Obligations — contract (the Roman consensual contracts)

The Roman consensual contracts are the point at which Roman private law most obviously ceases to look archaic. Earlier topics in Civil Law I have emphasised status, things, ownership, possession and conveyance. This week shifts from rights in things to rights against persons: obligatio as a legal bond by which one person is constrained to render something to another. The four consensual contracts are emptio venditio, sale; locatio conductio, hire; societas, partnership; and mandatum, mandate. Their defining feature is that obligation arises from agreement alone. No formal words, writing, delivery or religious ceremony is required. Consensus, if directed to a recognised contractual type, suffices.

That proposition requires two qualifications. First, Roman law was not a general law of contract in the modern common-law sense. The jurists did not begin with a universal doctrine that every serious bargain is enforceable. They worked through actions. A consensual contract mattered because it generated a particular action: actio empti and actio venditi for sale; actio locati and actio conducti for hire; actio pro socio for partnership; actio mandati directa and contraria for mandate. The question was not merely whether there was agreement, but whether the facts fitted an actional category. Secondly, consensus did not mean subjective harmony in the air. It meant agreement on the legally essential elements of that contract: thing and price in sale; thing or work and rent or remuneration in hire; common lawful venture in partnership; requested gratuitous service in mandate.

For Cambridge purposes, the topic is especially valuable because it forces you to think structurally. These contracts are not four isolated institutions to be memorised. They are a controlled experiment in Roman legal technique. Sale shows the separation between obligation and transfer of ownership: the contract binds, but traditio or another conveyance transfers dominium. Hire shows the difficulty of drawing boundaries between property use, labour and the production of a result. Partnership shows contract without legal personality, and the strong dependence of doctrine on personal trust. Mandate shows an enforceable gratuitous undertaking, thereby exposing the provincial character of modern assumptions about consideration.

The organising ideas are consensus, bona fides, reciprocity, risk, and actional remedies. The consensual contracts were bonae fidei iudicia: the judge was instructed to condemn according to what good faith required, rather than merely according to the literal performance of fixed words. That matters in problem questions. The content of the parties’ obligations was not exhausted by express terms. Good faith could require disclosure, care, cooperation, reimbursement, accounting and abstention from opportunistic conduct.

In an examination answer, avoid treating the consensual contracts as an early version of English contract law. There is no consideration requirement; no doctrine of privity in the common-law form; no single theory of offer and acceptance; and no general remedy of damages for breach of any promise. Conversely, avoid romanticising Roman law as pure moral consensus. It was intensely classificatory. Agreement was enforceable because the praetor and the classical jurists recognised these types and attached actions to them. The best Tripos answers hold both ideas together: Roman law is conceptually sophisticated, but its sophistication is institutional and actional, not codified around an abstract modern code of contract.

The consensual contracts sit between early formalism and later civilian generalisation. In the older Roman law, enforceable obligation was closely associated with form. Stipulatio, the paradigmatic verbal contract, required a question and answer in matching form. Nexum, in the archaic period, and certain solemn conveyances likewise reflected a society in which legal efficacy was secured by ritual. That did not make early Roman law primitive; formalism performed evidential, cautionary and status functions. But it was ill-suited to the expanding commercial life of the Republic, especially where dealings involved peregrines and transactions across distance, market and language.

The usual account connects consensual contracts with the ius gentium and the work of the praetor, especially the praetor peregrinus. The point should be stated carefully. Ius gentium was not international law in the modern sense. It was the body of institutions and modes of reasoning thought to be common, natural or convenient among peoples, and therefore apt for dealings beyond the strict citizen law. Sale, hire, partnership and mandate were obvious candidates. They were recurrent transactions in ordinary commerce: buying goods, letting land or labour, joining in ventures, asking another to act on one’s behalf.

Their emergence also reflects an actional transformation. The older legis actiones gave way, in practice, to the formulary procedure. The formula allowed a more flexible instruction to the judge. In bonae fidei actions, the judge was directed to assess what ought to be given or done according to good faith. This procedural setting is not incidental. It is one reason why Roman consensual contracts could be both informal and controlled. The absence of form did not entail unbounded enforceability, because the action’s formula and the juristic understanding of the contract supplied limits.

By the classical period, the four consensual contracts were established as a distinct class. Gaius treats them in Institutes book 3 after real, verbal and literal obligations. Justinian later preserves the taxonomy. The classification is familiar: obligations arise re, verbis, litteris or consensu. Real contracts, such as mutuum, commodatum, depositum and pignus, require delivery of a thing. Verbal contract requires formal words. Literal contract requires entry in writing. Consensual contract requires agreement alone. This scheme is pedagogically powerful, which is one reason it survived into the institutional literature and then into medieval and modern civilian thought.

The four contracts, however, did not all develop in the same way. Sale became the central commercial contract and generated elaborate doctrines on price, risk, latent defects and eviction. Hire became a broad and sometimes unstable category covering hire of things, hire of services and contracts for work. Partnership remained intensely personal, shaped by trust and dissolving readily on death or renunciation. Mandate preserved the moral and social world of amicitia: one person requested another to perform a service gratuitously, but the law nevertheless imposed obligations of care and indemnity.

The later reception of Roman law enlarged the conceptual importance of these categories. Medieval civilians and canonists used Roman materials to think about pact, cause, good faith and enforceability. Early modern natural lawyers and later civilian codifications moved towards broader general doctrines of contract. Yet the Roman starting point remains distinctive. It is not that Rome already had the modern law of contract in embryo. Rather, Roman law supplied a disciplined set of transaction types and a language of obligation which later jurists could generalise. For Part IA, the historical lesson is methodological: Roman law did not derive rules from a single abstract principle of promise-keeping; it refined a small number of recognised actions until they became capable of sophisticated commercial reasoning.

