Human agency and impersonal determinants in historical causation: A response to David Lindenfeld