David Ibbetson, The 'Obligation' in Roman Law and Society
The words obligatio and obligare derived from the root ligare, meaning to bind. Binding might stem from any source that generated a claim in personam..
David Ibbetson wrote:
Chapter 43
Obligatio in Roman Law and Society
1. Introduction
Four centuries ago, in his commentaries on the civil law, Donellus analysed the nature of obligatio in Roman law. Subsequent scholars by the score have continued his work and continue to do so. It would be impossible in a short piece to engage with all of this literature, so the purpose of the present discussion is simply to outline what is known and to locate it within the Roman world.
Lawyers today rather take it for granted that there is a branch of the law called the law of obligations, including at the very least contract and tort, and that there is something called an obligation, designating the situation where one person is legally constrained to do or abstain from doing something to another. But neither of these is obvious, and it is highly likely that both are the products of Roman jurisprudence. It is true that in his treatment of justice in the Nicomachean Ethics Aristotle linked together contract and wrongdoing, the voluntary and involuntary types of sunallagmata—transactions or interactions—that might generate a requirement of corrective justice, but it goes too far to identify Aristotle’s sunallagma with the Romans’ obligatio: the former is an interaction that gives rise to corrective justice, the latter the abstract consequence of such a transaction.
To understand the nature of obligatio we should begin with the verb form, obligare, and see its etymology and meaning. Its core meaning, already by the time of Plautus, is to bind, tie up or fasten, with its root ligare derived from the Greek λυγόω, with the same sense. Its secondary, abstract meaning of putting a person under a duty, which is found by the end of the Republic, maintains its link with this concrete meaning of binding or tying. In addition, it could refer to the binding of a thing, as where its owner pledged it to another. It is from here that we get, at the latest by the time of Cicero, the noun obligatio.
Related to it is nexum. We need not enter into the controversy about the institution, abolished in the fourth century BC, but may observe that it too is linguistically related to tying, the verb nectere being effectively a synonym of ligare. It involved one person falling into the bondage of another by a formal transaction per aes et libram, and perhaps encapsulated an idea or image of being bound and led off into captivity. By the time of the classical jurists it seems to have been practically synonymous with obligatio.
But we should remain with the verb obligare for the moment. In the active voice it referred to the momentary act of binding or constraining, and once that momentary act had passed no additional meaning could be given to it. The perfect obligavi, that is, did no more than refer back to the event that had already occurred. This focus of the verb placed the active participant centre stage: he was the person who bound the other. It follows from this that as we move into the abstract, legal domain it was the act of obliging that lay at the core of the word in the active voice. It was possible to use a reflexive form of the verb, where it was the active party who came under the duty. According to Quintus Mucius, a borrower who misused the thing lent was liable as a thief, furti se obligavit; again the stress is on the way in which the liability had been created, not on the continuing state. The passive voice was rather different, since it could refer both to the momentary act of becoming bound and the continuing state of being bound. Hence, a person might be obligatus; that is to say, he might have become obliged and still be obliged. The lex de Gallia Cisalpina (terminus ante quem of 42 BC), for example, deals with the situation where someone has confessed himself to be obliged, “obligatum se … esse confessus erit”, using the past participle.
The juristic texts in the Digest reflect this usage. When the active voice is used in the present tense it is normally to make a general statement, such as that an institor binds his principal or that a slave obliges someone to his master, and in the perfect tense it looks back to an act of obliging that has occurred in the past, as where a landowner has obliged himself to compensate for damnum infectum or to make a gift. In the passive, the usage is similar. In the present tense it makes a general statement, such as that we are obliged in some circumstances, and in the perfect tense it refers to something which has happened in the past which has consequences for the present. The person who has dug a pit where he should not do so, with the result that an animal has fallen into it and been injured, has thereby become obliged—obligatus est—under the lex Aquilia.
