NEARLY EVERYONE AGREES THAT William Shakespeare’s career as a playwright began in about 1590, but there is much less agreement on which plays began it. Depending on whose authority you favor, Shakespeare’s debut written offering might be any of at least eight works: The Comedy of Errors, The Two Gentleman of Verona, The Taming of the Shrew, Titus Andronicus, King John, or the three parts of Henry VI.

The American authority Sylvan Barnet lists The Comedy of Errors as Shakespeare’s first play with Love’s Labour’s Lost second, but more recently Stanley Wells and Gary Taylor, in the Oxford Complete Works, credit him with ten other plays-more than a quarter of his output-before either of those two comes along. Wells and Taylor place The Two Gentlemen of Verona at the head of their list-not on any documentary evidence, as they freely concede, but simply because it is notably unpolished (or has “an uncertainty of technique suggestive of inexperience,” as they rather more elegantly put it). The Arden Shakespeare, meanwhile, puts The Taming of the Shrew first, while the Riverside Shakespeare places the first part of Henry VI first. Hardly any two lists are the same.

For many plays all we can confidently adduce is a terminus ad quem-a date beyond which they could not have been written. Sometimes evidence of timing is seen in allusions to external events, as in A Midsummer Night’s Dream, in which seemingly pointed references are made to unseasonable weather and bad harvests (and England had very bad harvests in 1594 and 1595), or in Romeo and Juliet when Nurse speaks of an earthquake of eleven years before (London had a brief but startling one in 1580), but such hints are rare and often doubtful anyway. Many other judgments are made on little more than style. Thus The Comedy of Errors and Titus Andronicus “convey an aroma of youth,” in the words of Samuel Schoenbaum, while Barnet can, without blushing, suggest that Romeo and Juliet came before Othello simply because “one feels Othello is later.”

Arguments would run far deeper were it not for the existence of a small, plump book by one Francis Meres called Palladis Tamia: Wit’s Treasury. Published in 1598, it is a 700-page compendium of platitudes and philosophical musings, little of it original and even less of it of interest to history except for one immeasurably helpful passage first noticed by scholars some two hundred years after Shakespeare’s death: “As Plautus and Seneca are accounted the best for comedy and tragedy among the Latins, so Shakespeare among the English is the most excellent in both kinds for the stage. For comedy, witness his Gentlemen of Verona, his Errors, his Love Labour’s Lost, his Love Labour’s Won, his Midsummer Night’s Dream, and his Merchant of Venice; for tragedy, his Richard the Second, Richard the Third, Henry the Fourth, King John, Titus Andronicus, and his Romeo and Juliet.”

This was rich stuff indeed. It provided the first published mention of four of Shakespeare’s plays-The Merchant of Venice, King John, The Two Gentlemen of Verona, and A Midsummer Night’s Dream-and additionally, in a separate passage, established that he had written at least some sonnets by this time, though they wouldn’t be published as a collected work for a further eleven years.

Rather more puzzling is the mention of Love’s Labour’s Won, about which nothing else is known. For a long time it was assumed that this was an alternative name for some play that we already possess-in all likelihood The Taming of the Shrew, which is notably absent from Meres’s list. Shakespeare’s plays were occasionally known by other names: Twelfth Night was sometimes called Malvolio, and Much Ado About Nothing was sometimes Benedick and Beatrice, so the possibility of a second title was plausible.

In 1953 the mystery deepened when an antiquarian book dealer in London, while moving stock, chanced upon a fragment of a bookseller’s inventory from 1603, which listed Love’s Labour’s Won and The Taming of the Shrew together-clearly suggesting that they weren’t the same play after all, and giving further evidence that Love’s Labour Won really was a separate play. If, as the inventory equally suggests, it existed in published form, there may once have been as many as 1,500 copies in circulation, so there is every chance that the play may one day turn up somewhere (a prospect thought most unlikely for Shakespeare’s other lost play, Cardenio, which appears to have existed only in manuscript). It is all a little puzzling. If Love’s Labour’s Won is a real and separate play, and was published, a natural question is why Heminges and Condell didn’t include it in the First Folio. No one can say.

In whatever order the plays came, thanks to Meres we know that by 1598, when he had been at it for probably much less than a decade, Shakespeare had already proved himself a dab hand at comedy, history, and tragedy, and had done enough-much more than enough, in fact-to achieve a lasting reputation. His success was not, it must be said, without its shortcuts. Shakespeare didn’t scruple to steal plots, dialogue, names, and titles-whatever suited his purpose. To paraphrase George Bernard Shaw, Shakespeare was a wonderful teller of stories so long as someone else had told them first.

But then this was a charge that could be laid against nearly all writers of the day. To Elizabethan playwrights plots and characters were common property. Marlowe took his Doctor Faustus from a German Historia von D. Johann Fausten (by way of an English translation) and Dido Queen of Carthage directly from the Virgil’s Aeneid. Shakespeare’s Hamlet was preceded by an earlier Hamlet play, unfortunately now lost and its author unknown (though some believe it was the hazy genius Thomas Kyd), leaving us to guess how much his version owed to the original. His King Lear was similarly inspired by an earlier KingLeir. His Most Excellent and Lamentable Tragedy of Romeo and Juliet (to give it its formal original title) was freely based on the poem The Tragicall History of Romeus and Juliet by a promising young talent named Arthur Brooke, who wrote it in 1562 and then unfortunately drowned. Brooke in turn had taken the story from an Italian named Matteo Bandello. As You Like It was borrowed quite transparently from a work called Rosalynde, by Thomas Lodge, and The Winter’s Tale is likewise a reworking of Pandosto, a forgotten novel by Shakespeare’s bitter critic Robert Greene. Only a few of Shakespeare’s works-in particular the comedies A Midsummer Night’s Dream, Love’s Labour’s Lost, and The Tempest- appear to have borrowed from no one.

What Shakespeare did, of course, was take pedestrian pieces of work and endow them with distinction and, very often, greatness. Before he reworked it Othello was insipid melodrama. In Lear’s earlier manifestation, the king was not mad and the story had a happy ending. Twelfth Night and Much Ado About Nothing were inconsequential tales in a collection of popular Italian fiction. Shakespeare’s particular genius was to take an engaging notion and make it better yet. In The Comedy of Errors, he borrows a simple but effective plot device from Plautus-having twin brothers who have never met appear in the same town at the same time-but increases the comic potential exponentially by giving the brothers twin servants who are similarly underinformed.

Slightly more jarring to modern sensibilities was Shakespeare’s habit of lifting passages of text almost verbatim from other sources and dropping them into his plays. Julius Caesar and Antony and Cleopatra both contain considerable passages taken with only scant alteration from Sir Thomas North’s magisterial translation of Plutarch, and The Tempest pays a similar uncredited tribute to a popular translation of Ovid. Marlowe’s “Whoever loved that loved not at first sight?” from Hero and Leander reappears unchanged in Shakespeare’s As You Like It, and a couplet from Marlowe’s Tamburlaine-