1. Identify a domain of practices, such as scientific inference (chapter 2) or ethical reasoning
(chapter 9).
2. Identify candidate norms for these practices, such as inference to the best explanation (chapter 2)
or consequentialism (chapter 9).
3. Identify the appropriate goals of the practices in the given domain, such as truth (chapter 4) and
vital needs (chapter 8).
4. Evaluate the extent to which different practices accomplish the relevant goals.
5. Adopt as domain norms those practices that best accomplish the relevant goals.
Step 3 is the trickiest, because it requires complex consideration of relevant goals, taking into
account evidence, theory, and practices, as shown in figure 10.1. To establish goals for inference
about what to believe and do, we can ask such questions as the following. What do people aim for?
Why do they have those aims? Are the aims coherent with other goals? Step 4 is also difficult,
because it requires evidence about what practices causally produce goal satisfaction, not just
correlations between practices and goals. Nevertheless, these steps provide a way of using
descriptive evidence to address normative questions, as I will illustrate later in this chapter with
respect to the nature of government