In Chapter 7, we examined the general applicability of our model to alliance formation, concluding that European alliances between 1871 and 1914 are consistent with a calculation of coalitional value derived from the concept of system stability, and consistent as well with the respective roles of Germany and Britain as central power and balancer. However, system stability does not imply resource stability, which leaves room at a more microlevel for threats and counterthreats to be directed at achieving, if not the “slight superiority” to which Haas refers, at least some advantage. This view of instability within stability leads us to different predictions about conflict resolution than others offer. Translating his assertion into our terms, Chatterjee sees resource instability yielding system stability only if the game is not constant-sum - if members of the system can expropriate from external entities. We argue, though, that alliance flexibility allows wholly endogenous adjustments, but only if those adjustments are consistent with the requirements of such stability. We must check, then, whether we can interpret actual resource adjustments as being constrained if not dictated by system stability, and subsidiarily to see what insights we can gain about alliance formation and maintenance.