1. Consensus is the mode of formation, not a universal theory of enforceability. The Roman consensual contracts were formed by agreement alone. That is the sense in which they differ from real, verbal and literal contracts. But the agreement had to be agreement of a kind recognised by the law. A bare pact, nudum pactum, did not ordinarily generate a civil action, though it might have other effects, especially by way of defence or later praetorian intervention. The consensual contracts therefore illustrate both liberality and limitation. They reject formal words; they do not recognise every serious arrangement as actionable.

1. The contracts are nominate. Roman law’s nominate categories are central. In sale, the parties must agree on the merx and pretium. In hire, they must agree on the thing, service or work and the merces. In partnership, they must agree to pursue a common lawful object and share gains, and usually losses, in some agreed or legally implied proportion. In mandate, one party must request another to perform a gratuitous service or transaction, normally in the interest of the mandator, a third party, or both. If the facts do not fit, the answer may lie in stipulation, in an innominate contract, in unjustified enrichment, or in no action at all.

1. Sale does not transfer ownership. Emptio venditio creates obligations. The seller must make the thing available, protect the buyer against eviction, and answer for certain defects; the buyer must pay the price. But dominium passes only by an appropriate conveyance, usually traditio for res nec mancipi in the classical period, and subject to the transferor’s ownership and intention to transfer. This separation between contract and conveyance is one of the most important Roman ideas for comparative private law. It prevents the student from assuming that a valid sale itself makes the buyer owner.

1. Sale requires a price in money. Classical Roman law treated barter as distinct from sale. The Proculian view, ultimately dominant in Justinian’s presentation, insisted that pretium be pecunia numerata, money, not another thing. This was not a mere semantic quibble. It determined which party was buyer, which seller, which actions were available, and how risk and warranties operated. The price also had to be certain or capable of ascertainment by agreed method. Later law’s concern with just price should not be projected too readily into the classical law.

1. Risk in sale is a technical doctrine. The maxim periculum est emptoris is often rendered as the risk is the buyer’s. It must not be turned into a crude rule that the buyer always bears every loss after informal agreement. The sale must be perfecta: the thing, price and relevant conditions must be settled. The doctrine concerns accidental loss or deterioration, not loss caused by the seller’s fault. Nor does it mean that ownership has passed. Roman law could put risk on the buyer before ownership, a feature which modern students often find counterintuitive. The doctrine belongs to the structure of sale as an obligatory contract with separate conveyance.

Roman law does not possess landmark cases in the common-law sense. Its leading authorities are juristic texts, imperial constitutions and institutional statements. For this topic, the most examinable authorities are not litigation reports but passages in Gaius, Justinian and the Digest which establish the working grammar of consensual obligation. The keyCases array below therefore identifies leading Roman law source-texts rather than English cases.

The first cluster concerns formation by consensus. Gaius’ treatment of obligations consensu is indispensable. It explains why sale, hire, partnership and mandate differ from real contracts. Real contracts require delivery because the obligation is generated by the handing over of a thing. Consensual contracts, by contrast, are complete by agreement. This is the beginning of any answer. If a student cannot state precisely why emptio venditio differs from mutuum or commodatum, the answer will lack the institutional foundation required in Part IA.

The second cluster concerns sale. Gaius’ discussion of emptio venditio gives the essentialia: thing and price. Justinian’s Institutes add the settled presentation of controversies about price, including the rejection of barter as sale. The doctrinal consequences are substantial. If A agrees to give B a horse in exchange for B’s slave, the transaction may be an exchange, but it is not sale in the strict Roman analysis if there is no monetary price. This matters because the contractual actions of buyer and seller presuppose identifiable roles.

The third cluster concerns hire. Gaius’ treatment of locatio conductio shows Roman law’s willingness to group several modern categories under one contract. The hire of land, the employment of labour and the commissioning of work are conceptually linked by the idea of placing something at another’s disposal for reward. But the unity should not conceal the distinctions. In a contract for work, the result may matter more than the labour; in hire of a thing, the obligation concerns use and return; in services, the labour itself is central.

The fourth cluster concerns partnership. Gaius and Justinian present societas as consensual, personal and terminable. Its importance lies partly in what it is not. It is not a corporation. It is not a proprietary fund separated from the partners. It is not indestructible by unilateral withdrawal. The actio pro socio enforces the relationship internally and carries the moral pressure of infamia. That feature illustrates how Roman private law often combined patrimonial liability with status consequences.

The fifth cluster concerns mandate. Gaius’ mandate materials show that a gratuitous undertaking may bind. That is doctrinally important for comparison with English consideration, but it should also be understood on its own terms. Mandate belongs to a social world of friendship, service, trust and representation. Its gratuitousness distinguishes it from hire; its enforceability distinguishes it from mere moral favour. If the mandatary incurs expenses properly, the actio mandati contraria protects him; if he acts badly or outside instructions, the mandator has the actio mandati directa.

When using these authorities in an essay, avoid artificial case-law language. Do not write that Gaius held as a court would hold. Say that Gaius states, Justinian presents, or the Digest preserves a juristic view. The strongest answers nevertheless treat these texts as authorities for propositions. Roman legal reasoning is not less rigorous because its leading materials are not law reports. It simply belongs to a different legal culture.

See practice questions section below.

See further reading section below.