Voices of Spring

Voices of Spring
Yuhui Choe and Alexander Campbell
© ROH/ Tristram Kenton

< back to News & Events


25 June 2019 at the Wallace Collection, Hertford House, Manchester Square, London W1.
Presented by the Frederick Ashton Foundation in association with the Wallace Collection.
Speaker: Sir Nicholas Hytner

“We must leave the killing out.” Keeping the classics of the stage alive

It's a very great honour to be asked to give the first Frederick Ashton Lecture, but it is good to be reassured that the subject matter of what promises to be a wonderful series will be the arts in their widest sense. I love the art to which Sir Fred devoted his life, but I'm really not qualified to talk about it in any depth.

It could be worse. There are things I know less about. Nobody would want to hear my Diego Maradonna lecture. But I have knocked around the theatre for quite a long time, and I've been in and out of opera houses throughout my career. So, let me try to make some observations, and ask some questions, about the staging of the great classics of the stage based mainly on my experience of those I've directed.

I'll start with my title:

“We must leave the killing out.”

Which may sound like I'm about to argue for the return of censorship. But in fact, it's from A Midsummer Night's Dream. It's what the tailor, Starveling, says when Bottom warns the company of amateur actors that they're skating on thin ice with their plans to present Pyramus and Thisbe before the Duke.

"There are things in this comedy of Pyramus and Thisbe that will never please," says Bottom. "First, Pyramus must draw a sword to kill himself, which the ladies cannot abide. How answer you that?"

Consternation among the actors.

"I believe we must leave the killing out, when all is done," says Starveling.

Bottom, of course, has a more creative device to solve the problem. He suggests a big re-write - a prologue, which seems to say that:

"We will do no harm with our swords and that Pyramus is not killed indeed. And for the more better assurance, tell them that I, Pyramus, am not Pyramus, but Bottom the weaver. This will put them out of fear."

I am entirely sympathetic to the actors' concerns.

They're in rehearsal for an old play that has things in it that the audience might not understand.

Or even if the audience understands the play, they may find it offensive, or confusing, or out of date.

And even if they're happy to be offended, will the play have anything to say to an audience of newly-weds? How interested will they be in a story where love ends unhappily in a double suicide, as a result of the surprise intervention of a lion?

So good on the actors, you might say, for deciding to cut it, change it, and generally tart it up. How otherwise will they be able to keep their old play alive?

In fact, Bottom, Starveling and company are thinking their way through the challenges that confront everyone who's job it is to bring to the stage an old work from the past. I can't pretend that their solutions are successful. Their performance of Pyramus and Thisbe is a disaster, though I've seen worse, even on London's grandest stages.

But it's worth asking Bottom's questions again. Here's a great old play, or opera, or ballet. We know it's great, but we also know that the audience that it was made for is long dead, and that today's audience brings to it a significantly different way of looking at the world. What do we do to bridge the gap?

Do we stage it as it would have been staged originally - that is as a new work that speaks directly to our own contemporaries? Or do we insist that it's an old work and ask our audience to imagine itself back in the world for which it was written?

Do we expect our audience to understand the formal constraints of the theatre that the work was originally created for? Or do we allow the work to speak through the dramatic conventions that communicate most directly to today's audience?

Do we ask our audience to share the concerns and the world view of the original audience? Or do we look in the old play only for what speaks to today? Or look only for what is universal to human experience over the centuries?

If I've learned anything from repeatedly asking myself these questions, it is that there's no right answer to any of them.

Suppose first that the most vivid response to a particular play is to treat it as if it's hot off the press, a response to what's going on in our own world. This, you could say, is the most authentic way of tackling the great classics of the English stage. It was only in the twentieth century that the idea took hold that we might best do justice to Shakespeare by recreating the world he wrote his plays for. For at least three centuries, it was assumed that his plays should be heavily adapted and subjugated to contemporary theatrical expectations. Shakespeare himself was cavalier about history. His Roman plays may be based on Plutarch's Lives of the Ancient Romans, but historical authenticity doesn't appear to figure on his agenda. The Roman mob, just for a start, is lifted straight from the streets of Elizabethan London And rehearsing Julius Caesar last year at the Bridge Theatre, it was impossible not to see in it an uncanny reflection of our own predicament.

