expressions borrowed from Shakespeare or remarks which sound like allusions to something of his. Here, if we know the date of the other man's book, we learn that the play of Shakespeare from which he borrowed must have been in existence before that date. Thus, when the poet Barksted prints a poem in 1607 and borrows a passage in it from Measure for Measure, we conclude that Measure for Measure must have been produced before 1607, or Barksted could not have copied from it. This form of evidence has its dangers, since occasionally we cannot tell whether Shakespeare borrowed from the other man or the other man from him; nevertheless it is often valuable.
Furthermore, we sometimes find in contemporary books or papers, which are dated, an account of the acting of some play. A law student named John Manningham left a diary in which he records that on February 2, 1602 he saw a play called Twelfth Night or What You Will in the Hall of the Middle Temple; and his account of the play shows that it was Shakespeare's. Dr. Simon Forman, in a similar diary, describes the performance of three Shakespearean plays, two of the accounts being dated. Still more important in this class is the famous allusion, already quoted, by Francis Meres in his Palladis Tamia, a book published in 1598. In this he mentions with high praise six comedies of Shakespeare: The Two Gentlemen of Verona, The Comedy of Errors, Love's Labour's Lost, Love's Labour's Won,[[1]] A Midsummer Night's Dream, and The Merchant of Venice; and six "tragedies": Richard II, Richard III, Henry IV, King John, Titus Andronicus, and Romeo and Juliet.[[2]] Hence, we know that all these plays were written and acted somewhere before 1598, although three of them did not appear in print until 1623.
The above list does not exhaust all the forms of external evidence, but merely shows its general nature. External evidence, as can be seen, is not something mysterious and peculiar, but simply an application of common sense to the problem in hand.
Frequently two pieces of external evidence will accomplish what neither one could do alone. Often one fact will show that a play came somewhere before a certain date, but not show how long before, and another will prove that the play came after another date, without telling how long after. For example, King Lear was written before 1606, for we have a definite statement that it was performed then. It was written after 1603, for it borrowed material from a book printed in that year. This method of hemming in a play between its earliest and its latest possible date is common and useful, both with Shakespeare and with other writers.
Internal Evidence.—By the above methods a few plays have been dated quite accurately, and many others confined between limits only two or three years apart. But many plays are still dated very vaguely, and some are not dated at all. For further results we must fall back on internal evidence. The first, though by no means the most important, form of this consists of allusions within the play to contemporary events. If a boy should read in an old diary of his grandmother's that she had just heard of the fight at Gettysburg, he would feel certain that the words were written a few days after that great battle, even if there were no date anywhere in the manuscript. In the same way, when the Prologue of Shakespeare's Henry V alludes to the fact that Elizabeth's general (the Earl of Essex) is in Ireland quelling a rebellion, we know that this was written between April and September of 1599, the period during which Essex actually was in Ireland. Similarly, certain details in The Tempest appear to have been borrowed from accounts of the wreck of Sir George Somers's ship in 1609. As Shakespeare could not have borrowed from these accounts before they existed, he must have written his comedy sometime after 1609.[[3]]