The indeterminate macro-social: new traps for old players in HIV / AIDS social research.