ON THE DUTY TO COMPLY WITH PEACE AGREEMENTS
What Peace Agreements Do: Declarative, Specifying, and Creative Functions
The maxim pacta sunt servanda is a cornerstone of international law. It is also taken entirely for granted. Yet, in so far as compliance might not work to the advantage of signatories, and absent robust mechanisms for enforcement, it pays to inquire into the normative foundations of the obligation to abide by the terms of international treaties in general and peace agreements in particular. At first sight, the answer seems obvious: we all are under a duty to promote peace; in so far as violating peace treaties threatens peace, we all are under an obligation to comply with their provisions. But there are two difficulties with this move. First, in so far as we all are under a general obligation to promote peace, the view under scrutiny cannot on its own explain what is specific to the duty to comply with peace agreements—as distinct, for example, from the duty not to invade another country without just cause. Second, and relatedly, it does not account for the intuition that the victims of treaty violations are themselves wronged, and that their being wronged (as distinct from peace being undermined) provides at least part of the justification for the obligation to comply (and the right to seek redress for noncompliance). Germany’s invasion of Belgium and France in the Spring of 1940, in violation of the Versailles Treaty, may well have been a crime against peace: but it was also and foremost a crime against French and Belgian citizens, who did have a special claim against their 1914 aggressor that it comply with the Treaty—a specific claim which other signatories to the Treaty, for example the United Kingdom and the United States, did not have.
How, then, can we best justify pacta sunt servanda? One can think of a peace agreement as a tool whereby a victorious belligerent imposes its will on its defeated enemy (for example, the 1919 Versailles Treaty), as a contract between two equally situated belligerents seeking to put an end to their dispute (a vision which seems encapsulated in the 1969 Vienna Convention on the Law of Treaties), or as a joint undertaking between two moral equals who agree to set their past grievances aside and to cooperate with one another so as to bring about peace (for example, the 1998 Good Friday Agreements between the UK, the Republic of Ireland, and eight political parties from Northern Ireland). On the first model, treaties are a tool of statecraft suited to the demands of realpolitik. On the second and third of those models, treaty obligations are voluntary and thus relevantly similar to promissory and contractual duties.
In so far as the first model is incompatible with cosmopolitan morality, it is to the literature on promissory and contractual obligations that we should turn, which provides a useful set of lenses for unearthing the moral foundations of treaty obligations. That literature is too vast for me fully to account for it here. In particular, its preoccupation with the similarities and differences between promissory and contractual obligations need not be ours. Its other preoccupation—why we are under a duty to comply with promises and contracts—is relevant since we must defend the view that signatories are under an obligation to comply with just peace agreements. As a preliminary step, however, we need to elucidate what it is that those agreements do. At first sight, we might think that they confer on their signatories special, as distinct from general, rights and obligations. The distinction between general and special rights is usually drawn as follows.5 General rights are rights which we have in virtue of (depending on one’s account of rights) being human, or being a person. Special rights, by contrast, are standardly bestowed on their holders either as a result of some past event or deed, or through transactions such as contracts, agreement, and promises. However, while peace agreements to a large extent confer rights and obligations, the jural relations so established are not always special in that sense. For peace agreements have three functions as far as jural relations between signatories are concerned: a declarative function, whereby they state independently justified rights and obligations which can themselves be either special or general; a specifying function, whereby they particularize and specify independently justified rights and obligations which (again) can be either special or general; and a creative function, whereby they create rights and obligations whose sole justification lies in the agreement itself.
To illustrate the declarative function, consider the 1973 Paris Peace Accords, which put an end to the Vietnam War and whose signatories were the United States, the Democratic Republic of Vietnam, and the South Vietnamese Provisional Revolutionary Government. Article 1 states that ‘The United States and all other countries respect the independence, sovereignty, unity and territorial integrity of Vietnam as recognised by the 1954 Geneva Agreements on Vietnam’. Assuming for the sake of argument that the Vietnamese have a jointly held right to political self-determination, article 1 does nothing but affirm that (independently justified) right—which is a general right held by human beings in virtue of being moral and rational agents. Similarly, consider the 1919 Versailles Treaty—and in particular Part VIII, which sets out Germany’s reparative obligations to the Allies. Let us assume for the sake of argument that the Germans did have reparative obligations on account of their unlawful and wrongful invasion of Belgium and France. Reparative obligations and their correlative rights are clearly special—not general—in so far as they arise as a result of a past event, and are not held in virtue of being human or of some feature of agency. However, here too, the relevant clauses of the Treaty (articles 231—47) merely state a set of independently justified rights and obligations—justified, that is, at the bar of reparative justice.
