Everything went beautifully with Susan Graham. She’s such a complete artist; her voice is magnificent, and she’s an extraordinarly committed tragic actress with a marvellous feeling for the text. In the end we didn’t know who was doing what… I shared stylistic elements and performance ideas with her, and she on the other hand gave me other ideas, and especially her innate stagecraft. She’s really one of the world’s great Didos, simply because she is exactly that character and has that character’s voice. I cast the rest of the opera around her.

You didn’t base your choices on the vocal aspect alone; you made very personal choices.

Dido is a rather mysterious work. Consider the circumstances of the first peformance, for instance. For a long time people thought the premiere was given by young girls in a school. We now more or less know that the work was given at court, although we don’t really know if girls or older musicians took part. The same is true for the orchestra. Nothing is really certain; we’re not sure if the orchestra was small or large. Since there are so many musicological uncertainties, I let myself be inspired by my Dido when planning the rest. When I knew Susan would sing Dido, I chose the magician accordingly, and opted for a larger orchestra that included woodwinds because that would have been possible at the time. There are some rather obscure indications in the score, like “horrible music” and “chaconne”, and we interpreted them the way we thought they should be done.

In any case, you managed to create a very English atmosphere.

Dido is a very English work, and it was important that Susan be a native English speaker, like most of the cast. That made for a familiarity with the language that would have been very difficult to imitate. Language defines musicality to an enormous extent in early music. The same was true for the choir: I wanted a British group there too, and chose The European Voices for their sound and their typically English homogeneity. At first I thought about putting together a French choir, but reckoned it would be too difficult to get the English colour we needed in time. Working with the orchestra was a lot of work in itself. In any case, I eventually intend to found my own choir.

You and Le Concert d’Astrée recently tackled a couple of great choral works – Handel’s Dixit Dominus and Bach’s Magnificat. Up until recently you spent more time on baroque bel canto, didn’t you?

That’s right. We did some short motets from the French repertoire, which is music I really like, very contemplative, very expressive. The next logical step was to take on these big ‘cathedrals’ of sound, real choral music. Our goal is to perform Handel’s great oratorios one day. My personal development played a big role here, especially in terms of Theodora at Glyndebourne. That score showed me an aspect of Handel I hadn’t worked on before; it’s almost as if another composer wrote it. I love early Handel, the Italian Handel whose music is full of inspiration and amazing light. He was a real charmer who picked up Italian in no time. That was the Handel who gave the premiere of Rinaldo in London…an incredible violin solo here, a staggering harpsichord cadenza there, mad arias for extraordinary singers and so on. But at the end of his life, when he wrote Theodora, he had nothing left to prove and that made for a completely different atmosphere… a mystical ambience, long over-arching lines, and a choir which is almost the main character. The personal psychological concerns of opera seria were far behind him; he was on a completely different spiritual quest. It’s almost as if he were saying good-bye to the world, as Theodora herself does in her aria Fond flatt’ring world in the first act. After this production, which was directed by Peter Sellars and was very moving because it was so contemplative - it included the execution of the two martyrs on stage - choosing Dixit Dominus and the Magnificat was a natural next step for me. But I’ll certainly tackle Handel and Bach’s great choral works one day as well.