As we saw on Monday, Gaus believes that the externalist view of having a reason carries with it serious problems. Furthermore, the attempt to decrease the diversity of reasons that one has through idealization is beset by the twin problems of indeterminacy and path-dependence. Even with radical idealization of our cognitive faculties, we would still not necessarily, or even likely, end up sharing all of our reasons. This leads Gaus to give up on the idea of full rationality as a possibility.
Once we give up on the idea of full rationality, we are led, Gaus argues, to theorize from the point of view of what John Pollock called real rationality. Pollock distinguishes, helpfully, between justified and warranted choices. Justified choices are the products of epistemically valid procedures of reasoning; warranted choices are the product of all possible relevant reasoning. Gaus argues “in a world of less than perfect information and cognitive capacities, we need some concept to indicate when a person’s reasoning about the world is up to acceptable standards and when it is not.” (247)
This conception of justification cannot be equated with truth, however. The fact that there is a reason does not necessarily mean that anyone actual person will necessarily be justified in acting on that reason. One can be justified in having a reason, but what ultimately matters in terms of interpersonal justification is whether or not that reason is warranted. One can be reasonably said to have a reason, however, if they do not have any defeaters that are accessible to them. This standard is importantly not that there are no defeaters, there may be, but they are not accessible to a person that has done a reasonable amount of reflection and investigation. Gaus argues that “the reasons you have must be accessible to you, and as a real rational agent in a world in which cognitive activity has significant costs, rationality does not demand one keep on with the quest to discover less and less accessible reasons.” (253)
What counts as a respectable amount of deliberation is often vague and will vary with context. Gaus compares what counts as a respectable amount of deliberation in a physics seminar to what counts for a baseball umpire. Baseball umpires need to make calls quickly and, hence, deliberation must be very quick, understanding that there will be a certain amount of error. Not so in a physics seminar.
Morality, after all, is not meant to be the esoteric doctrine of the epistemic elite. We expect normal adults to be able to grasp and follow the rules of social morality. Gaus claims that “normal moral agents have accessible undefeated reasons to affirm” the rules of social morality. (255) The idea that normal adults should be able to recognize their moral reasons sets a maximum limit on the epistemic demands of normal moral reasoning. But, we must be careful not to set the bar too low. People do wrong and we often think it is because they did not take sufficient care or deliberation before they acted. The thing to do is not always glaringly obvious. After all, our conceptions of social morality are not static, we genuinely learn both from our own actions and from what others tell us. The possibility of moral change, both progressive and regressive, occurs because finding out what reasons we have is often a social and collaborative venture.
As Gaus admits, though, “this is a messy and indeterminate sort of standard – different people will have different reasons, and we may not always be able to say what people’s reasons are.” (258) We may wish for something more, but real rationality with all of its problems is the cognitive situation in which we find ourselves. If we are concerned with justifying social morality and political authority to beings like us, this is all we can hope for. The silver lining, according to Gaus is that if an account of moral justification can be developed in this framework, it will be more robust. The big question will be whether such an account can indeed be developed. A lot of the heavy lifting will be done in the next chapter.
Some questions come up at this point. For one, the account that Gaus has presented here will be, most likely, extremely controversial. There are more than a few externalists out there, both in the sense of what reasons there are and what reasons we have, and it is hard to see how they will be able to sign on to much of what comes after this chapter if they reject his account of reasons. Of course, if someone rejects this account of reasons wholesale, it is hard to see why they would care about the justificatory project that concerns Gaus in the first place, so maybe it is not much of a problem.
One of Gaus’ concerns is with moral esotericism, a concern that also motivated Baier, Williams, and Rawls. Moral rules need to be accessible to ordinary moral agents. Idealizing too far will make this difficult, but I wonder if there isn’t another problem. The rules we have and the emotions that help to organize our motivation and behavior to recognize and follow rules are the products of biological and cultural evolution. We know that all of these rules and dispositions were, at least at one time, adaptive but the actual origin and function of any given disposition or rule is hidden from us.
Now, of course, we don’t need to know why we have fingers to know how to use them or that they are valuable. Similarly, the fact that a disposition to resent those who have harmed us maliciously is the result of millennia of evolution does not make the sentiment any less real or justified. But this does raise the question of what the limits of the reasons we have are. Some of the later parts of the book will deal with this issue in more detail, especially Section 20.
I am curious though, how important other people find this issue of moral esotericism or moral accessibility as a desideratum of social morality. It is clearly very important to Gaus, but certainly many other moral philosophers are less concerned with it. I would also be interested to hear how troubling other people found the problem of indeterminacy and path-dependence with regards to our sets of reasons even post idealization. It seems that someone could, potentially, admit that there is, indeed, indeterminacy with the beliefs and reasons of idealized agents but claim that the indeterminacy is still less than it would be for real agents and so there may still be a reason to have an idealized standard. We could still ask how the idealized standard is meant to be a normative ideal for lowly reasoners like us, but the general line of argument doesn’t seem completely hopeless to me.
Another approach might be to reject any idealization at all and to argue that the Reason Affirmation Thesis is closer to the truth. There are, no doubt, problems with this view as well, but one might be able to run many of the arguments that Gaus uses against radical epistemic idealization against his more modest form of idealization. If so, Gaus might find himself in a dilemma between radical idealization and the Reason Affirmation Thesis.