Verbs precede abstract nouns. David Daube has shown that in many situations the Roman jurists never got as far as creating abstract nouns, using instead verbal forms, or that they clearly preferred verbal forms to nominal. Obligatio, however, is a frequent occurrence in the texts, hardly less frequent in the Digest than the various forms of the verb. We can say that the development of the noun presupposes a sophistication of thought about the institution, an ability to treat an active relationship as a thing, and that at least by the time of Labeo, the first jurist known to have used the noun, Roman law had taken that step.
As a noun, obligatio has none of the imbalance of the verb, for it can refer both to the act and to the state of obliging, from both the active and passive standpoint. That said, there is probably a slant towards the passive end: it makes perfect sense to speak of a person being obliged without specifying who it is that is doing or has done the obliging, whereas we cannot so easily speak of the active party to the obligation without saying or implying who it is that is being or has been obliged. Perhaps more to the point, though, the noun carries with it a greater sense of the relationship between two persons than does the verb. It is this that accounts for its function in the institutional structure of Roman law.
It is noteworthy that, despite the fact that the etymology of obligatio lies in lay usage, the word itself is very unusual in classical Latin outside the legal context; it seems not to be found before Tertullian around 200 AD. The bilaterality, which was so useful in the law in expressing the relationship between two people, was out of place in an ethical context where what was important was the duty under which one person found him- or herself. Here it was only when the word had developed sufficiently that it could refer simply to the duty without any connotation of a correlative right that it could be applied freely in the field of ethics.
2. Obligations: The Institutional Structure
The place of obligations in the Institutes of Gaius and Justinian can be easily described. According to Gaius, all law relates to persons, things or actions. Obligations are seen as forming part of the law of things. In this category Gaius deals first with the law of property, the relations between persons and things, and then moves on to obligations. The transition occurs at Inst.Gai.3.88:
Nunc transeamus ad obligationes. Quarum summa divisio in duas species deducitur: omnis enim obligatio vel ex contractu nascitur vel ex delicto.
Let us proceed to obligations. Their basic division is into two species: for every obligation arises either from contract or from delict.
This then provides the shape for Gaius’ treatment of personal obligations: first contracts, then delicts or wrongs. The same duality is found in Gaius’ treatment of personal actions: these, he says, seek to enforce an obligation imposed by contract or delict; though we should observe that already in the Institutes he has referred to the personal action to enforce a legacy per damnationem. In truth, contract and wrongdoing are merely the two principal sources of obligatio, and in the Digest Gaius is seen to be giving a slightly expanded categorisation, adding an additional residuary group:
Obligationes aut ex contractu nascuntur aut ex maleficio aut proprio quodam iure ex variis causarum figuris.
Obligations arise either from contract or from wrongdoing or by some special right from various types of causes.
Noteworthily, although the sources of obligations are given in the Institutes, Gaius gives nothing like a definition of obligatio. We might say that his primary reason for introducing the category of obligations into the institutional framework at this point has nothing to do with obligations themselves, but serves rather to provide a peg on which to hang the discussion of contracts and delicts.
Justinian’s Institutes 3.13pr-2 are more forthcoming, though perhaps little more revealing:
Nunc transeamus ad obligationes …
Omnium autem obligationum summa divisio in duo genera deducitur: namque aut civiles sunt aut praetoriae. Civiles sunt, quae aut legibus constitutae aut certe iure civili comprobatae sunt. Praetoriae sunt, quas praetor ex sua iurisdictione constituit, quae etiam honorariae vocantur.
Sequens divisio in quattuor species deducitur: aut enim ex contractu sunt aut quasi ex contractu aut ex maleficio aut quasi ex maleficio.
Let us proceed to obligations …
The basic division of all obligations is into two genera: for they are either civil or praeto- rian. Civil are those which are constituted by statutes or at least recognised by the ius civile. Praetorian are those which the praetor has established out of his own authority, which are also called honorary.
The next division is into four species; for they arise from contract, or as if from contract, or from wrongdoing, or as if from wrongdoing.