As the play starts, the leaders of the metropolitan elite (Brutus, Cassius and company) are terrified that the state could slide at any moment into tyranny. They're convinced that they're working for the common good, but they're equally concerned for their own position in a system that works well for them. Many of them went to school together. Their privilege comes with their class and education.

So, they decide to get rid of the tyrant Caesar before he gets the chance to abuse his power any further. They suppose the streets will echo to the cry of freedom. They have no doubt that the people will share their fear for the future.

But their arrogance is their undoing. Their remoteness from the street blinds them to the fragility of their grip on popular opinion. The liberal establishment is trounced by a demagogue (Mark Anthony) who appeals to the gut, and who tells the stories the mob wants to hear. The masses turn on the liberals. The tyrant is replaced by a younger tyrant, who’s more ruthless. Once the body politic is infected with the virus of authoritarianism, it can't be eradicated.

The recreation of Ancient Rome for Shakespeare was a means to an end. He often used history as a way of exploring dangerous subject matter that would have had him thrown in jail if he'd gone at it directly. The Elizabethan authorities would not have looked kindly on a play about any of the several attempts to assassinate Queen Elizabeth. Better to write about the assassination of Caesar. Still, the Elizabethan audience expected to see themselves on the stage. Great plays are worth reviving because they continue to hold the mirror up to us.

Of course, you can ask a modern audience to look for themselves in a painstaking reconstruction of the late Roman Republic. You can also do what the Elizabethan actors did: they threw sheets over their street clothes in a rough approximation of togas, so their Ancient Rome was a mix of ancient and modern. But whatever you do, you have to acknowledge that all great works of the past change with every new company of performers and with every new audience.

A case in point: a few months before we staged Julius Caesar at the Bridge, it was produced by New York's Public Theatre, who described Caesar as

"a force unlike any the city has seen. Magnetic, populist, irreverent, he seems bent on absolute power. A small band of patriots, devoted to the country's democratic traditions, must decide how to oppose him."

And they went for it. Caesar had bouffant blonde hair, and a red tie that hung pendulously below his waist. In a borough that voted 9 to 1 against Donald Trump, the production delivered a sensational commentary on the state of the nation, though outraged Trumpists who hadn't seen it bullied the Public Theatre's commercial sponsors into withdrawing their support. It's hard to imagine better evidence of Shakespeare's continued pugnacity as an analyst of contemporary politics.

In New York, faced with a Trump presidency, you can see how necessary it must have been to give Brutus, Cassius and their "small band of patriots" custody of "the country's democratic traditions." In London, however, I found it impossible to see in them an unequivocal endorsement of liberal presumptions.

Julius Caesar seems to me to address the failure of dismayed liberals to understand and overcome the appeal of populism. It exposes the manipulative half-truths and outright falsehoods that are the populists' stock in trade. It is unsentimental about the gullibility of the multitude. As a member of the establishment says, if Caesar had stabbed their mothers, they would have forgiven him. Mark Antony later produces Caesar's will and promises every Roman a hand-out of 75 drachmas. Job done, he conspires to change the will because he has a war to finance. 75 drachmas, £350 million per week, whatever it takes.

Over my years at the National, I often staged plays by Shakespeare that seemed to speak directly about what was happening in the world outside the theatre, across the river in Westminster, on the streets of the West End. It would have seemed perverse not to treat them as new plays.

Henry V, in 2003, in the context of the build-up to the Iraq War, was quite clearly about a charismatic young English leader who commits his troops to a dangerous foreign invasion, for which he has to struggle to find justification in international law. Timon of Athens is about a plutocrat and philanthropist, a man whose estimation of his own worth is entirely financial. After he runs out of money, he faces a catastrophic credit crunch. Those he thought to be his friends - a bunch of unscrupulous bankers, politicians and their disgusting hangers-on in the arts world - desert him. He eventually dies of self-disgust. It's one of Shakespeare's least performed plays, but it is often rediscovered in the wake of a financial crisis. We did it at the National shortly after 2008.

The urge to treat these plays as contemporary testimony is often irresistible. You submit to the impression that the great playwrights are writing for us, now. What you risk losing when you do that, of course, is that they were undoubtedly writing for their own audience, then. It's a serious loss, but not a permanent one. There's always next time.

