in trying to stay "original" when i am working on a new effort, what i listen to all depends on that particular project.
example...
if i am working on my (normal) heavy project... a few weeks before i begin writing, i'll pick about 5-10 cd's by various artists who's sounds/styles i would at least like to attempt to blend together (this past time it was Megadeth, Tourniquet, Dream Theater, Cradle of Filth, and Deliverance). then once i actually begin writing, i will completely switch gears on what i listen to while i am writing (again, this past time i switched to Prince, Enigma, Russ Taff, Marz, and SFC).
This is exactly the kind of thing I mean by twisting and recombining influences to make something that sounds new. I think that just about everyone who is aiming for something "original" does something like this, which is why I say "originality" strictly does not exist. The process of sole author invention that people think they are referring to when they say "originality" is not, in fact, what is happening. Instead, it's something much vaster and more interesting, a process in which the songwriter is a participant and a focus but not the
source.
It's more like a vast group- or hive-mind process. If you follow me, I'd call it an iterative process, which is to say that you have a complex system, and the output of the system is fed back into the system to produce more output, which is in turn fed back into the system, ad infinitum, as in fractals and chaotic systems in general. This is just another way of saying that we listen to different things that have gone before and put them together in a novel way, and the result, if all goes well, is something really distinctive, new, and unpredictable. And then somebody takes that, and the process goes on.
Look at it this way: None of us would come up with a rock song if we had never heard one. And we would certainly never invent something like "post-rock" without being completely steeped in rock music. So instead of sole inventor and font of all genius, I think the songwriter acts as more of a lens that focuses various elements of the cuture in its own idiosyncratic way, and if he does it right, the result is something interesting (which is, in turn, reused). The genius is only partly in the songwriter. The rest is in the system. To me, only this kind of iterative or fractal system can explain the genius we see in music or in any "artistic" field. The theory of originality, which is to say, the "author" theory, simply can't account for the facts.
There are very specific and cynical reasons why the "author" theory is and has been vigorously promoted by certain people in society, but that's an analysis for another time. I think for now I just want to notice that it's mistaken. As this thread (along with countless similar discussions) shows, the concept tends to baffle those of us who set out to make songs. Once we see what's really going on, I think it ceases to be a source of trouble, and we can get on with our role in the vast machine of creation.
I haven't even touched on language specifically, or the role of the "audience" in
actually creating the song. I mean, what I've written here will already sound like gibberish to a lot of people. I just want to encourage people to let go of anxiety about "originality" and get on with the process of how music is actually created in the real world. It is less like the classical concept of artistry and more like tinkering. So go ahead and tinker is all I'm saying.