The transition is the same as in Gaius’ Institutes, from property to obligations, but we now have two divisiones of the sources of obligations, one formal and one substantive. It is the latter on which we should concentrate. Four types are given—from contract, as if from contract, from wrongdoing and as if from wrongdoing—adopting the same structural approach as Gaius but expanding the content. It is this fourfold classification that shapes Justinian’s Institutes’ subsequent treatment of the substance of personal obligations. Notwithstanding this expanded classification, however, in book IV of the Institutes personal actions are still treated as being derived from either contract or delict.
3. Definitions of Obligatio
As well as expanding the sources of obligationes, Justinian’s Institutes (3.13pr) includes what we might take to be a definition of obligatio:
Obligatio est iuris vinculum, quo necessitate adstringimur alicuius solvendae rei, secundum nostrae civitatis iura.
An obligatio is a tie of law by which we are of necessity constrained to pay some thing according to the laws of our civitas.
As a workable definition of obligatio this looks less than ideal. First of all, there appears to be some redundancy in the use of both necessitate and adstringimur: if we are constrained to do something, then we cannot but be constrained to do it of necessity. It may, however, be that the correct translation of the Latin is something like “An obligatio is a tie of law, whereby we are constrained by the necessity of paying some thing, according to the laws of our civitas”, which would avoid this problem. Secondly, taking the Latin at face value, the only thing that is described as the object of the obligatio is a payment, a handing over; there is no consideration that there might be an obligation to do something, an obligatio faciendi. Thirdly, the constraint is by the laws of our civitas, hardly the phrase to be used when describing the law of a world Empire. Moreover, since we have just learned that the summa divisio of obligationes is between those which arise from the ius civile and those which derive from the ius honorarium, the reference to the laws of our civitas cannot be a reference simply to the Roman ius civile, though it may reflect an earlier understanding according to which the term obligatio was only appropriate to relationships arising under the ius civile. And fourthly, the focus on constraint or enforceability is too narrow to deal with the class of unenforceable natural obligations, something of which we know from elsewhere in the Roman legal corpus.
The definition, if we can call it that, has all the marks of having been lifted from a classical or immediately post-classical source, possibly from Papinian (to whom all but two of the Digest’s uses of vinculum in the sense of an abstract bond are attributed) … The latter is perhaps more likely; even leaving out stylistic considerations, it is easier to imagine Justinian’s compilers having turned to the Gaian text than to Papinian to find their “definition”. In any event, whatever its source, it has probably been decontextualised without much thought for its generality or its appropriateness to the sixth century rather than the second or third. If what was to follow was an analysis of the nature of obligationes, a definition of this sort would have been near disastrous, but as a preliminary to the discussion of the different causae it does no harm. It stresses the personal nature of contractual and delictual obligations and their associates, contrasting with the relationship between person and thing which characterised the law of property.
Of central importance to this definition, though, is the description of the obligatio bond as a tie of law. Other relationships dependent on trust, fides, might generate a bond between two people, but that bond was not a tie of law. It was not an obligatio. For the original author of the definition, be it Gaius or Papinian, it may be that the reflexive relationship between obligatio and actio may have been essential. If this were so, obligatio might perhaps then have referred exclusively to an enforceable relationship and hence have excluded any idea of natural obligation. It does not absolutely follow that the vinculum iuris necessitated an actionable obligation, though; so long as the bond was one which had legal ramifications it could be described as a tie of law.
A rather different definition, or description, is attributed in the Digest to Paul:
Obligationum substantia non in eo consistit, ut aliquod corpus nostrum aut servitutem nostram faciat, sed ut alium nobis obstringat ad dandum aliquid vel faciendum vel praestandum.
The substance of obligations does not consist in that it makes some property or servitude ours, but it binds another person to us to give, do or be responsible for something.
The contrast with property is here made absolutely explicit. The obligatio does not make something ours, but constrains another person; it is therefore personal rather than real. It does not have as its object only a payment, but a transfer, a doing or a standing responsible for something, thereby avoiding the narrowness of the definition in the Institutes. The localism inherent in the reference to civitas in the Institutional definition is absent, and there is no verbal redundancy as may have occurred with the inclusion of necessitate. Most importantly, though, while the definition in the Institutes has a very firm focus on the person who is obliged, Paul’s approach is more two-sided: another person is bound to us to give or do something.
