Given Nagano's extensive discography, as well as repertoire both on the concert platform and in the opera house, he has covered a number of problematic works which have been left unfinished and completed (or otherwise) by someone else.  Why this particular interest?

I'm not actually that interested in it! I end up recording a lot of them because people ask me to record them; but I ask them the same questions as about all repertoire.  The score sits on my desk for a long, long time; sometimes it takes years of study before it finally makes its way onto the podium for performance.  It is important to research as thoroughly as you are able to at that particular time and, with the great repertoire we have, one of the common characteristics is that the depth and substance of these works are so vast that you can spend a lifetime studying and researching.  As your own understanding grows so does your ability to then invest sharper and more provocative research skills into your interpretations.

Whether problematic pieces or standards of the repertoire, it is important to try to look at the evolution of a piece.  If the work has been through a number of versions - the Bruckner Symphonies for example - it is important to find some clues as to the composer's intent; to decide whether the final decision made by the composer was truly in keeping with the composer's intentions, from your point of view.  There may have been exterior forces at works - friends, critics, enemies - you never know, we're all human beings, and that includes composers as well, and one can never actually be totally comfortable standing in somebody else's shoes. So one should never have the slightest pretence to say what is right or wrong, because we can't. 

We should certainly examine works and have a feeling of what a composer's intentions might have been.  In the case of the original version of Strauss and Hofmannsthal's Ariadne auf Naxos there is both the Molière play and the revolutionary structure of the incidental music, which make for a very, very long play lasting nearly three-and-a-half hours.  Somehow very naturally it organically transforms its  structure into a stage opera.  This was so intriguing and was so revolutionary for Strauss, who is normally - unrightfully so - considered to be a composer who looked to the past for ideas instead of towards the future, that I felt that it was certainly important to examine the original «wild' version, as I call it.  Then, when the Strauss Institute in Munich visited me and asked me to give a performance, I was finally convinced in discussion with them that it was an important thing to do.

Both versions of Ariadne auf Naxos are in my repertoire, and I would never suggest that the first version should supplant or supersede the final version.  The final version was the version that Strauss preferred and this is important.  It is brilliant, but it is more of a standard operatic structure.

I perform Billy Budd in both versions.  The final version was the one that Britten left and sanctioned, but the first, four act version again seems to me - after much debate, conversation and research - to solve a number of theatrical problems, which are left as loose ends in the final version.  In the original version - which essentially restores a number of cuts and redivides the opera into four acts - there is a very natural and organic focus to the story, which maintains a coherency of character development.  In the second version both the story and character development are arrested at certain points, making it more challenging and less organic to follow.

With regards to Busoni's Doktor Faust, on stage we used the ending by Busoni's student, Phillip Jarnach.  We considered Anthony Beaumont's ending for a while, but finally the stage director Peter Musbach felt that for his stage concept the earlier ending made more sense.  However, as we had performed the Beaumont ending in concert and had both «endings' in our repertoire, we recorded them both, so listeners could come to their own conclusions. I think Beaumont's ending reflects some very important research he had done and shouldn't be at all overlooked or dismissed; so it was important that we recorded it so that people could hear it as a viable ending.  But the other ending is closer in time to Busoni (although you can hear the style change, as is to be expected I suppose), and it gives quite a bit of a dramatic and theatrical crescendo throughout the evening, which is why also chose it for our stage production in Lyon.

It's very important never to allow yourself to think you know the answer, because this does not serve the composer nor the music.  It depends really upon what your research tells you is the composer's intention.  For example I do Bruckner's Seventh Symphony without the cymbal crash in the slow movement.  Sometimes it's very contradictory.  I don't perform Mahler's Tenth Symphony - I don't somehow believe in it, it doesn't sound like Mahler - in any of the completions.  To me, it just doesn't feel right, since the piece isn't Mahler.  Likewise Elgar's Third Symphony: I've looked carefully at the score, but it's doubtful that I'll ever conduct it - it remains dangerously controversial.

That being said, I did conduct (and record) Debussy's incomplete operatic attempt Rodrique et Chimène which Pierre Boulez told me he had some serious problems with, which of course I accepted.  Everything was a bit different with this.  I had seen the sketches from Durand and the material from Rodrique et Chimène is brilliant writing.  It also it offered clues or suggestions to one of the big questions that I and many people have about Debussy's compositional development: what happened from the time of both Prélude ð L'Après-midi d'un faune and L'Enfant prodique and the composition of Pelléas et Mélisande?  Something enormous happened!  Pelléas et Mélisande is one of the greatest masterpieces of all time.  The earlier works are beautifully refined music, but you can't really say they are the greatest masterpieces.  What happened?

The clues as to what happened are contained in those pages of Rodrique et Chimène.  I was lucky enough to see the sketches (unless you had access to them, you could only read about something called Rodrique et Chimène in a biography). Whilst fully acknowledging the controversy and fully accepting the criticism that it could not be called 100% Debussy, what is available in terms of an insight as to Debussy the artist and Debussy's language as a composer is invaluable.  So I conducted and recorded Edison Denisov's orchestration of Richard Langham Smith's reconstruction, so others can hear it.