The Digest of Justinian

[ULPIAN, On the Edict, book 4]

​Of agreements under the law of nations, some generate actions, others exceptions.

​Those which generate actions do not remain under their general name, but pass over into the specific name of a contract: such as purchase and sale, letting and hiring, partnership, loan for use, deposit, and other similar contracts.

​But even if the matter does not pass over into another contract, yet there is an underlying cause, Aristo elegantly answered Celsus that there is an obligation. For example, “I gave you a thing so that you might give me another,” “I gave so that you might do something”: that this is a κυνάλλαγμα, and from this arises a civil obligation. And therefore I think Julian was rightly criticized by Mauricianus in this case: “I gave you Stichus so that you might manumit Pamphilus; you manumitted him; Stichus has been evicted.”

​Julian writes that an action on the facts should be granted by the praetor; but the latter (Mauricianus) says that a civil action for an uncertain amount, that is, one with prescribed words, is sufficient: for there is a contract, which Aristo calls κυνάλλαγμα, from which this action arises.

​If a promise is made in order that a delict may not be committed, no obligation arises from this agreement.

​But when there is no underlying cause, it is clear that an obligation cannot be established on account of the agreement; therefore, a nude pact does not generate an obligation, but it does generate an exception.

​Nay, indeed, sometimes it shapes the action itself, as in trials of good faith; for we are accustomed to say that agreed-upon pacts are inherent in trials of good faith. But this must be understood in such a way that if indeed the pacts followed immediately, they are inherent even on the part of the plaintiff; if after an interval, they will not be inherent, nor will they be valid if he brings an action, lest an action arise from a pact. For example, if after a divorce it was agreed that the dowry should be returned immediately, rather than at the established period of deferment: this will not be valid, lest an action arise from a pact; Marcellus writes the same. And if in an action on guardianship it was agreed that higher interest than that which is established should be paid, it will have no place, lest an action arise from a pact; for those pacts are inherent which give law to the contract, that is, which were made at the inception of the contract. I know that the same was answered by Papinian, that if after a purchase, after an interval, something outside the nature of the contract is agreed upon, an action cannot be brought on the purchase for this reason, due to the same rule, lest an action arise from a pact. And this must be said in all trials of good faith. But on the part of the defendant, the pact will have place, because even those pacts which are interposed afterwards are accustomed to generate exceptions.

​Moreover, exceptions made afterwards, which arise from the same contract, are inherent in trials of good faith to such an extent that it is established that in purchase and other trials of good faith, before performance has taken place, one can withdraw from the purchase. If therefore it is possible as a whole, why can a part of it not be changed by agreement? And Pomponius writes thus in the sixth book On the Edict. Since this is so, the agreement also has place on the part of the plaintiff, so that, by the same reason, it may benefit the action before performance has taken place. For if the whole matter can be abolished, why can it not also be remodeled? So that in a certain way the contract appears as if renewed. Which can be said not without subtlety. Hence I equally do not reject what Pomponius approves in his books of readings, that one can withdraw in part from a purchase by agreement, as if the purchase of that part were repeated. But when two heirs to the buyer existed, and the seller agreed with one of them that there should be a withdrawal from the purchase: Julian says that the agreement is valid and the purchase is dissolved in part; since even from another contract, by making an agreement, one of the heirs could acquire an exception for himself. Therefore both are rightly approved, both that of Julian and that of Pomponius.

​The praetor says: “Agreements which shall be made neither by malice, nor against laws, plebiscites, decrees of the senate, imperial edicts, and decrees of the emperors, nor whereby fraud is done to any of them, I will uphold.” Of pacts, some are in rem, others in personam. They are in rem whenever I agree generally that I will not sue; they are in personam whenever I agree that I will not sue a specific person, that is, that I will not sue Lucius Titius. Moreover, whether a pact is made in rem or in personam must be judged not less from the words than from the intention of the contracting parties; for generally, as Pedius says, a person is inserted in a pact, not that a personal pact may be made, but that it may be shown with whom the pact was made.