There's a playwriting exercise that illustrates the difference between story and plot. You take two plays with very particular plotting strategies - say, Bertolt Brecht's Galileo and Henrik Ibsen's Ghosts - and you imagine how the other playwright would have written them. So, Brecht's Ghosts would take us, scene by scene, through Mrs Alving's attempt to leave her drunken husband, her rejection by Pastor Manders, her lying letters to her absent son about his father's virtues, and the son's discovery of his inherited syphilis, before returning home to expose all of the above. Ibsen's Galileo would all be set in the closing scene in which the aged Galileo, under virtual house arrest by the Inquisition, confesses to his former disciple that, by recanting his beliefs, he has betrayed his calling.
The point of the exercise is to show how structure implies not only tone (the longer the scenes, the greater the emotional intensity), but also meaning. In Ibsen, the protagonists' crimes and betrayals have occurred before the play begins, and are thus unalterable. Brecht shows events as they happen, when the protagonist still has the option of acting differently.
These considerations were in my mind when the director Trevor Nunn and I debated how to dramatise Gitta Sereny's Albert Speer: his battle with truth. Speer's story was first told during his evidence at the 1946-47 Nuremburg trial; he gave his own version in his hugely successful memoirs (Inside the Third Reich), which were published after his release from 20 years of imprisonment. His version was answered by the American writer Matthias Schmidt, in his 1982 book Albert Speer: the end of a myth; and a further attack on Speer's view of himself was mounted by Dan van der Vat in his 1997 book The Good Nazi.
Between these two books came Sereny's 1995 biography, which painted a complex picture of a highly ambiguous man. The book has a thesis about Speer; but as it summarises the interviews on which it is based, it also contains alternative views. And because Sereny successfully confronted Speer with aspects of his own past that he had hitherto denied, the making of the book itself became part of Speer's story.
In our initial discussions, I considered a number of alternative dramatic strategies. The German writer Esther Vilar had already done one highly pared-down version (her play Speer, performed at the Almeida last year, was a two-hander set in postwar East Germany). There were various other small-scale options: Speer was interrogated by a very starry selection of American intelligence officers (including Paul Nitze and J K Galbraith) which could have formed an excellent frame for a discussion of his wartime career; in prison, Speer built a garden in which he conducted an imaginary walk round the world, and I could certainly have built a play out of Speer's relationship with his guards, his fellow prisoners and his fantasy life. And there was the rich drama of the Nuremburg trial itself.
Well, Speer at Nuremburg had already been done brilliantly by Richard Norton-Taylor at the Tricycle Theatre. And while the interrogation or prison-play formats could frame flashbacks to Speer's past, they couldn't give me his life after prison, when he refashioned himself, reforged his career and was forced to confront his own evasions. And we were talking about a big-cast play suitable for the vast Lyttelton stage.
So it was going to be an epic play, moving from the Nuremburg rallies to the Nuremburg trial, from Berlin to Paris to Ukraine to postwar Germany, much closer in form to Shaw and Brecht than to Chekhov or Ibsen.
But the problem with epic plays is that they tend to foreground a single authorial voice: with Shaw and Brecht in particular, the author's message is often contained in the choice of scene (in St Joan, why has Shaw put us into a tent with three men?; in Galileo, why has Brecht suddenly pitched us into a carnival?).
In this play, I wanted to dramatise the contest between Speer's own telling of his story and that of his critics. In addition, I wanted to find a special dramatic device to express Speer's inner struggle, as drawn out by Sereny.
Accordingly, the first act shows Speer constructing his own version of his life, flashing back from conversations with a Calvinist pastor in Spandau to his career as a Nazi, a story dominated by his relationship with Hitler. In the second act, other witnesses come forward as we follow Speer through his post-prison life. Surrounding all of these frames is the confrontation between Speer and an imagined interrogator in the audience itself. So the play is consciously broken-backed, as was Speer's life: the first act consisting of a linear narrative told by one man to another, the second of a series of scenes, directly presented, but complemented by comment from other characters.
There was one final problem. The years after publication of Speer's first book can be seen as a struggle with the big and small lies he had told in it. However, this process was unexpectedly and (from a dramatic point of view) inconveniently interrupted by Speer's falling in love with a young woman who essentially gave him permission to abandon the struggle with his conscience. Maybe Speer would have returned to the fight, but in fact he died in her arms.
In this sense, the real story was unfinished. I wanted to find a device that would show what Speer's full confession might have consisted of and might have meant. So I wrote a nightmare - occurring, proverbially, at the very moment of death - in which Speer is forced to confront not only what he knew but also the flaw in him that allowed him to deny it. The last lines of the play - criticised by Nina Raine in last week's New Statesman for being historically inaccurate - are, in fact, reasonably close to what Speer agreed in conversation with Sereny. But, as is obvious in the play on stage, they are at the climax of a scene that occurs inside Speer's imagination.
The last years have seen a move back to highly contained, psychologically intense dramas, played with small casts in one location and often in real time. One of the best, Michael Frayn's Copenhagen, is also multidimensional, as its three characters posit many different readings of a 15-minute conversation.
I wanted to find a way of telling a panoramic story panoramically, breaking free of real-time, single-set constraints, but maintaining the consciousness of the various points of view from which the story is told. Albert Speer is the result.
Albert Speer by David Edgar, based on Gitta Sereny's Albert Speer: his battle with truth, runs in repertoire at the National Theatre, London, until 8 July (020-7452 3000)