This double aspect of the noun obligatio is fundamental. It can be seen from the standpoint of the beneficiary of the obligation, the person who is under the obligation, or both. Moreover, it has the capacity to be reified, so that the relationship between person and person can be seen as a thing, an asset of the beneficiary and a liability of the person obliged.
4. Obligatio as Relationship
Although the obligatio connotes a relationship between persons, in a specific context it might refer to that relationship from the standpoint of the person bound. This is clear, for example, [when]… One person makes a stipulatio with his neighbour that his eaves might project over the neighbour’s land, and then buys a second property. It is said that the original stipulatio (probably) does not apply to the second house, lest the obligatio promissionis be increased. When the senatus consultum Velleianum regulated obligationes of women, it was transparently referring to their being bound. Similarly, a transfer of a slave to be freed by the transferee after the death of the transferor is said to create an obligatio, i.e. an obligation on the transferee to free the slave; and a fideiussor can fall under an obligatio. On entry into a hereditas, the heir takes on (suscipit) the obligationes of an inheritance, a procurator may take on the obligationes of his principal, a son or slave should be relieved of an obligatio, and more generally a person may be freed from an obligatio. An obligatio could be transferred from fideiussor to freedman, or from a solvent to an insolvent debtor, or a noxal obligatio can be transferred where another person confesses that he or she is the owner of a slave whose wrong is the basis of a claim. When it is said that an obligatio would be made more burdensome, it cannot be anything but the burden to the person who is under the obligation that can be meant. Finally, although it is a new relationship that is created when an obligatio is novated, the focus is on the new liability that comes into existence rather than on the new right.
Equally, the obligatio can be seen from the standpoint of the beneficiary. A stipulatio, for example, could be described as an act to obtain an obligatio (actus ad obligationem comparandam); it is something from which the stipulator acquires an obligatio, though where the name of one’s son is added as an alternative payee no obligatio is created for him thereby. There are frequent references in the Digest to the acquisition of an obligatio; a slave acquires obligationes for his master, an obligatio of mortgage cannot be acquired through a free person, where one person lends money in the name of another an obligatio is acquired by the person in whose name the money is lent. An obligatio of the ancestor accrues to the heir, and if there are co-heirs the obligatio must be divided between them. Should the heir sell the hereditas he must transfer any obligationes he has created to the purchaser. That it was the beneficiary’s side of the obligatio that was being referred to could be stressed by describing it as a ius obligationis, as something that could be sold.
Although references to obligatio normally refer primarily to one side of the relationship or the other, they might refer to the relationship itself. Hence, Ulpian (following Labeo) describes a contract as ultro citroque obligatio, stressing the bilaterality of the obligatio, and more specifically in the context of the contract of commodatum Paul writes of the obligatio inter dandum accipiendumque, thereby bringing into the foreground that the obligatio is a link between the two parties and not something that adheres to one of them or the other.
5. Obligatio as Thing
Although the obligatio was at its core a relationship between two parties, as a noun it was reified so that it was seen as an asset in the hands of the beneficiary. In his Institutes, Gaius places obligationes in his list of incorporeal things, alongside inheritances, usufructs and servitudes; something that is repeated by Justinian. An obligatio therefore had a duration; it could be said to be born and to be extinguished or brought to an end. It could endure or remain, or it could be perpetuated. If a transaction was undone so that there was restitutio in integrum, a former obligatio could be restored. When a novation occurred it could be said that the obligatio had been changed, and as a thing it could be sold or transferred. Indeed, so proprietary was the obligatio that both Ulpian and Julian say that it can be made the subject of a condictio; where a promise had been made without a causa the promisor could bring a condictio to reclaim the obligatio itself, not the sum of money that had been promised.