And maybe the losses are just as serious if you start from the premise that the best way to communicate a great classic is through a recreation of the world it was written for. A couple of years after I did Henry V, I staged both parts of Henry IV.

I felt more constrained by them than I did by Henry V. I was wary of bringing too much of the present to them. I wasn't convinced that a narrative about civil war in England, the consequences of regicide, and the stricken conscience of the usurper would make much sense if the plays were presented in a modern context, as they were when staged originally in the 1590s. I thought in 2005 that they were driven by a specifically Elizabethan terror of a return to the bloody chaos of the Wars of the Roses. Now, in 2019, that terror feels much closer. Town and country are divided, and so are North and South, Scotland and England.

In 2005, I overestimated the stability of our body politic and I wish I could have another go at tapping into folk memories of civil strife.

But there's a danger of becoming too fixated on period. I can think of many fine productions from my time at the National that were meticulous in their recreation of an old world: the director Howard Davies staged a wonderful series of early twentieth century Russian plays. Marianne Elliott brilliantly directed Ibsen's Pillars of the Community. I was very fond of my own Much Ado About Nothing, which was set in an approximation of late sixteenth century Sicily, which is where the play purports to be set, though in fact it must originally have been set - if set is the right word - on an empty stage in the middle of London, because the conventions of the Elizabethan stage were nothing like our own.

And in fact, our approach to the theatrical conventions of the great works of the past is as important as our imaginative response to the worlds they explore. Bottom and his company of amateur actors are just as keen to grapple with how to present their terrible old play as they are with what it says and how it might affect the audience. How will they bring moonlight into a chamber? How will they signify a wall? Should their acting be lofty, or condoling? Should we, following their example, recreate the stylistic conventions of the past or remake them anew?

As it happens, thanks to the work of Shakespeare's Globe, we can get a pretty good idea of what happens if we recreate the conventions of Elizabethan theatre. It's always illuminating to see a play outside on a summer's afternoon, or inside the tiny candlelit Wanamaker Theatre, a beautiful copy of the Jacobean Blackfriars Theatre. What can’t be recreated is the Jacobean audience, or the Jacobean context. We are obliged to experience so-called authentic productions inauthentically. Their authenticity is remote and strange, where it would once have been the norm.

Some of the conventions of the English classical theatre have, in fact, survived. It is maybe more astonishing that it often seems to be that good actors can so effectively communicate text that is more than four centuries old. In many respects, acting an old play requires the same dedication to the basics of the craft as singing an opera or dancing a classical ballet. An actor must think, breathe and feel through long, sinuous paragraphs. She has to reconcile this with the ever-evolving demand to be real and natural, knowing that what seemed real ten years ago seems stagey today, and what seemed natural a hundred years ago now seems ridiculous. And all this as she lets us see who she is and feel the workings of her heart.

And she'll quickly realise that the demands made by the contemporary repertoire are often no different to those made by the classical repertoire. Here's Shakespeare's Beatrice, who requires from an actor pirouettes of intellectual dexterity:

"He that hath a beard is more than a youth, and he that hath no beard is less than a man; and he that is more than a youth is not for me, and he that is less than a man, I am not for him."

And here's Wilde's Lady Bracknell three hundred years later, who demands the same attention to phrasing and breath control:

"To be born, or at any rate bred, in a hand-bag, whether it had handles or not, seems to me to display a contempt for the ordinary decencies of family life that reminds one of the works excesses of the French Revolution."

And here's Alan Bennett's Mrs Lintott a hundred years after that, still in the high style:

"Can you, for a moment, imagine how dispiriting it is to teach five centuries of masculine ineptitude? History is a commentary on the various and continuing incapabilities of men. What is history? History is women following behind with the bucket."

Of course, none of these people speak the way people actually spoke, even then. Their wit is contrived, their fluency heightened. The actor's job, like the singer's and the dancer's, is to make people believe the extreme articulacy of the way they communicate. The director's is to create a stage world where that form of communication seems spontaneous and inevitable.

Which is a good place to think for a while about opera, where spontaneity would seem to be directly in conflict with the need to communicate a precisely written series of musical instructions.

