Origin. The tale of Bernabò da Genova in the Decamerone, which Shakespeare had probably read in the original. There was an imitation of it in a tract called Westward for Smelts, of which, however, no edition earlier than 1620 is known. For the historical part, he, of course, had resorted to Hollinshed.
In these plays we may, I think, distinguish four different phases of composition, in each of which the thoughts and the language of the poet present a peculiar appearance.
The first phase extends we may say from 1593 to 1598, and contains the plays in Meres's list—except 1 Henry IV., and The Merchant of Venice, and The Taming of the Shrew. It is distinguished by a continual play on words and by frequent rimes—both in couplets and in stanzas—while the blank verse, which is as yet unformed, is harmonious and almost always decasyllabic. Richard III. seems to form the connecting link between this and the next phase; for it is free from both rimes and play on words, while the blank verse has not yet acquired its appropriate form.
The second phase would seem to extend from 1598 to 1603. It contains The Merchant of Venice, As You Like It, Much Ado about Nothing, The Merry Wives of Windsor, Twelfth Night, 1 and 2 Henry IV., Henry V., Hamlet, Othello, Julius Cæsar, Antony and Cleopatra. Here the dramatic blank verse is perfect, trisyllabic feet being admitted, and the lines running into each other, rimes only appearing in final couplets. There rarely occurs a play on words, and the language is in general easy and natural.
The third phase may extend from 1603 to 1609. It contains Measure for Measure, Lear, Macbeth, Troilus and Cressida, Timon of Athens, and Coriolanus. In this the poet returned to the practice of giving passages of several lines in rime, though not in stanzas, and his language is obscured by periphrases, inversions, and ellipses to such an extent that many places—the speeches of Ulysses, for instance, in Troilus and Cressida—must have been perfectly unintelligible to an ordinary audience. He had already, as in Antony and Cleopatra, begun to place more frequently the preposition or conjunction at the end of one line and the word connected with it at the beginning of the next, and he continues to do so here, chiefly in Coriolanus, though hardly at all in Troilus and Cressida, or in Timon of Athens.
The fourth and last phase contains The Winter's Tale, The Tempest, Henry VIII., Cymbeline. He here seems to have made a return to the simpler language of the second phase. In Henry VIII. and The Tempest, what has been said of prepositions and conjunctions goes on to a great extent.
The plays above noticed—thirty-two in number—are the genuine productions of the poet. Two of them, King John and The Taming of the Shrew, were founded on plays that are still extant, and we may see that he used them precisely as he did the tales and chronicles on which he founded so many of his other plays, taking the story, the incidents, the characters, and, when it suited his purpose, the language which they contained.
But beside these, we find in the folio four other plays of a different kind, of which the most that any critic ventures to assert is that they were retouched, improved, and enlarged by Shakespeare. Of two of these, namely The Second and Third Parts of Henry VI. this would seem to be the truth; for we have the two plays in their original form, and there can be little doubt that it was them chiefly that Green had in view in the passage quoted above from his Groat's Worth of Wit, &c.; and upon examination it appears that in the first of them Shakespeare's additions and improvements amount to a fifth, in the second to only an eighth part of the text. Of the other two, The First Part of Henry VI. and Titus Andronicus, after a very careful study of them, my decided opinion, and apparently that of Mr. Dyce also, is that, with an exception presently to be noticed, neither the one nor the other contains a single speech or even a single line from the pen of Shakespeare. How they got into the folio is a question not easy to answer. Heminge and Condell, no doubt, may not have been critics, and so may have fancied that he had had to do with The First Part of Henry VI. also; or they may have merely inserted it as being connected with the other Parts. As to Titus Andronicus, I have already given a reason for its appearance in Meres's list. He had probably heard that it was by Shakespeare, and he made no exact inquiry, and so ascribed it to him; and the editors of the folio may have taken it on his authority, or have followed the same tradition. I do not believe that it was at any time in Shakespeare's nature to write the horrors of one of these plays, or to treat the noble Maid of Orleans as she is treated in the other, or even to labour on and improve the pieces that contained them. Besides, there are nowhere to be found plays more entirely of one single cast than these are. There is also displayed in them an acquaintance with Horace and others of the ancient Classics which Shakespeare did not possess. They may have been written by either Kyd or Marlow, each of whom had this acquaintance, and also a taste for horrors, and abundant talent for their composition. At the same time I think it possible that, as there is a Clown in Titus Andronicus—the only instance I believe out of the plays of Shakespeare—the two short, trifling, and needless scenes in which he appears may be from our poet's pen, and that hence the play was hastily ascribed to him.