Reasons-Responsiveness: One More Look at the Agent-Based and Mechanism-Based Accounts

Abstract

According to Fischer and Ravizza, having alternative possibilities of action is not a necessary condition of moral responsibility. The authors are convinced by the Frankfurt examples and motivated to hold this position due to the unresolved issue of compatibility of determinism and the possibility to act otherwise. Instead, they propose that what we need for holding people morally responsible is that they possess a kind of control called “guidance control”; and this is dependent on the mechanism which leads to action’s being moderately reasonresponsive. The notion of moderate reasons-responsiveness (MRR) is explicated as follows: a type of mechanism M is MRR if and only if (1) in many possible worlds where it operates and S has a sufficient reason to do otherwise, S recognizes that reason as sufficient; (2) S's disposition to recognize different reasons forms an intelligible pattern; (3) in at least one of those worlds, S acts on the recognized sufficient reason. After the authors present their account in their book Responsibility and Control. A Theory of Moral Responsibility (1998), a lot of discussion and criticism emerged. One of the main points of criticism was the vagueness of the notion of mechanism and the lack of identity criteria for different types of mechanisms. This led some philosophers, Carl Ginet and Michael McKenna among them, to adopt an agent-based reasonsresponsiveness theory. This modification faces other problems. First, it has been suggested that it is vulnerable to Frankfurt-like examples, since the agent with whom the counterfactual intervener is connected is not reasons-responsive, although she is morally responsible. This charge is rejected both by Ginet and McKenna. However, the latter admits that there are possible examples, concerned with “intervener” who is internal to the agent, that may threat the agent-based theory. One of his undeveloped examples, constituting the second problem, is the situation in which the mentally ill agent acts while his sickness is not active, but it would activate and prevent him from acting otherwise in alternative scenarios. In this paper, I present examples based on McKenna’s suggestion and argue that they pose a much more serious problem for agentbased account than McKenna would like to admit. I also present a possible counterexample to both agentbased and mechanisms-based account concerning a recovered addict who refuses to take a drug, yet seems to be morally responsible (in this case – praiseworthy) for her resistance and yet is not reasons-responsive in the right way.