Welcome to
      Home    Listen    Site Index    BLOG  

032 < >

How Not to Write a Musical

    Some years ago, I had the misfortune to offer my services to a lady who had written a musical. This had apparently been her hobby for some years, and it would have been a perfectly harmless pastime except that she wanted it to be performed, which meant that there were going to be musicians involved who would have to decode her instructions and somehow bring this voluminous enterprise to life. I seem to recall she was in her eighties, and had spent most of her life writing what she called her 'baby.' She therefore did not seem interesting in hearing any suggestions, most of which were of a purely technical variety on my part, though if I had been honest I would have suggested quite a bit of rewriting as well.  The musical had at least been printed on a popular computer scoring program so it was legible, but there were so many things that made it awkward and nearly impossible to execute (not because it was hard to play but because so much of it did not make sense; the musical equivalent of spelling errors and poor syntax throughout made for very rough sledding).  In a few cases during rehearsals, when trying to deal with several of these problems at once made me ready to spit nails, I tried very politely to explain a few of these concepts to her, but they did not take root for whatever reason, and, since I did not want to crush her ego, I kept quiet about most of it. Eventually, after struggling over the issue, I decided that nothing useful would come from offering to help her "revise" her score, (which I'm sure she didn't think was in any way necessary) but the teacher in me would still like to do whatever is possible to prevent something like this from happening to anyone else who would like to avoid it.

Therefore, consider this the Marley's ghost of compositional advice--please, for the love of Pete, if you are writing something that you want people to perform, consider the following items. They are not here to ruin your inspiration, merely to keep the musicians around you from losing their sanity. Most of them seem like common sense. However, when you are writing a large and complex musical work, there are an astonishing number of things that you can do wrong. Since this lady did practically all of them, I have it in my power to warn you against making the same mistakes.

There are many people who have the quaint idea that a composer is someone who works in solitude on in a mountain villa by an open window, blissfully being bombarded with musical ideas like neutrinos and simply writing them down without any need for second-guessing anything. No technical knowledge is required; anything learned can only get in the way of a good inspiration. These people's patron saint is Mozart, who may have actually come the closest to really composing in this manner, except for the small fact that he had a teacher-composer father who gave young Mozart all kinds of technical instruction from the age of three, and who took Mozart all over Europe, admittedly showing him off before astonished royalty (astonishing royalty is not as hard to do as you might think), but also giving Mozart exposure to the best minds on the continent so that before he was very old he knew a lot of stuff about music. This, and the fact that, while many of his papers were destroyed, there are also plenty of sketches and unfinished compositions that show that Mozart didn't just take dictation from heaven and never change his mind about anything. He learned, he tried, he failed, he assimilated, he thought about how to write music constantly.

Now, the history of music is also filled with narrow-minded pedants who liked to tell composers that their most audacious and ground-breaking ideas simply would not do at all, and like to pour cold water on their best inspirations. But every undergraduate theory student who gets his or her compositional assignment corrected by a teacher is liable to put themselves in such a grandiose position too, so you see how that philosophy can be taken too far.

Basically, anybody who really wants to be a composer ought to get some kind of training; you can always decide later which advice to heed and which to ignore. Just because you know how to do something doesn't mean you have to do it that way, but not knowing in the first place means you are greatly limited; for example to the three chords that this lady was using. If she had known how to use them properly she might have kept us entertained. There is, after all, some charm in simplicity. But then, there is what I call simplicity by choice and simplicity by default. The first kind is some of the most beautiful music in the world. When a great genius in music, who could write a lengthy 6-voice fugue chooses instead to write a simple melody, the effect is often astonishing. But the second kind of simplicity is asking for trouble--when a chord or a turn of phrase is the only one the composer has in his or her imagination, even though it doesn't fit very well. Maybe some of your notes don't get along with the others.

It is not necessary to obtain a license to compose music--most means of expression, artistic or otherwise, are about as unregulated as the food supplements industry. And since you are entitled to "lose 50 lbs fast!" simply by swallowing some tasty pills, you might as well be entitled to command certain sounds to arrange themselves in a certain order without bothering yourself about whether there is a better or worse way to do it.  It is inspiration, after all. Which, when you think about it, is a lot like opinions. You can arrive at them however you like.

There are, in the course of any kind of composition, any number of decisions in which judgment and taste play a role. But most of the suggestions below are of a purely practical nature. It generally does not mess with a novelist's artistic vision if his publisher numbers the pages. Most people consider it helpful to be able to find page 336 when they want to. It is the same in music.