Handel's Xerxes, in 1985 at English National Opera, was my high-water mark as an opera director. No opera production I've done since has come near it. And that's maybe because Handel's Italian operas are so remote from what the modern audience thinks of as the real world, that they liberate you, if you're the director, to lead an audience into a world of your own invention.

Xerxes follows the rigid conventions of Baroque Opera Seria: a succession of long arias, interrupted by recitative, the sparsely musicalised dialogue that carries the narrative. Each aria expands on a single emotion in an A-section, turns a corner in a shorter B-section, then repeats the A-section all over again. Ad nauseam, you might say, if you don't like it.

And Opera Seria, even more than the rest of the operatic repertoire, calls for a form of acting that would be beyond most actors in the spoken theatre, where time is in the actor's hands, where thought follows thought at something like the same pace thoughts follow each other in life. In opera, the performer is often required to stay with a thought far longer than she'd hold it in reality. Time moves at the pace of the music, and it's the singer's job to persuade the audience that she's showing them something like the truth.

This is something to bear in mind when people talk about whether opera singers can act or not: the way they have to act would flummox actors who don't sing for a living. Imagine asking a brilliant young star of the spoken stage to play Tosca. She wouldn't be able to sing it - that's a given. But she probably wouldn't be able to act it either, because she'd find herself confined by Puccini.

In Xerxes, the King was played by the great mezzo soprano Ann Murray. Opera, of course, embraced gender fluidity long before the spoken theatre caught up with its temptations. In a typical aria, she had to stay with one single idea about her beloved:
When I see her, I would hold her,
For my heart aches to enfold her,
And swells with desire.

She had to sing it again and again and again, embellish desire with long roulades, ache and swell for four minutes until she was allowed to move on to a B section about cherishing and perishing and consuming fire, and then she was back for another four minutes of holding, enfolding and swelling with desire.

Here in a nutshell is why opera should theoretically be unstageable. It would be hard to argue that anything that you would conventionally define as drama happens during Xerxes' nine-minute aria. It contains no conflict. It doesn't move the story forward. Xerxes discovers one thing about himself - that he's fallen hard for a woman - and he goes on discovering the same damn thing until the composer has finished satisfying the demands of his chosen form. Then after a brief burst of dialogue that advances a frankly ridiculous plot, somebody else spends nine minutes singing about how miserable they are.

It worked in 1985 because Ann Murray mesmerised the audience with nothing but a row of green deck-chairs, her voice, her musicality, and an opera-singer's supernatural concentration. She had the support of a great conductor, Charles Mackerras, as much a man of the theatre as a musician, who brought palpable erotic tension to music that in other hands could sound merely beautiful.

Xerxes also presented such an extreme challenge to the director that it elicited from me a degree of invention and critical thought that I’ve been hard pressed to match in the opera house in the years since.

Because a worm of doubt was gnawing away at me. What makes an opera worth staging is always its score. Its principal language of communication is music. Its libretto is effective to the extent that it draws genuinely dramatic music from the composer: music that reveals character, creates conflict, builds momentum, generates feeling. The literary quality of the text is secondary. It need not make much narrative sense: the dramatic coherence of an opera is more a consequence of its music than its plot.

So, I kept asking myself whether I shouldn't be leaving more space for the music to create its own poetry.

I returned to ENO in 1988 with The Magic Flute. This is popular entertainment, I thought, its libretto famously written by an impresario and star of the Viennese popular theatre. It's a magical genre-piece. Mozart dignifies it with music of extraordinary spirituality, but it has to be straightforwardly done, funny and beautiful. Leave it alone. Don’t over-think it.

But one problem with The Magic Flute is the enormous gulf between what the music seems to be saying and the very narrow idea of brotherhood that the entire text is at pains to promote. Its sexual politics are beyond redemption: the only tolerable thing about the brotherhood is the music they sing. The heroine of the opera, Pamina, is abducted at the brotherhood's behest, imprisoned, subjugated to a lustful moor, then hectored by the brotherhood about her mother in particular and women in general. She's never allowed the kind of music that Mozart lavished on the heroines of his Italian operas. The libretti of Don Giovanni and Così fan Tutte may be as misogynistic as the society that gave birth to them, but you can just about persuade yourself that Mozart writes music that asserts the wit, intelligence and independence of their female protagonists.

