I know that there's a specific term for this, but I can't for the life of me remember what it is.

When you propose a theory for a phenomenon, people will try to disprove your idea. If they're successful, then you can either drop your theory all together, or make modifications to account for others' criticisms.

There's a problem when you change your theory too much though, so that it's too... something. So, for example, when somebody would show Freud a behaviour that he didn't have a defence mechanism for, he would just invent a new defence mechanism, which makes his theory of defence mechanisms a... something or other theory.

Does anybody know what I'm trying to get at?