As well, since it could be conceived of as a thing, an obligatio could be split into parts or divided up. It followed therefore that part of an obligatio could be released, leaving the remainder of the obligatio intact. It was easy to deal with co-heirs, since it could be said that an obligatio owed by or to the testator could be divided between them, each having or being liable for a share. Exactly the same applied to co-ownership; when a promise was made to a co-owned slave, for example, the obligatio could be divided up among the co-owners. It is at this point that we can see the flexibility that is achieved by the use of the noun rather than the verb. No doubt it would have been possible to have reached the same results solely by using the verb, but it would hardly have been linguistically economical to do so.
Further than this, the use of the noun made it possible to formulate more sophisticated legal rules than would have been possible without it. Suppose Titius enters a contract to pay 100 to Maevius; there is now an obligatio for the 100. Subsequently he undertakes to pay the 100, a constitutum debiti; there is now a further obligatio to pay the 100, so that there are two obligationes to pay the same sum. Payment of the 100 will discharge both obligationes, but it is held by Ulpian that litis contestatio in an action on the initial contract did not destroy the obligatio arising under the constitutum debiti and vice versa. The two obligationes to pay the same sum of money can to that extent be treated as independent of each other. In the same way, a person could have a right under a will and under a stipulatio against the same person for the same thing: here again it could be said that there were two obligationes, each independent of the other.
6. Obligatio and Actio
The rather rough and ready definition of obligatio as a vinculum iuris found in Justinian’s Institutes, a tie of law by which we are of necessity constrained to pay some thing according to the laws of our civitas, brings into the foreground the relationship between obligatio and actio. If the effect of the obligatio was that one person could be forced to pay (or do) something by the beneficiary, this would be achieved by the beneficiary bringing an action against the person under the obligatio. To that extent, therefore, there was an intimate connection between obligatio and actio, the one being a corollary of the other.
So intimate was this link that book 44.7 of the Digest has as its title De Obligationibus et Actionibus.
Yet it is not quite accurate to say simply that obligatio is the correlative of actio. As Gaius states, the connection is between obligations and personal actions:
In personam actio est, qua agimus, quotiens litigamus cum aliquo, qui nobis vel ex contractu vel ex delicto obligatus est, id est, cum intendimus dare facere praestare oportere.
An action in personam is one in which we proceed against someone who is obliged to us either from contract or delict, an action, that is, in which we claim that he ought to give, do, or be responsible for something.
Moreover, it is only an intimate relationship, not an equivalence. The obligatio gives rise to the actio, the actio is grounded on the obligatio. There is, therefore, a separation in time between the two; the purpose of the actio is to obtain performance of the obligatio and thereby to dissolve it. This is an important aspect of the relationship between the two ideas. While the obligatio exists, as well as the legal relationship between the parties there is a social relationship of dominance and subordination. The inevitable period of time between the creation of the obligatio and its dissolution (whether it be by actio or by performance) marks the continuance of this social relationship. The obligatio might continue indefinitely, it might never be dissolved, thereby stretching the duration of the power relationship. It may even be the case that the debtor might sometimes be expected not to perform, since it might not be appropriate for the person in the subordinate position to determine when the relationship should end.
The establishment of this distinction between actio and obligatio was significant, for it enabled the Romans to develop a type of obligation that was not actionable; the so-called natural obligation, obligatio naturalis. Although no action lay to enforce the obligation it was nonetheless owed, so that no condictio indebiti lay to recover it back if it was in fact performed. Though the language of natural obligation might have been applicable to situations actionable under the ius gentium rather than the ius civile, its importance for our present purposes lies in its applicability to situations where there was only a social duty. Hence a freedman who performed day labour for his patron in the erroneous belief that he was under a legal liability to do so could not bring an action for the value of the services: although there was no liability at law there was a social duty—of- ficium—to perform. A woman erroneously believing herself liable to give a dowry does so; she may not recover it back since there was an underlying motive of pietas which still existed even when the mistaken belief was removed. A person who borrowed money could not reclaim it if he was sued for it and paid after litis contestatio but before judgement, since even if he was absolved in the action the natural obligation would have remained. Although there are insufficient texts to enable us to delimit the concept with precision, the idea of obligatio naturalis meant that it was possible for the Roman jurists to construct a non-legal sense of obligatio, giving secondary legal effect to such Roman virtues as fides, honor and pietas.
