I’ve just finished listening to a 29-hour audiobook by one of my favorite writers, historian Jill Lepore. Unlike most recordings of books, this one was read by the author herself. She’s not a professional voiceover talent, so it was interesting to note how her delivery varied — I could easily tell when she was engaged, or bored, or worn out. (When she was especially tired, her voice got hoarse and she read at breakneck speed, as if to get it over with.) Ordinarily I would find that inconsistency distracting, but it helped that her speaking voice perfectly matches her authorial voice, so her reading just felt human, as if Lepore was sitting nearby and speaking to me.
That’s not always the case. There are some writers whose speaking is so different from their writing that it’s hard to accept that they both come from the same person. In the 1980s, I was a great admirer of Michael Kinsley’s political writing, and I used to eagerly await each new issue of The New Republic to see what smart, incisive things he had to say. He gradually made the transition to being a TV pundit, and the first time I heard him speak, I was immediately let down by his weak, nasal voice and Michigan twang. I never enjoyed his written work as much after that, because I mentally heard the words in his voice as I was reading.
I have to confess that my own speaking voice is more in the Michael Kinsley category than the Jill Lepore category. My writing style is confident and articulate — or at least I like to think so —but my speaking is the opposite: My voice is thin, often strained, and somewhat doofy, and I tend to mumble and stammer and slur words together. Much of the time, before a sentence is fully out of my mouth, I know that the person I’m talking to is going to say, “What?”
For much of my life, I had a parallel voice, what I called my “narrator voice.” It was a voice that I started to cultivate when I got my first tape recorder at eight years old, and that came into full flower when I reached my teens. In ninth grade, I got out of writing a term paper by volunteering to record a dramatized series of African folktales, complete with sound effects and original music. Throughout high school, I wrote and produced radio-style commercials for the shows we were doing in the drama club, which got played over the school’s PA system during the morning announcements. All of these recordings featured my narrator voice — a credible imitation of a 1940s radio announcer, all rounded vowels and clipped consonants. It was a voice that, in retrospect, was corny even in the 1970s, when people on the radio were beginning to adopt the more laid-back, conversational style that’s standard today. It was a voice that I certainly couldn’t use socially, but defaulted to using onstage, which is presumably why I so often got cast as professors and judges.
I also tended to use that voice when I was singing. I never realized how strange that was until I started taking singing lessons in my 20s, and my teacher — hearing me perform a song that I was then doing in a children’s play — said, “Why are you over-enunciating your words that way?” It was the first time I really became aware of the phoniness of it. Why, indeed?
I don’t think it happened consciously, but from that time on, I gradually shed all vestiges of mannerism in both speaking and singing. I may not like my voice very much, but at least I know it’s authentic. This, surprisingly, has become something of a handicap when I try to sing popular songs. Have you ever noticed that singers of folk and rock music pronounce their words with a sort of pseudo-western or southern accent — the kind of accent where “I’m singing” comes out as “Ah’m singin’ “? Everybody does it, and I don’t think it’s done deliberately; it’s just the way people learn to sing. I don’t talk that way, and I just can’t bring myself to sing that way. When I try, it comes out sounding awkward and unnatural. When I hear someone else do it, It feels as much like an affectation of informality as my narrator voice was an affectation of formality.
Recent Comments