Saturday, March 26, 2005
What do internalists believe anyway?
In an unfinished post that's still languishing in the draft queue, I described moral internalism as follows:
Moral internalism is the metaethical view that moral theory is first and foremost concerned with the project of providing an account of internalist reasons for moral behavior. Internalist reasons, on this understanding, are considerations which are psychologically available to the agent and which are understood by the agent to bear on the question of what action is to be done. Importantly, this means that any consideration which is an internalist reason is such that the consideration has some degree of motivational force for the agent. That is, part of what is meant in saying that the agent understands a reason is that the agent is moved to act by the consideration embodied in the reason.
In articulating this conception of moral internalism, I mean to be saying that moral internalism is committed to the claim that the psychological availability and motivational efficacy of a consideration is a necessary condition of that consideration's being a reason for the agent. For reasons which will become apparent, I will call this view weak internalism.
There is room, it should be clear, for the possibility that a consideration might be psychologically available and motivating in the relevant sense and yet not be a reason. For example, if it were a bit warmer I might have a desire to walk down to Dairy Queen for an ice cream cone, and this desire would have at least some motivational force. In such a case the necessary conditions for having a reason would be met, and yet it is at least plausible to say that meeting these conditions is not sufficient for the ascription of a reason. Or, if this is not persuasive, consider a case which is logically equivalent: if someone offends me, I might have a desire to assault them, perhaps a very strong desire. This desire will have at least some motivational force, and yet it seems quite plausible say that I have no reason to commit an assault.
Some people have a temptation here to say that I do have a reason, but that other reasons for other actions are stronger. I don't deny that there are cases which work in that way, but I mean to be saying something different. While it is possible for someone to deliberate about whether or not to bash a disrespectful interlocutor upside the head, this isn't what always happens. In many cases the urge occurs but is immediately dismissed -- the agent is aware of the desire, but doesn't think that the desire has any bearing on her choice of actions. In such a case it isn't as if the violent desire was trumped by pacifistic or prudential reasons. It simply wasn't in the running.
The availability of such cases indicates that we sometimes enter into motivationally animate psychological states without having reasons. A fortiori, something beyond motivational force and psychological access is required in order to have a reason. Hence, weak internalism does not state a sufficient condition for having a reason.
What, then, would stating such a sufficient condition look like? Consider this passage from Bernard Williams' postscript to his "Internal and External Reasons" paper:
The formulation of the internalist position which I now prefer is: A has reason to F only if there is a sound deliberative route from A's subjective motivational set (which I label "S," as in the original article) to A's F-ing. Whether this is a sufficient condition of A's having a reason to F is a question which I have left aside; the essence of the internalist position is that it is a necessary condition.
In this passage Williams seems to be endorsing weak internalism while remaining agnostic as to what further might be needed. He does, however, provide some indication of how this further condition might be filled out by suggesting that having a reason is tied to the availability of 'a sound deliberative route.' In some other work, and in particular in Williams' reply to an objection posed by John McDowell, Williams has articulated this requirement in a way which some have taken to hint at an endorsement of a stronger form of internalism. In that reply, Williams' central point was that the notion of a sound deliberative route couldn't be explicated by appeal to the notion of an ideal deliberator, but, rather, that soundness depended on the route conforming the the deliberative standards accepted by the agent herself.
It is worth noting that the criticism Williams offers of McDowell requires nothing stronger than weak internalism. The problem with appealing to the ideal deliberator is that the normative standards which such a deliberator employs cannot be part of a non-ideal agent's subjective motivational set. As such, those normative standards fail to satisfy the necessary condition posited by weak internalism.
That said, interpreters like Evan Tiffany take Williams to be saying something stronger. They take him to be denying that rational criticism can ever be licensed by an external standard. On this view, the problem with the ideal deliberator lies in the fact that the standards such a deliberator employs are independent of the particular agent's moral psychology. Put another way, the claim is that the necessary condition lying at the center of the internalist position means that the sufficiency condition, whatever it comes to, cannot appeal to standards that are external to the agent's subjective motivational set. Call this position strong internalism.
To see clearly what is at stake between weak and strong internalism, it may help to consider a non-ideal case. So, we might imagine a community of agents who share a number of practices, one of which they call science. Science, broadly construed, is a practice concerned with discovering predictive hypotheses about the external world. This process of discovery, it has been discovered, is greatly facilitated if certain inferential standards are followed. For example, progress in science is furthered by adherance to the rule that correllation does not imply causation.
Now, suppose that there is a member of this community, we'll call him Vik, who claims (a) that he is engaged in the practice of science; and (b) that correllation implies causation. Suppose further that Vik asserts a predictive hypothesis H which, Vik further asserts, is licensed by various correllations he has noticed. Lastly, suppose that some other member of the community objects and says to Vik, "Science gives you no reason to believe H."
The dispute between Vik and his interlocutor has to do with which inferences are legitimate for a person engaged in the practice of science. The interlocultor appeals to norms which are publicly accepted as belonging to the practice and says to Vik that since he has violated those norms he doesn't have the reasons that he takes himself to have. In making this objection, the interlocutor denies that Vik's own commitment to b is sufficient grounds for Vik to assert that he has a reason to accept H.
If weak internalism is accepted, then this analysis of the exchange might possibly be right. That is, it might be that Vik fails to have a reason because he fails to properly apply the norms of the practice of science. If, on the other hand, strong internalism is right, then the interlocultor's objection is badly off the track. According to strong internalism, it is enough for Vik to have a reason if he has properly followed a deliberative route which he accepts and which originates in his subjective motivational set. Since these conditions have been satisfied, the strong internalist will maintain that Vik does, in fact, have a reason to believe H.
This is enough, I think, to make clear the distinction between strong and weak internalism. However, a few words are in order to distinguish weak internalism from externalism. Externalism, as I have written before, is the view that there are robust external reasons. Williams has characterized externalism with regard to some action F as the view that for an agent to have a reason to F, "F-ing will have to be an action that an agent could rationally decide to do as a result of deliberation whatever his S might be." Weak internalism is compatible with the truth of the claim that there are actions F such that any agent, regardless of her S, will decide as a result of deliberation that she has reason to F. Weak internalism denies, however, that the truth of that claim is either a necessary or a sufficient condition for the agent's having a reason to F. It is not a necessary condition because a particular agent may have reasons which could not be arrived at from just any subjective motivational set. It is not a sufficient condition because the availability of such a deliberative route will not guarantee that the route is followed by a given agent.