7. Obligatio and Contract
As has been seen, the words obligatio and obligare were derived from the root ligare, meaning to bind. In classical law the binding might stem from any source that generated a claim in personam. Its breadth and lack of focus are indicated in passing by Ulpian:
Si quis tutelam vel curam vel negotia vel argentariam vel quid aliud, unde obligatio oritur, certo loci administravit: etsi ibi domicilium non habuit, ibi se debebit defendere et, si non defendat neque ibi domicilium habeat, bona possideri patietur.
If a person has administered a tutelage, a curacy, business activities, banking, or anything else from which an obligatio arises, in a certain place, he will have to defend himself there even if he did not have his home there, and if he neither defends nor has his home there his goods will be seized.
More specifically it might flow from a delict, from the giving of a dowry, from a legacy or from a judgement.
Above all, an obligatio might flow from a contract, so much so that it could sometimes be treated as synonymous with contract. This is clearest in Ulpian’s description of the nature of contract as something which stemmed from an agreement:
Conventionis nomen generale est, ut eleganter dicat Pedius nullum esse contractum, nullam obligationem, quae non habeat in se conventionem, sive re sive verbis fiat: nam et stipulatio, quae verbis fit, nisi habeat consensum, nulla est.
The term agreement (conventio) is general, so that Pedius elegantly says that there is no contract, no obligatio, that does not have an agreement within it, whether it arises by delivery or words: for even the stipulatio, which arises by words, is void unless there is agreement.
Here, unequivocally, there is an identification of obligatio with contract; and even if the reference to obligatio is interpolated (and there is no strong reason to think that it is), the text shows that someone at some time believed them to be equivalent, and that this was not so outlandish that it was excised by Justinian’s compilers. While it is, of course, nonsense within the institutional structure to say that every obligatio is based on agreement—most obviously those obligations which arise out of delict have no element at all of agreement within them—there is nonetheless a close connection between obligatio and contract.
Tryphoninus, for example, in the context of a sale, writes of liability having arisen out of a contract or an obligatio of whatever sort (ex contractu et qualiquali obligatione a debitore interposita) on the part of the debtor. And when it is said that a slave acquires an obligatio for his master it must be a contract which is at the forefront of the mind; in other situations where a master might become the beneficiary of an obligation through the involvement of a slave—for example, where the slave was injured and the master thereby had a claim under the third chapter of the lex Aquilia—it would not really be appropriate to speak of the slave acquiring an obligatio at all. Moreover, it is not uncommon to find an obligatio identified with the contract which underlies it, as where Pomponius writes of an obligatio empti et venditi, or an obligation arising out of a loan is called an obligatio mutui, an obligation arising out of a mandate an obligatio mandati. Most common is the obligatio pignoris; so much so that obligatio standing alone might refer to a mortgage, where it is the land that is bound rather than just the person. The similitude is implicitly visible when obligationes round off a list of contracts: ab emptione venditione, locatione conductione ceterisque similibus obligationibus. In other words, we can point here to an equivalence between contractus and obligatio contracta.
8. Conclusion
This apparently close link between obligatio and contract may be more than an accident flowing from the survival of sources. In Akkadian and Hittite, for example, the abstract nouns cognate with the verb denoting the act of binding refer specifically to contracts or treaties; and in Hittite the active sense implicit in the verb form is carried across into the noun to the extent that it might refer to a treaty imposed on a vassal state and not (or not only) to one founded on an agreement. It may be, then, that in Latin too there was originally a focal sense of active binding behind the use of the noun, and hence an association with the voluntary act of contracting. Whether or not this is so—and it is no more than speculation—it remains the case that the Roman’s development of the obligatio in the sense of any legal bond between persons represents a major development in legal thought—one which survives almost unthinkingly into present-day legal discourse. [END]
The Source:
David Ibbetson, ‘Obligatio in Roman Law and Society’, The Oxford Handbook of Roman Law and Society, Oxford University Press 2016
Evolutions of social order from the earliest humans to the present day and future machine age.