This chapter considers two additional forms of flexibility: imprecision in the language used to describe an agreement's substantive terms and reservations added to agreements that allow states to opt-out of certain provisions. The COIL sample displays great variation on both of these dimensions.

In fact, the variation at first glance appears to be issue-area driven, with 32 percent of the human rights agreements having reservations attached to them at the time of entry into force as opposed to only 1 percent of the economics agreements. Moreover, only 2 percent of the economics sample can be characterized by very or somewhat vague language, whereas almost 40 percent of the human rights sample can. Do these descriptive statistics imply that human rights agreements are far less serious forms of international cooperation than are economics agreements, or even that human rights agreements are just scraps of paper or meaningless agreements, perhaps meant to placate an interest group, as some scholars like to argue?

In this chapter, I challenge the quotes that open this chapter, both of which engage the issue area of human rights, by providing a theoretical explanation for imprecision and reservations based on the underlying cooperation problems and the number of states involved in the cooperative endeavor. In doing so, not only can these two flexibility devices be understood as rational choices by states whose goal is indeed to cooperate, but the issue area of human rights loses much if not all of its explanatory power. Put differently, the design of human rights agreements can be explained within COIL's general theoretical framework.