In fact, when the original idea package for this novel arrived on my mental doorstep, the main character was already a mute boy. I didn't know the boy's name, why he was mute, whether he had been mute all his life or whether he would always be mute. All I knew was that his inability to speak was going to be essential to the story, which would (I guessed) mainly treat it as a fact of his life. There were a couple of other items in that idea package, including a 90-acre farm I knew very well, because I had grown up on it, and dogs—lots of dogs—dogs with some quality worth fighting for. Because of his speech deficit, I suspected that the main character (who was going to be smart, a prodigy really) would find a way to channel his natural desire for speech into a gift for working with the dogs. I don't how ideas arrive for other writers, but that's what I was handed. All that, plus a very old, very well-known storyline that seemed to me to have unfairly lost its power to move people.
So where did those ideas come from? I only have guesses. I had long been interested in language, both human and artificial, as a phenomenon. Much of my career in software research and development has been spent creating programs that analyze human language for various purposes, and I'd ended up working alongside anthropologists, linguists, lexicographers and psychologists, picking up tidbits and anecdotes, learning to see the world the way they see it. If you eat lunch with a bunch of linguists, you'll soon notice them stopping with their forks half-raised, cocking their heads at some offhand expression of yours, exchanging meaningful linguist-glances and muttering, "Can you get that? I can get that." Meaning, they can see how what you just blurted out might be a grammatical utterance. This can be unnerving, though interesting, especially if you are willing to ask dumb questions.
By hanging around with people like that, I was exposed to the scientific literature describing language—how it works, and also how it sometimes fails to work. This in turn led me to some popular books on language: Steven Pinker's The Language Instinct, for example, and Oliver Sacks' book Seeing Voices, a wonderfully thoughtful history of sign language. In typical Sacks fashion, Seeing Voices includes tangential explorations of many topics, including the story of Victor, the "wild boy of Avignon," one of the most famous cases of a child growing up in the wild and coming late to human language. I saw connections to my literary interests: as a lifelong reader and re-reader of Kipling's Jungle Book, I was fascinated by the idea of a character who can talk with wolves. Finally, perpetually fascinated as I am by dogs, I was struck all over again by the uncannily communicative fabric of our relationship with dogs—how we "read" each other so well, without the benefit (or curse) of language. Somewhere along the way, I discovered a book by Vicki Hearne entitled Adam's Task, which took as one of its main arguments that training a dog meant undertaking a morally loaded and highly linguistic joint activity. For example, Hearne talks at length about the "syntax" of the communication between a rider and her horse, and the thoughtful effort required for a dog and a person to come to agreement about the meaning of the simple word, "fetch."
So all those ideas were already in the mix. Life experience had some role as well. I once had some minor oral surgery. Nothing very serious, but for about a week I was left with some stitches that made it hard to talk intelligibly. Rather than continually explain what was going on, I took it as a challenge to see how much of my life I could transact without talking. The experience made a huge impression on me. Muteness, I discovered, exempts you from a certain social overhead. It frees you up to watch. And so as soon as I knew Edgar was mute, I also knew he had a compensating gift for intense observation.
As it turns out, I wasn't especially interested in imagining or diagnosing a specific cause for his muteness—I simply wanted to remove one aspect of language from his character, to make him inward, self-reliant and perceptive. In the end, Edgar's condition is more literary than medical. At the time, the writing felt like ignorant, blind groping. It never progressed as rationally as the above paragraphs might suggest. Only in retrospect, with the finished story in hand, do certain ingredients seem clear. What drove the writing on a day-to-day basis was simple fascination with the events in Edgar's world and a desire to know how they would play out.