The Magic Flute libretto is fine when it's childish, innocent, farcical, or magical. But when it takes itself seriously, which is often, it's a massive embarrassment. In 1988, I was so busy making an enchanting show that I ignored or sanitised everything that is offensive about it. I thought it was enough to turn the lustful moor into a lustful white guy; and in what I thought was a radical intervention, I allowed the heroine, rather than the hero, to play the magic flute. In retrospect, it all seems rather feeble.

You can take The Magic Flute at face value, in which case you have to take it warts and all, racism and misogyny intact. You can distance the production from the libretto, make theatre out of a critical commentary on it and find a contemporary framework that acknowledges the gulf between the music and the mediocrity of a text that reflects a vanished way of thought. Or you can substantially rewrite the text and reconceive the action to bring it closer in spirit to the score.

Three possible solutions, all of them legitimate, all of them potentially exciting, all of them implied by the discussions of the amateur actors in A Midsummer Night's Dream.

Of course, the most straightforward option is always to take the work at face value, warts and all. You can assume that the audience understands its original frame of reference. Often, that doesn't present a problem. The audience gets it. Bad idea to divide the kingdom into three parts. Bad idea to kill the king.

Just as often, the audience will recoil from a play or an opera's underlying assumptions. Bad idea to disobey your husband. Bad idea to do business with a Jewish moneylender. Good idea to dump your boyfriend because his father is worried that you'll bring shame on the family.

So, the director, knowing that the audience will otherwise detach itself from the play, can legitimately provide a framework for the production that acknowledges the philosophical gulf between then and now. A production that might even invite the audience to consider what was once right about the play but now seems wrong. You don't really have a choice if you do The Taming of the Shrew or The Merchant of Venice as written.

Which often leads the director to consider the option of adapting or even rewriting a play in order to recapture some of its original spirit.

But I didn't have the nerve to embrace any of these three options when I did The Magic Flute, and in any event, music is not so easily interfered with as text. Nor would you want to interfere with the music of The Magic Flute, but seriously - how sacred should the libretto be? Bottom's colleagues couldn't be happier to mess with their text. How sacred is the text of any old piece?

Of course, it helps if you're confident you know what the original text is. It's not necessarily what survives on the page. How many old plays, or operas for that matter, exist only in one edition? If we know for sure what happened at the first performance, are we sure that the first performance worked as well as it worked once it had been properly road tested in front of an audience? I have literally never directed a new play that didn't change - often considerably - during its run. Even the greatest and most experienced playwrights adapt their work when they see what happens to it when it's lifted off the page and embodied in flesh and blood.

It occurs to me that as a theatre director, I can talk about text and be confident that you know what I mean by it. I can't talk confidently about the text of a classical ballet, because I don’t know exactly where it resides. In the score? The libretto? The original choreography? How certain can we be that any of them have been passed down in exactly the form that their creators intended? And how certain were those creators that they ever achieved the perfect version of what they were trying to do? Many of you have spent years thinking about exactly these questions, and your answers will be much better than mine.

And I've been exhilarated by some of the answers I've seen on the dance stage. Matthew Bourne has made a dazzling series of ballets by working on the premise that the libretto and choreography of the great classics is entirely his business. Akram Khan's Giselle has a brand-new score, too. It was no less Giselle when Tamara Rojo, with an impresario's flamboyance, programmed it alongside the traditional Giselle.

Maybe what keeps classical dance honest, and I hope keeps the spoken theatre honest too, is the continuing vitality of the contemporary repertoire. In the spoken theatre nothing attracts a bigger audience than the newest hit play. The popular theatre repertoire has never stopped expanding. Neither has the ballet repertoire. New ballets feature heavily in the programming of all the major British ballet companies.

But the creation of a popular opera repertoire slowed down when classical music parted company from the wider public early in the twentieth century. The Royal Opera’s tenacity in the presentation of new opera has been rewarded by some sensational successes - by Thomas Adès, George Benjamin and Mark-Anthony Turnage among others; ENO has been equally persistent. But opera houses world-wide are largely dependent on often-performed classics. So, innovation has been out-sourced from composers and writers to directors. Opera companies compete with each other for the most daring Rigoletto or Tosca.