I'll begin with the most technical, detailed explanation and work outward. You can skip it if you are not in the mood for music theory right at the moment.

First of all, listen to your performers. If they are having trouble singing something you have written, it might be because they are poor singers; on the other hand, perhaps what you have written is awkward. If they are forever instinctively changing one note, that may be because the phrase is much smoother that way. Let me submit the following for your consideration:



The first phrase represents how everyone is singing your tune; the second is how it was written. However, the second way contains a few problems. The first is that there is an odd interval called a tri-tone between the second and third note, which is hard to sing. Not impossible--it can even be used effectively, for example in Bernstein's West Side Story when Tony sings of his new found love Maria on these notes:

The tri-tone was once called "the devil in music" which I take as a sign that early music theorists didn't care for it a lot.

Just to show why things like these take a bit of discussion with a person who knows what they are talking about--Your first phrase has the same three notes at the beginning, and the matter turns out very differently. This is because what is really going on here is that we are singing two things at once, known as a "compound line." The first and third notes are leading us up to the final note, the door prize (disclaimer: not an actual musical term), and the second and fourth are as well. Well, sort of. What the second line needs to do, not-obviously-enough, is establish the "dominant" harmony, which is always five notes up from the key of the piece, in this case D. That would make our dominant pitch an A. This dominant harmony functions as a kind of harmonic glue to get us to our "door prize" harmony. You've heard this formula millions of times in your lifetime whether you know it or not. It became, for about 300 years, the harmonic pattern our ears were itching to hear. Three centuries is a pretty long time for a fad, if indeed that is all it was. And we're still not over it.

Interestingly enough, this dominant harmony, in the key of D Major, contains the notes A, C# and E. C# is particularly important because it is the tone that leads the ear the conclude on the note D. But in the first example, the C# doesn't go where it ought to. It is false advertising. It goes in the wrong direction, and it suggests the wrong harmony. This is not a good time to introduce a note belonging to a sub-dominant chord; if you wrote an essay for English class in which you discussed the need for aid to Africa, and in the last paragraph decided to talk about the Cubs' game last night, your teacher would not think this was a stroke of brilliance on your part. True, there was a musical phenomena, called a "Landini cadence" in which the leading tone always gave way to the very same extraneous note we've got going on above, before concluding on the keynote, which is guaranteed to drive the modern ear mad, but that went out of style about 700 years ago.

Then why does it work if it goes down to an A instead of a B? Because the harmony doesn't change. Since it is still part of the dominant harmony, the ear hears two complimentary things going on here, instead of one unfocused, disjointed thing.

Most singers could probably not explain this phenomenon, but, at our rehearsals, I noticed that they all changed the note B to an A instinctively, which shows you how sometimes a technical explanation can be avoided if your ear knows what to do.  What is interesting about the two examples is that, while they both contain tri-tones, the second version makes it obvious, whereas in the first, you don't notice. If I furnished the left hand, you would also notice a nasty disagreement between the harmony and the melody. What is also interesting is that everybody in that room seemed to know what was musically natural except the composer!

Being able to explain why something works in words and concepts helps in ways that, alas, might take some explanation. Some people are quite against any explaining, and if they really can manage the same results purely by instinct they have my blessing. But it is rare that a good composer did not consider that his work involved craftsmanship, and it is just as rare for the people who have no idea how to make music themselves to think that it does.

If the above explanation does not make any sense to you, you may wish to learn some basic music theory. If you cannot honestly hear any difference in the two examples above, you might think about something besides composing, unless you are really innovative and plan to avoid traditional harmony altogether.

It is amazing how one note can make such a difference. Since I have spent so much time dissecting the matter, I'll spare you the hundreds of other examples in this musical, which were considerably more egregious, and could not have been fixed simply be changing one note. When Peter Schaffer, in the movie "Amadeus" has Salieri exclaim admiringly "Change one note [of Mozart's] and the piece is diminished." he isn't just blowing smoke. This is probably the kind of thing the kept Chopin up nights.


There are standard rules that people generally follow when they are writing words. One is being able to spell them properly. There is an agreed upon spelling of most words (maybe two). If you don't feel bound to this you may cause people reading your work to spend more time trying to decode your writing then they would like. There are similar rules in music.


