Conflict settlement is a process rather than a singular act. At its most basic, a peace process comprises three phases: the negotiation, implementation and operation of an agreement meant to enable the conflict parties to resolve their disputes by nonviolent, political means.
Yet the successful conclusion of a peace process is by no means a foregone conclusion—they can, and do, fail. Sometimes negotiations break down and no agreements are concluded, leading conflict parties back to violence. In other cases, disagreements about the meaning of particular provisions arise after an agreement has been reached. In the absence of effective dispute resolution mechanisms, one or more parties are then likely to defect from a negotiated deal, reopening armed conflict. In yet other cases, spoilers might torpedo the operation of an agreement, inciting another round of fighting, because they and their constituencies have been excluded from negotiations or have not seen the expected returns from an agreement they signed up to. Spoilers might also be external powers with their own stakes in a conflict and its settlement, at times using proxy forces to derail a deal.
Peace agreements are only one dimension of peace processes, yet they provide the crucial link between war and peace—they need to be negotiated, implemented and operated, often in conditions that are anything but conducive to sustainable peace. Ideally, peace agreements offer incentives to conflict parties that make a return to violence unattractive; by establishing institutions and processes, agreements give the parties a reasonable prospect of seeing their interests accommodated at lower costs than by recourse to violence. In that sense, a peace agreement will stick if the content of the actual settlement really addresses the grievances of the conflict parties.