I was living with my wife, Kristine (née Allen), in Orem, Utah, when I made the first breakthrough in creating this book. The two ideas were still quite separate, and the speaker-for-the-dead idea was still in a very primitive form. In fact, I had decided that the funeral "oration" should be in song--that it should be a "singer of death." I suppose I thought of this because I had sung at a few funerals and found it a moving experience even when I didn't know the deceased. But when I mentioned this singer-of-death idea to Kristine, she winced. "You've already written 'Unaccompanied Sonata' and Songmaster."she reminded me. "They were both about music. If you do another music story people will think that's all you can do." I realized that she was even more right than she knew! It happened that "Unaccompanied Sonata" and the original short story, "Mikal's Songbird," on which Songmaster had been based were also two of my stories that had been nominated for awards. In fact a novella called "Songhouse," which was really the opening chapters of Songmaster, had also been nominated for a Hugo. The only story of mine which had been nominated for awards and that wasn't about music was the novelet version of "Ender's Game"! So Kristine had inadvertently caught me in the unconscious process of imitating my own past successes. I knew she was right--the music motif may have won me some favorable attention, but it was time to set aside that crutch and do something else.

So it would be a speaker of death in my story, not a singer. That felt right. But here's the silly part. Perhaps I was still unconsciously trying to lean on my most successful previous work, but I immediately wondered, What if the Speaker of Death was Ender Wiggin? It was obvious to me what I was doing--if I can't do the music thing, I can still bring the kid-who-saves-the-world back for another round! And yet the idea appealed to me. I didn't trust it yet, but it appealed to me.