Editing Shakespeare is like painting the Forth Bridge," Jonathan Bate remarked lately. "As soon as you finish the job, you have to start all over again." Perfectly true, if only because the analogy is built into the job description. No editor dreams today of the definitive Shakespeare. We now have provisional, updated, state-of-the-art editions.

The occasion for Bate's remark was the launching of the Arden Shakespeare, Third Series. Arden Two was a slow and courtly process, spanning the 1950s and 1980s. It was never completed. The Sonnets did not appear, two editors having expired before their labours could be crowned. Arden Three should be a brisker affair. What can it offer?

The text is subject to continuous minor adjustment. The balance of argument between Folio and Quarto tends to swing, and Bad Quartos get a better press these days. More important is the ancillary material. An Introduction that was respectable a generation ago - T. S. Dorsch's Julius Caesar (1955), say - is now clearly outmoded. Above all, the record of stage production creates innovation and obsolescence. One seldom hears nowadays the "high academic" line, that the best performances are in the head. The lifespan of a current text may not be much more than a decade or so.

The case for a re-edited text shifts with each title, as can be seen with the first three titles just brought out by Routledge. Titus Andronicus is a remote, peripheral play, notwithstanding the productions by Peter Brook and Deborah Warner. What has changed is our own culture. "But fashions in taste go around and come around, and in its willingness to confront violence, often in ways that are simultaneously shocking and playful, our culture resembles that of the Elizabethans more than that of Dr Johnson." As Bate remarks, movie-goers at least will be familiar with bloody revenge, dismemberment, miscegenation, rape and cannibalism. The rediscovery of Titus Andronicus is founded on cultural engagement, and the new Arden offers committed advocacy.

ÁñÁ«ÊÓƵ

With Henry V, the case is largely text-driven. The relations between the Quarto (1600) and Folio are always subject to editorial dispute; but Gary Taylor's Oxford edition of 1984 went far in its acceptance of Quarto readings. In particular, that itch to make radical change, which convulsed Oxford in the 1980s, impelled Taylor to accept the Quarto substitution of "Bourbon" for "Dauphin" throughout 3.7. The pre-Agincourt scene loses some of its force if the boastful Frenchman is not identified as the Dauphin. The new Arden editor, T. W. Craik, is judicious and balanced in his textual decisions. He also prints, as an appendix, a reduced photostat of the Quarto. With the exceptionally thorough Arden notes, and extensive editorial coverage, including recent stage history, this Henry V is the one to have.

The case for the Arden Two Antony and Cleopatra is critical, not textual. M. R. Ridley's Arden Two edition (1954) broke important ground but is now hopelessly out of date. A new editorial survey of critical issues was needed. So John Wilders's Antony and Cleopatra is most welcome. Wilders was literary advisor to the BBC TV Shakespeare series and has an acute eye for staging problems. I admire the masterly Introduction, in which he moves from considerations of staging Antony and Cleopatra to its structure.

ÁñÁ«ÊÓƵ

The revised view of the play accompanies the stage's re-examination. In 1943, James Agate wrote: "If ever a Shakespeare play calls for music, processions, and Tadema-like excesses in bathroom marble, Antony and Cleopatra is that play." The best recent practice goes the other way. Swift, short scenes need an uncluttered stage. They carry the unstable impressionism that is the key to the play's moral issues. Arden Three is a brilliant and necessary synthesis of what we now know about Antony and Cleopatra.

The Story That The Sonnets Tell is the record of an obsession. A. D. Wraight holds that Christopher Marlowe did not die in 1593. His killing was faked, with Establishment connivance, after which he travelled on the Continent. Marlowe, not the Stratford actor, is the true author of the works we know as Shakespeare's, and his autobiography is to be read in the Sonnets. "The prey of worms, my body being dead,/The coward conquest of a wretch's knife" is the keenest stroke of self-revelation.

Few will accept this thesis. The book has however the virtue of its single thesis, a single-minded drive to re-examine all the relevant data. In a familiar landscape, a great deal looks different. "Shakescene", Greene's term, might not refer to Shakespeare. Alleyn has excellent credentials. The murky waters of Marlowe's killing were not cleared by the inquest. The Italian plays show a detailed familiarity with Italy (topography, canals, proverbs), more extensive than has been thought.

The Sonnets come out looking different, too. The "lovely boy" and the Patron, says Wraight, are not the same. She sees a commissioned group 1-17 addressed to Southampton. Another group is addressed to "Mr W. H.", for whom Wraight accepts Hotson's identification with Will Hatcliffe. The Patron is Thomas Walsingham, to whom many Sonnets are addressed. These multiple identifications create more problems than they solve, but restricting the entire sequence to Southampton and the Dark Lady may be a falsification. This book will stimulate readers to reconsider issues they had thought closed.

ÁñÁ«ÊÓƵ

Ralph Berry is the author of Shakespeare in Performance: Castings and Metamorphoses.