Often, of course, peace agreements do more than declare those rights and obligations: they specify them in a way which creates further rights and obligations. Here is a simple domestic analogy. Suppose that while out on a walk the two of us stumble upon a child who has got separated from her parents and is in considerable distress, having broken her arm and fallen into a pond. Let us assume (plausibly) that this child has a right against us that we assist her, in virtue of a general right to help which we all have against everyone else. We quickly agree that I will get her out of the pond and that you will gather material for a makeshift splint. As a result of our agreement, we each bear a specific obligation which contributes to 
the fulfilment of our general obligation to help. Note, crucially, that there is no prior, independent justification for allocating to me, rather than you, the obligation to get the child out of the pond, and to you, rather than me, the obligation to prepare the splint: I am not a lifeguard, and you are not a nurse. If we each respectively occupied those roles, our agreement would be declarative, not specifying, since we each would have an independently justified obligation to help the child in those particular ways—justified, that is, on the grounds that the best qualified person for a particular task should perform that task. Note, further, that agreements of the kind just described can specify independently justified special rights and obligations. Suppose that you and I accidentally damage another person’s property. We clearly are under a special obligation to provide her with assistance (at a greater cost to us than would be the case if we were not responsible for her predicament) and to pay compensation for the harm we caused her. However, there are many morally equivalent ways in which we can discharge those reparative obligations, and the two of us might thus come to an agreement as to what you and I should do precisely.
Let me now bring those considerations to bear on the issue of international treaties in general and peace agreements in particular. Recall that at the bar of cosmopolitan justice, all individuals wherever they reside have rights to the freedoms and resources which they need in order to lead a flourishing life. Those rights in turn impose duties on everyone else. But while there is little difficulty in assigning on everyone in the world a general negative obligation not to, e.g., kill, torture, or lie, positive duties to provide assistance must be assigned to specific parties. Simply to say that those in a position to help must do so will not work, since vast numbers of agents may well be in a position to help, coordination might be required to deliver assistance, and so on. The point is familiar, of course: less familiar is the thought that treaties can function as a mechanism for the assignment of such duties. Consider the right not to be left to die of starvation. It correlates into an obligation on the part of everyone to provide food to anyone who is starving. But we still need to know who among those who have more than enough must provide food to whom among the starving. Or consider the right to political self-determination. It correlates into a negative duty on the part of all others not to mount an unwarranted military attack on politically sovereign communities and a positive duty to provide military assistance to those who are attacked (subject to costs). But we still need to know who must provide which kind of assistance for which threatened community. In the case of food, it is often said that domestic institutions, or states, are the best way to allocate the general obligation to help. Suppose however that a political community and its domestic institutions are unable on their own to discharge that obligation to their own members. A further layer of duty-holders is needed. Likewise, in the case of help to thwart a military aggression, some agent other than the state who is attacked must be identified. In both cases, bilateral and multilateral agreements or treaties can serve to identify who is under the relevant obligations to whom. They can also serve to specify how those obligations will be met.
