Alan H. Goldman

Genuine rules cannot capture our intuitive moral judgments because, if usable, they mention only a limited number of factors as relevant to decisions. But morally relevant factors are both numerous and unpredictable in the ways they interact to change priorities among them. Particularists have pointed this out, but their account of moral judgment is also inadequate, leaving no room for genuine reasoning or argument. Reasons must be general even if not universal. Particularists can insist that our judgments be reflective, unbiased, informed, and sensitive, requiring a background of experiences that expand sympathy and empathy for others. But beyond this, our judgments must be coherent. This requirement provides a way to reason to the correct answer to a controversial issue—the answer most coherent with or body of settled judgments. Rawls' account of coherence in terms of reflective equilibrium, where we adjust particular judgments to match rules and adjust rules to match judgments, is rejected since rules have no independent force. Instead, the central requirement is that we not judge cases differently without being able to cite a morally relevant difference between them. Such differences must make a difference else-where as well, although they need not do so universally. Factors cannot be relevant in only one context because they reflect values that must recur to be maintained. The method of moral reasoning based on this requirement is specified as follows: first, the specification of competing values or interests in the problematic case; second, the location of paradigm cases in which these competing values are prioritized, making sure that these settled judgments are reflective, informed, and sensitive; third, the search for relevant differences between the settled and problematic cases or the location of alternative, more closely analogous paradigms. The paper ends with an illustration of the method applied to the issue of doctor assisted suicide.

To say that an object is beautiful or ugly is seemingly to refer to a property of the object. But it is also to express a positive or negative response to it, a set of aesthetic values, and to suggest that others ought to respond in the same way. Such judg- ments are descriptive, expressive, and normative or prescriptive at once. These multiple features are captured well by Humean accounts that analyze the judgments as ascribing relational properties. To say that an object is beautiful is to say, in part, that it is such as to elicit a response expressing pleasure in certain observers. The observers in question must not be ignorant, biased, insensitive, or of poor taste, and they must not base their evaluations on aesthetically irrelevant properties of the objects they judge. The reference to the object's "being such . . ." captures the objective side of the relation; refer- ence to the pleasurable response captures the expressive function of these judgments; and the ideal properties of the observers suggest that others ought to judge in the same way