As soon as it is admitted, as it must be, that received opinion has to be consulted if a normative moral theory is to be formulated, one is forced from a methodological position like Hare's to a position like Rawls's. One is forced to recognise that there is not one-way traffic from the normative theory to the moral judgments which are correct in individual types of case, but a two-way traffic between the normative theory and the moral judgment. To quote, for the sake of variety, from Stuart Hampshire's recent book, "As Aristotle argued, there is a two-way traffic by which intuitions are modified by reflection on the general principles that explain them, and the general principles are qualified by particular cases that do not fit and fall under them. Aristotle therefore claimed that there is a strong practical interest, a guide to action, in moral theory."
If one is to think of moral philosophy in terms of some wide sweeping theory such as Utilitarianism or Contractualism, one is forced to adopt the methodological position which is adopted by Rawls and has Aristotle's ancient backing. But while accepting that, it is also impossible not to feel, apparently inconsistently, that Hare is perfectly right to object to the way in which Rawls tailors the theoretical structure until it yields his considered moral judgments. It is altogether too reminiscent of cooking the results of one's schoolboy chemistry experiments.
I distinguished earlier, by quoting from Rawls, between a moral theory designed to introduce "order and system into our considered judgments" (i.e., an explanatory theory) and a moral theory designed to provide answers to moral questions. And I remarked that if Rawls's theory was intended to come under the first description, there would be no blemish at all in tailoring the theoretical structure to match our considered judgments. Of course not! The trouble is that such a theory inevitably comes to have a limited settlability function (if I may be permitted the barbarism) in that, in thinking about the judgments and thinking about the theory, one will sometimes find it more acceptable to sacrifice a judgment rather than modify the theory. This is simply the way that the situation forces itself upon one. It is because of this limited settlability function, which the theory inevitably has, that the tailoring of the original position which goes on in A Theory of Justice looks so bogus. To some extent, a theory which is going to tell us what is right and wrong is being rigged up to match our antecedent judgments of what is right and wrong. Hare's criticism correctly fastens onto this point; but it is able to do so only because Rawls had done his work with a thoroughness which has seldom been approached by any moral philosopher before him.
(G. R. Grice, "Moral Theories and Received Opinion," The Aristotelian Society, supplementary volume 52 [1978]: 1-12, at 10-2 [endnote omitted])
Note from KBJ: What we have here is a dispute between progressives and conservatives as regards moral method. The progressive accords no weight to judgments, considered or otherwise, in the construction of a normative ethical theory. The conservative accords considered judgments some weight (at the limit, with infinite weight). Most utilitarians are progressives in this sense, with R. M. Hare, J. J. C. Smart, and Peter Singer being prominent examples. They believe that people's judgments (even their considered ones) are infected with bias, prejudice, ignorance, and superstition, so if considered judgments are allowed to influence one's ethical theory, these things (bias &c.) get built into one's theory. Conservatives, by contrast, reject the belief that people's considered judgments are suspect. To the contrary, they say: People's considered judgments reflect experience, trial and error, and wisdom. The progressive puts his or her faith in theory, which is viewed as a disinfectant. The conservative puts his or her faith in received opinion, which is viewed as the embodiment of collective wisdom and rationality. Note that John Rawls, a political progressive (specifically, an egalitarian), is a conservative as regards moral method. This shows that one can be a conservative in one respect or in one area while being a progressive in another respect or in another area.