How does this apply to a peace agreement? Whether one conceives of it as a contract, an instrument for the imposition of the victor’s will on the vanquished, or a joint undertaking between moral equals, a peace agreement is passed between parties which in virtue of the conflict which opposes them stand in a relationship characterized by the commission of wrongdoings and attendant grievances on (usually) both sides (even if one side is in a position to claim that its war was just in toto). The rights and obligations set out in the agreement are special in that they are justified by past events and are constitutive of erstwhile belligerents’ relationship with one another qua erstwhile belligerents. Moreover, as we saw a few paragraphs ago, in so far as many of its clauses pertain to compensation, restitution, and punishment for wrongdoings, the content of the terms it imposes is determined by prior principles of compensatory, restitutive, and retributive justice. However, to the extent that discharging those obligations necessitates joint action by multiple parties and admits of morally equivalent decisions, a peace agreement will cash out those independently justified obligations into specific obligations. Examples include a schedule of reparation payments, as set out (e.g.) in article 235 of the Versailles Treaty, or a timeframe for organizing a handover of legislative and administrative competences from Occupier to Occupied and the precise modalities of such transfers, as set out (e.g.) in the 1993 Oslo Accords between Israel and the Palestinians. Moreover, a peace agreement might also specify existing cosmopolitan and thus general rights and duties. Suppose that one of the parties, B, has been devastated by the war, and that its population is on the brink of starvation. To anticipate on ch. 6, at the bar of cosmopolitan morality, those individuals have a right to assistance which imposes obligations of assistance not just on the belligerent responsible for those harms but on outsiders as well. The peace agreement might serve as a tool to allocate those obligations if those outsiders (for example, some subsections of the international community) are willing to oversee the peace process.
Finally, treaties often create rights and duties de novo, which are necessarily special since they derive their justificatory force solely from treaties. Suppose that it is in X’s interest to get Z but that Y is not under an obligation to provide it. Under some circumstances, Y might be morally permitted, and have the moral power, to bind himself to help X. If you and I agree via a contract of sale that you will sell me your car for 5000 pounds, I acquire a right against you to that car, and you acquire a right against me to my money: we both will it to be the case that the content of both rights is as stipulated by the contract. Likewise, a peace agreement creates rights and obligations de novo when, for example, its parties use the peace process and resulting settlement as an opportunity to set up a military or trade alliance. Thus, treaties concluded by Britain and China in the middle of the nineteenth century following the two Opium Wars (1839—42 and 1856—60) stipulated terms (highly unequal to the Chinese) under which Britain and other Western merchants could trade with and within China. Under those terms, Britain was granted the status of so-called Most Preferred Nation (for trade) with no obligation on its part to reciprocate. On no plausible grounds did Britain have a prior, independent right to be granted such status, and it is appropriate therefore to regard the relevant treaty provisions as creative.
Importantly, the creation and conferral of de novo special rights and obligations are constrained by moral norms, for there are limits to the extent to which others can justifiably take advantage of a superior negotiating position to get us to agree to terms which we would otherwise reject. As we shall see in s.4.3, the prohibition on defrauding, deceiving, and exploiting the other side delineates such limits on the making of peace agreements, whereas the imposition of otherwise warranted coercion does not.
-  For a classic defence of the view that contractual obligations are a species of promissory obligations, see C. Fried, Contract as Promise: A Theory of Contractual Obligation (Cambridge, Mass.:Harvard University Press, 1981). For a illuminating critique, see D. Kimmel, From Promise toContract—Towards a Liberal Theory of Contract (Oxford: Hart Publishing, 2003). The view that atreaty is a promise by definition is crucial to Vattel’s and Wolff’s accounts of treaty obligations. SeeWolff, Jus gentium methodo scientificapertractatum, ch. IV; and Vattel, Le droit des gens, ou, Principes dela loi naturelle appliques a la conduite et aux affaires des nations et des souverains, Bk IV, ch. 3.
-  See H. L. A. Hart, Are There Any Natural Rights?’, The Philosophical Review 64 (1955): 175—91;J. Waldron, The Right to Private Property (Oxford: Oxford University Press, 1988), ch. 4.
-  See, e.g., R. E. Goodin, ‘What Is So Special about Our Fellow Countrymen?’, Ethics 98 (1988):663—86; H. Shue, ‘Mediating Duties’, Ethics 98 (1988): 687—704.
-  See, e.g., D. Wang, China's Unequal Treaties: Narrating National History (Lanham, Md.: Rowman& Littlefield, 2005), ch. 1. I am grateful to Vivienne Shue for bringing this case to my attention.