Some American companies cling grimly to tradition and play safe to a dwindling audience of ageing patrons who want their Rigoletto embalmed. But in the pursuit of novelty, it's too easy for an opera director to succumb to the temptation of playing to a small coterie of other opera directors: it becomes a badge of honour whose Rigoletto was booed loudest. The British public is slow to boo, but it feels left out if it doesn't understand what you’re trying to do with your Rigoletto. It will go anywhere with you if you let it in on the secret, but it dislikes being made to feel stupid.

I don't think singers like it either, but they put up with it. If directors can't persuade actors in the spoken theatre of the validity of their ideas, if the actors can't find a way to embody the director's ideas, they'll grind to a halt. The most banal truth about an opera is that the music never stops; and unless the conductor puts down the baton, the orchestra will keep playing. So, it doesn't matter what a director throws at the stage: the performance still keeps going. The music powers on whether there are too many ideas, or none at all, or whether the ideas are good or bad.

It's an interesting paradox that as the staging of opera has become increasingly unmoored from the stage conventions that used to govern it, musicians have become ever more obsessed with musical authenticity. The default position is to perform with absolute fidelity every single note the composer wrote, even when the composer took out his own blue pencil and cut his own opera. The result is that very different answers to Bottom's questions can co-exist, often very effectively, in the same show. The director can leave the killing out, the conductor can put it straight back in. Or more likely, the director can put a great deal more killing in than anyone ever envisaged, the bloodier the better.

Returning to Shakespeare, I've become more and more convinced over the years, that to be faithful to the spirit of the plays it is often a good idea not to be too fanatical about the letter of them. Certainly, it's rarely a good idea to play the whole text, exactly as published. It's unlikely that the playwright ever proof-read even the Quarto editions of the plays that were published during his life time. He wrote for a specific company of actors, and I'm sure they questioned his text in rehearsal, and altered it, every bit as much as Bottom's company does.

We can be certain that they added stuff of their own. This is what Hamlet says, in his famous advice to the players who come to Elsinore to put on a play.

"Let those that play your clowns speak no more than is set down for them. For there be some of them that will themselves laugh to set on some quantity of barren spectators to laugh too, though in the meantime some necessary question of the play be then to be considered. That's villainous."

I admit this gave me pause when I found myself allowing some of the actors in A Midsummer Night's Dream to ad lib during rehearsals. It sounds like the playwright is warning us across the centuries not to mess with his text.

But of course, it isn't Shakespeare speaking these lines. It's Hamlet, a fictional prince. Shakespeare would surely have identified with the pissed off actors whose own business Hamlet is trying to explain to them. He's the wealthy patron, he's paying the artists, so he gets to tell them how to make art. Who's to know whether he's right? Maybe Shakespeare, as a share-holder in his company, was delighted when the clowns perked up somebody else's boring play. Or even, his own boring play.

In any event, we can tell what Shakespeare's own plays were like when they weren't subjected to the rigour of rehearsal and performance. There is no record of Timon of Athens ever reaching the stage during Shakespeare's life. It's an unfinished first draft, full of inconsistencies. The plot doesn't make sense. If it arrived on my desk from a living playwright, I'd send it straight back with a frosty suggestion that it needed more work.

In the absence of a living playwright, when I staged it at the National, I did the work myself, cutting large chunks, adding others to make more sense of what was left. I granted myself some of the translator's license. Many productions of Shakespeare in translation appear to be freer and more radical than our own. In another language, you can bend the text to your will, cut it and rewrite it when it contradicts the story you want to tell, or - in the case of Timon of Athens - the story the playwright tried but failed to tell.

Nobody noticed what I'd done, or if they did, they didn't care. They didn't care when Laurence Olivier messed with the text Henry V for his great 1944 film. And as a matter of fact, he did considerably more violence to it than I did to it in 2003.

The play begins with a Council meeting. The King needs rock-solid legal justification for the invasion of France. He's in icy control, cueing the Archbishop of Canterbury to give him the right answer to the question that matters: "May I with right and conscience make this claim?" The Archbishop obliges him, at tortuous length, with an analysis of the ancient law governing succession to the French throne, and the King is as gratified as Tony Blair must have been when the Attorney General delivered to him the notorious justification in international law for the invasion of Iraq.

