Active control, agent-causation and free action

Abstract

Key elements of Randolph Clarke's libertarian account of freedom that requires both agent-causation and non-deterministic event-causation in the production of free action is assessed with an eye toward determining whether agent-causal accounts can accommodate the truth of judgments of moral obligation.

Acknowledgements

I thank Al Mele and Derk Pereboom for their useful comments on early drafts.

Notes

See, for instance, Kane (1996), O'Connor (2000), Ginet (1990), Waddell Ekstrom (2000) and McCann (1998).

Eleonore Stump (1999) argues that libertarians need not be committed to the view that alternative possibilities are required for responsibility (see also Stump 1997). David Hunt expresses a similar view in ‘Freedom, foreknowledge, and Frankfurt’ (2003). But Stump and Hunt claim that libertarianism requires ultimate control that, in turn, requires the falsity of determinism. Linda Zagzebski (2000) defends an analogous position.

Non-causalist accounts are developed and defended by Ginet (1990), Goetz (1988, 1997) and McCann (1998). I do not discuss these views as they have a prima facie difficulty not shared by other libertarian competitors: if free actions are, at their core, uncaused, and hence are not determined to occur by anyone or anything, they seem not to be under anyone's control. Event-causal accounts have been defended or discussed by Mele (1995, Chaps 11, 12), Clarke (2000, 2002), Kane (1996), Dennett (1978) and Fischer (1995).

Mele outlines this sort of libertarianism (see 1995, Chap. 12; 1996, 1999).

Clarke argues for a similar conclusion (1996, 27).

Directly free actions are free independently of the agent's having had control over something that brought about these actions; their being free is not derived from the freedom of any other acts. Derivedly free actions are not directly free.

See Clarke (1996, 39–40). But see, also, the discussion below on the agent's role in action explanation.

Clarke introduces Peg's case in ‘Libertarian views (II)’ (2002, 366).

See O'Connor (2000, 2002) and Ginet (2002). For elucidation of the problem of causal exclusion and Clarke's responses, see Clarke (2003, sec. 8.3).

I owe this criticism to Derk Pereboom.

This argument is developed and defended in detail in Haji (2002).