First principles justification of a ``single wave model'' for electrostatic instabilities