Technically an A-flat and a G-sharp are the same note--at least on a piano (a string player may have other ideas). This gives rise to what we call enharmonic spellings, the choice of whether to refer a the note between the G and the A as A-flat or G-sharp, for example. Just like in English, we can't spell everything however we feel like it, even if it technically works. For instance, even though the "gh" is silent in the word light, that doesn't give you license to spell everything with a silent gh in the middle. There are similar rules regarding how to spell chords and melodies, and they are largely based on the function those notes have in the overall context.  If you are in the key of A-minor, for example, and you choose to write every leading-tone as an A-flat, you will drive your pianist up a wall. He will be expecting that A-flat to lead down to a G natural in the next measure, rather than up to the A (which is what makes something a leading tone--it leads the ear up to the key note--here the note A). Twelve pages of this can make your pianist very cranky. It is not impossible to read music written this way, it just takes a lot more painstaking effort, with no hints as to the music's overall construction.

knot beeing abull 2 speghl propirlee leeds two alll kiends uv airers thatgh wil mac yor pees vairy hrd too reid?

There was a point where the harmony in this musical was so weird I had no idea which notes to assume were misprints because the composer's grasp of standard rules of this sort (including which key signatures to use when) was so sketchy I could not deduce what the outline of the musical thought was. Key-signatures also have standard spellings, and they point out the overall plan and tendencies of the notes to go in various directions; in other words governing the behavior of the notes, priming our ears for what is typically expected and what is a surprise. A surprise is very difficult to carry off unless there is some kind of expectation. These key signatures are part of an overall system and are not as complicated as multiplication tables; elementary school students could learn them all, if music teachers would make the attempt.

The remaining points all deal with practical matters related to performance.

If you are determined to make an effect, you will naturally want to employ a small orchestra. The first thing you can do for their conductor is not to hide them somewhere that it is impossible to see the stage and the singers. Every amateur I've worked with in a musical is convinced that singers do not need to see the conductor, because in movies the music just comes out of nowhere and is perfectly attuned to when the singer opens her mouth. In the theater there is a guy in the pit, hidden away as usual, because nobody wants to imagine that there is any calculated effort involved in the magic of the theater, but visible to the singers and the instrumentalists. He is there because 20 people with their own opinion about when the piece ought to begin is not a great idea. Try getting 20 people to agree on what to have on their pizza and you'll see what I mean. That conductor has to coordinate everything--he literally makes the beat visible to everyone, so that they can see exactly when it is time to begin, where the next beat follows, and how to play a particular passage by whether his motions are fluid or vehement. Therefore, it is helpful if you do not position your orchestra so that it is impossible for the singers to see your conductor. That is not why large pillars were created.

The next thing you can do for your conductor is to give him a score. The "full-score" is a copy of the music in which you can see what all the instruments are playing. It is organized so that all of the instrumental lines are joined together and, though it is written left to right, a single line of music may be as tall as the page itself, if there are enough instruments all playing that line together, each on its own "staff." Since one of the functions of the conductor is to "cue" each instrument when it is time for them to resume playing after a break (for security reasons, partly, since the clarinetist should be able to keep his own place) it would be helpful if the conductor knew when the clarinet was actually supposed to begin playing.

Another thing that is just a bit helpful at rehearsals is to number your measures. If the conductor asks you to do this several times, take him seriously. What he is trying to avoid is the situation wherein the aforementioned clarinetist appears to have come in too early and is obviously in the wrong spot in the music. If this happens, oh, I don't, know, let's say five minutes into a seven minute piece, the options are as follows: go back to the very beginning of the piece and start over, confident in the knowledge that, five minutes later, the same thing will happen, probably because the clarinet part has been written incorrectly (i.e., the wrong number of measures of rest has been given), or, tell everybody to start at measure 310, which is only a few measures before the problem spot occurred, so you can fix the problem in a reasonable amount of time. If you had a full score, with numbered measures, you would know what was supposed to be going on--otherwise you will simply have to rely on your ear and your intuition.

Now, as it happens, most music writing software programs, like Finale, automatically number measures. It seems you would have to intentionally turn them off. This would be an indication of unfathomably poor judgment.

If you are writing for instruments, it is not necessary that they should all be playing all the time. Judicious use of them is a hallmark of good orchestration. This does not mean that the best thing to do is to save the instruments until you have reached the spot where the tempo is about to change radically, and the singers, having arrived at a rapturous climax, are preparing to make the audience's dreary lives worthwhile by hanging out on a high C while those of a less operatic inclination use this time to go out and feed the meter without missing anything. When the tempo is being altered and the opportunity is therefore much greater for a sloppy and glorious train-wreck of instrumental imprecision, by all means, throw every instrument you have into the mix, the better to cover up the singers, and make certain that none but the most stalwart tenor will bother with that high C ever again.  Once the music has resumed its normal, predictable pace, you can let the piano navigate these boring passages on its own. This way, you don't have to write nearly as many notes.



comments powered by Disqus