I got a YouTube comment the other day asking for tips on composing. Unfortunately, YouTube’s wonderful new commenting system didn’t allow me to reply.
I was about to answer privately, but realized others might get something out it, so why not share it with everybody?
To paraphrase the commenter, he could already come up with ideas and riffs, but didn’t know how to take it further and actually write a song or a composition. He also wanted to know if he should learn theory.
I’ll start by saying that theory is not necessary for writing good music, though I most definitely recommend it. When it comes right down to it, composing is more about problem-solving than anything else, and all that theory does is give you a greater toolset to work with when you’re trying to solve a problem.
So how do I go about composing? This is the process I usually go through, though it can vary greatly from song to song. This isn’t some master plan I came up with to write music, it’s just how it’s always happened naturally.
First Step: Inspiration & Collection
This is the part most people can do with no problem. That cool riff you discover while practicing, that melody that pops in your head on your way to work. These are the seeds that you’ll use as a foundation.
I’ve always been pretty inconsistent about remembering ideas, but lately I’ve gotten into the habit of recording everything right away into Evernote. It has really helped me become a lot more productive with my music. In fact, most of the music on Worldbuilding started with an idea or riff in my “Song Ideas” Notebook.
Step 2: Research & Development
This is usually my favorite step, and also the part where knowing theory will help you the most. I just take an idea and try to develop it as far as i can. I’ll try it in different keys, change a few notes around, try to come up with a contrasting part…basically, I jam with it without worrying too much about the final result.
While I’m doing this, I’ll start getting an idea of where the composition is headed. Maybe the original idea sounds like a great intro, and this one variation could be the chorus, etc. I usually come up with the first half of the song fairly quickly. That’s where the hardest part comes in: where to go from there and how to end the song.
Step 3: Putting it All Together
I try to make every composition feel like a story, and every good story has a climax. My thinking process is something like this: “Ok, I have this really cool first half, now something even cooler has to happen, and then we finish.” That something cooler sometimes comes easily, sometimes not so much.
I can’t give any specific advice since it’s always different. Like I said, composing is problem-solving, and every song will have its own solution. It’s up to you to discover it. The important thing for me, is that I always want the music to go somewhere.
Step 4: The Final Details
Once the composition has a form and has all its parts, I focus on the small stuff. Maybe find a way to make the transitions between parts a bit smoother, or maybe that one part in the chorus isn’t sounding quite how I’d like it too. Sometimes this step can take the longest, it happened to me with “You Are My Natural Selection.” I could never get the heavy section in the middle to sound quite right, and the transition was also kind of clunky. It took me months of experimenting to finally get it right.
And that’s my approach more or less. Sometimes I go through this whole process, sometimes I skip a step or two. And sometimes the music just writes itself (but that’s very rare).
The important thing is to keep trying. Even if you feel that what you’re coming up with is no good, the more you compose, the better you’ll get at it. Everybody has to start somewhere.
Other than that, if you’re finding it difficult to write your own music, the best advice I can give is to just listen to a lot of music, but with composer ears. Pay attention to the form of the song, how it builds, where the climax is, how they transition to one part to the next…the more you listen, the more possibilities you’ll have.