This assignment completes Dworkin’s discussion of precedents.

• §IV.B.2 explores some consequences of the ideas introduced in §IV.B.1. At its heart is the image of a “seamless web.” Dworkin has no regular contrasting phrase but that purpose could be served by the term patchwork, reflecting the idea of “piecemeal rationales” that appears in the first paragraph of the section (p. 1093). The contrast is the one Dworkin has pointed to several times (most recently in the argument of §IV.B.1) between the requirements of consistency imposed by arguments of principle and those imposed by arguments of policy (see pp. 1064f for his introduction to the issue). The section then lays out the considerations involved in interpreting precedent (as Hercules would do it) in a way that you can usefully compare with Hart’s account in The Concept of Law (which is limited to a single paragraph, pp. 134-5).

• §IV.B.3 is Dworkin’s theory of mistakes. In “The Model of Rules” (pp. 37ff), Dworkin argued that positivists have too few resources to explain the constraints on judges’ power to find the decisions of other judges to be mistaken, so it is important to see how much freedom he leaves to judges.

• Dworkin’s first topic (pp. 1096f) is not mistakes as such but the degree to which judges are bound by the principles that the authors of earlier decisions cited as their rationales.

• Dworkin then considers mistakes in their own right. He first considers why it can be reasonable for Hercules to conclude there has been a mistake (pp. 1097-1099) and then goes on to consider the theory of mistakes in two parts.

• The first part concerns what remains of the import of legislative and judicial decisions when they are deemed mistakes (pp. 1099f); notice in particular his distinction between gravitational force and “specific authority.”

• The second part (pp. 1100f) is the more interesting for our purposes. It embodies Dworkin’s account of the limits on judicial authority to count previous decisions as mistakes. Notice the two guidelines he states as well as the corresponding maxims.