At the National, the Archbishop handed copies of an elaborately produced dossier around the cabinet table, and he referred to it repeatedly as he explained England's right to take military action. We played his long speech virtually uncut. We were true to the text, but the audience knew about dodgy dossiers and they quickly realised how interested Shakespeare was in how leaders manipulate truth to justify going to war.

When Olivier made his film, who was interested in justifying the war? The cause spoke for itself, and Henry V has never been more necessary than it was when it raised the nation's morale six months after D Day. So, Olivier cut the Archbishop to the bone, and mocked what was left. In fact, Olivier ruthlessly cut the text to remove everything that shows the King in a poor light, and everything that insists that war can be bloody, treacherous and ignoble. That's a hell of a lot of the play, which is far more ambiguous about war and leadership than you'd guess from watching Olivier's film.

But I think Olivier was right. We change all the time. We changed entirely between 1944 and 2003. A nation fighting for its survival is a different nation to one racked with doubt about a foreign adventure. Sometimes, the changes in us seem like reason enough to change the classics.

Did Shakespeare really want the audience to withdraw in disgust from A Midsummer Night's Dream when Oberon, wanting to humiliate his wife Titania, drugs her and arranges for a donkey to have sex with her? How funny is that these days? What kind of enchantment can survive taking it at face value?

I've just opened the Dream at the Bridge Theatre. When I started to think about it, those were among the questions that I asked myself. The changes in us, including our much greater alertness to the ways in which an arrogant patriarchy subjugates and humiliates women, make the treatment of Titania ugly at best, criminal whichever way you look at it. It's hard to ask an audience to be enchanted by the lengths to which the King of the Fairies goes to bring the Queen to heel.

I can't imagine any of these questions bothered Frederick Ashton when he made his ballet in 1964. In the very best sense, Ashton's Dream seems to me to be frozen in time. His response to both Shakespeare and Mendelssohn is so complete that it would be hard to unpick any element of it, including its set design, which would seem hopelessly dated if you put a production of the play in front of it these days. In fact, many of the productions of the play that I've seen recently have been legitimately dark, even sour.

The trouble is that I don't think that it is a dark play. I don't think The Magic Flute is a dark opera. I think they're both life-affirming, celebratory, joyful - like Ashton's Dream. So, I did to Shakespeare's Dream what I didn't have the courage to do to Mozart's Flute. I rewrote it. Actually, I altered very few of the lines; but I reassigned almost all of Titania's lines to Oberon, and almost all of his to Titania.

I also adopted the more conventional doubling of Oberon with Theseus, Duke of Athens; and Titania with Hippolyta, Queen of the Amazons. You may remember that at the start of the play, Theseus forces Hippolyta to marry him. And then he tells a young Athenian woman, Hermia, that if she doesn’t marry the man her father has arranged for her, she must either spend her life in a nunnery or be executed.

The rest of the play, the dream in the enchanted forest, is often played as a kind of unconscious working out of the traumas of the first act. How much better, I thought, and how much truer to the spirit of the play, if in the dream the Queen took charge, the Queen drugged the King, the Queen tricked the King into a night of passion with the transformed Bottom, the Queen showed the King that women's bodies can't be disposed of according to patriarchal decree, that love and desire can't be controlled by law.

That's what I did, and it seems to me to work just fine. Even people who know the play don't really notice what we're up to until the moment when Oberon awakes and falls in love with Bottom. And to my great delight, they're thrilled that two big, domineering men go crazy for each other. And when Theseus, the morning after the dream, gets a second chance to dispose of Hermia, he's learned his lesson, so he lets her marry the man she wants to marry.

I think that the rewrite restores something true about the play that can easily be lost if you play it exactly as written. Though I look forward to seeing the play as written again soon. The text survives. There's always next time. There's no right way to do it.

At the end of the performance of Pyramus and Thisbe, Bottom offers an epilogue. "No epilogue, I pray you," says Theseus, who has been through a lot, particularly at the Bridge, "for your play needs no excuse." Despite the fact that Pyramus and Thisbe has in fact been a disaster, he's the kind of patron that all artists yearn for. He assures the actors that he's just seen a fine tragedy, "very notably discharged." I will follow his advice. No epilogue. And I hope you will be as kind about this performance as Theseus is about Bottom's.