Watching Smash, and thinking about untold stories about the making of a musical, reminded me that I recently enjoyed reading William Goldman’s book The Season: A Candid Look at Broadway. He saw and wrote about every Broadway show in the 1967-8 season, incorporating his own opinions about what was wrong with Broadway as well as interviews with many insiders (some anonymous, some not, all with axes to grind). I recommend the book both as a snapshot of the attitudes of the time and as a repository of Broadway gossip. It includes at least two fascinating stories of how highly anticipated musicals, The Happy Time and Golden Rainbow (hopefully you haven’t heard of them; they weren’t very successful) arrived on Broadway with their original stories distorted completely beyond recognition by the “Muscle” of the production, the person with the power to shape the show: a powerful director-choreographer in the first case, a powerful star in the other case.
But I thought this would be as good a place as any to present Goldman’s list of reasons why a song might bomb in front of an audience. Goldman, who had already written a Broadway musical with his brother James, was trying to explain why it’s so difficult to write a successful musical, and one reason is that a song can fail for any of the following reasons – which then necessitates that you throw out whatever you thought the structure of the show was supposed to be, and start fixing everything as fast as you can.
1. It doesn’t work because the goddam soprano can’t sing it. That’s one of the commonest reasons put forward by songwriters. The material and the performer are simply not in harmony. Sometimes, for example, a performer can have difficulty phrasing a song properly
, or the range is too great, and this brings on discomfort in rehearsal, which often turns to panic by the time the show is out of town. The performer is simply unhappy doing the song, and that’s it.
2. Sometimes the director can’t direct it.
3. Sometimes the choreographer can’t put a “button” on it, a button being the closing moment that leads the audience happily into applause. Stephen Sondheim, lyricist for West Side Story and Gypsy, thinks this is the commonest reason for a number not working: lack of a button.
4. Sometimes songs don’t fit the moment.
5. Sometimes the song tells us something we already know and don’t want to hear again.
6. Sometimes a song can be in the wrong scene.
7. Sometimes a song can be in the right scene, but the wrong part of the right scene.
8. Sometimes an audience will disagree with what comes out of the character’s mouth and simply refuse to believe what he’s saying.
9. Sometimes the audience doesn’t understand what leads up to the song — the reason for someone’s singing.
10. Sometimes a song falls between two stools; the music is one thing, the lyrics another, and the effect becomes diluted, watery.
11. Sometimes an audience will not want to spend the amount of time required for a character to sing a song, especially if the character is an unimportant one. (This brings up the whole question of musical time, which will not be gone into except to say that musical time is compressed; three minutes of a song might be equivalent to ten or fifteen minutes of spoken dialogue.)
12. Sometimes a song can simply be wrong: wrong for the moment, wrong for the character, wrong for the show itself.
13. Sometimes a song can be programmed badly: a slow song following too closely upon another slow song. Two funny numbers back to back can sometimes damage one or both.
14. Sometimes a song can rip the fabric of a play. (Sometimes this happens and it works, as with “Ol’ Man River.” It’s a specialty, a turn. Like a bar in the back of an automobile: it doesn’t make the motor go, but it helps make the trip enjoyable.
15. Sometimes a song can be a bad song. But this doesn’t happen as often as one might think on Broadway, because most of the songwriters entrusted with half a million have a certain minimum proficiency. Besides, don’t ask me to define what a bad song is; badness is more often a matter of personal bias than aesthetic judgment. In musicals, it is probably safe to say that songs often aren’t as bad as they are wrong…
…All right, hang in there, because now we get to be creative. We hide out in our hotel room and write the [replacement] song. That can take a couple of days. And it can take a week or two before we ever get to see it done on stage with an audience, because it has to be approved and choreographed and learned and orchestrated and rehearsed. But at last it goes in and it’s an improvement, so we’re done, right?
Wrong. Because this strange thing happens: with the substitution of the new number, some of what preceded it – at the very least the song lead-in – is different, and everything that follows it is in some way affected. So another number or scene that was working perfectly may suddenly, crankily, stop working. Or something following the old (bad) number, which seemed adequate, may, in the light of the new (good) number, turn rotten. So we’ve got to change that. And when you change that, everything else changes. Out of town, it’s a constant wild race: changing, trying like hell to fix and patch and forget what you said the show was about back before rehearsal: under the gun you go with what works. And slowly, without anyone knowing it at first, the whole giant structure begins to change direction.
This is from a chapter dealing with a 1967 musical called Henry, Sweet Henry, which looked like a hit out of town, and by the time it came into Broadway, the only things that were really stopping the show were the numbers for a 20 year-old supporting player named Alice Playten, who nobody had expected to be popular. Anything in show business is unpredictable, but nothing’s quite so unpredictable as the making of a musical, because it’s so hard to know in advance what numbers will work. And that’s probably as good a response as any to my doubts that the Marilyn numbers on Smash could be a successful musical: this is a real song from a real musical, and it isn’t so great, and it stopped the show.