You can do this -- and indeed, promoting constants to fields is a common technique in the analysis of field theories. But it's cheating if you say the constant

You can do this -- and indeed, promoting constants to fields is a common technique in the analysis of field theories. But it's cheating if you say the constant is dynamical and then don't say what the dynamics is. In field theory, you have to make additional choices here, and there are a lot of additional parameters that enter. In string theory, you don't have to/get to make a choice at all. The dynamics is uniquely determined. It's not exactly a free lunch, but it's a very unusual feature for a physical theory.