Internal Organisation: How Are Obligations Arranged?

We have been talking about the category of obligations as a single lump and looking at its place in the whole law. At this point we pass inside the category itself and look to see how it was divided into manageable parts.

It is useful to say at the start that there are several different ways in which obligations can be classified. The Roman experience, and also the Scottish and the English, show signs of a competition between two in particular. one is classification by reference to the content of the obligation. According to this, obligations to give a certain thing or certain quantity of things can form a distinct class—certum dare obliga­tions. Another class in the same series might be all obligations �to do' as opposed to �to give' (facere rather than dare). But of course the fact that one decides to classify obligations by content does not immediately determine what the divisions shall be, for content itself can be divided in different ways and at different levels of generality. The other system of classification is by causative events. That is, by the events which bring the obligation into being. Again there are choices to be made within such a system: How shall the events be classified?

i. The classification of obligations used by Gaius

He starts with a simple dichotomy, which soon gets into trouble. G.3.88:

Nunc transeamus ad obligationes. Quarum summa divisio in duas species diducitur: omnis enim obligatio vel ex contractu nascitur vel ex delicto.

Let us now move on to obligations. The principal division of these puts them into two species: for every obligation arises either from a contract or from a delict.

This is obviously a classification by causative events: omnis obligatio... nascitur. The question, literally, is, How are obligations born? And they are divided according as they come into existence by virtue of a contract or a delict.

This neat proposition encounters the proof that it is not exhaustive, and that it is therefore wrong, within three paragraphs. At G.3.91 Gaius encounters the obligation, undoubtedly recognised by the law of his time, to repay money received from one who mistakenly believed that he owed it. The event in question is receipt of a payment not due, non debitum (or indebitum) accipere. Of the obligation arising from that event Gaius rightly observes:

Sed haec species obligationis non videtur ex contractu consistere, quia is qui solvendi animo dat magis distrahere vult negotium quam contrahere.

But this type of obligation appears not to exist by virtue of contract. For one who pays with the intent to release himself wishes to discharge a deal rather than to sew one up.

He does not say that the obligation also does not arise from wrongdoing (delict) but that is quite obvious. So it follows that the event indebitum accipere is a non-contractual and non-delictual event which falsifies the simple dichotomy between contract and delict. And as we shall see there are others which also do not fit the two-term classification.

Gaius later adjusted to this difficulty. Before seeing how, it is useful to ask one more question about G.3.91. Why was Gaius brought into contact so quickly with the awkward case of mistaken payment? often people look away from the deficiencies of their expositions. Gaius was not like that. But it was not merely intellectual honesty which led him to confront this difficulty. He was drawn into it. The reason was this. His classification in terms of causative events was, at one point at least, in conflict with the classification instinctively or inarticulately adopted in the list of actions. Actions in personam on sale, hire, theft, contempt, for example, all constituted packets of law focused on events, com­pletely in harmony with the classification which Gaius wanted to use. But one very important action in personam was not. The Roman action of debt, the condictio, was based on a different unity.

Digest 44.7, �Concerning obligations and actions', begins with an excerpt from a book called Res Cottidianae (�Everyday Things'). �Day to Day Law' might better capture the sense of its name. It is also called Aurea (�Golden Law', or �Nuggets').

D.44.7.1 pr., Gaius libro secundo aureorum.Obligationes aut ex contractu nascuntur aut ex maleficio autproprio quodam iure ex variis causarum figuris.

D.44.7.1 pr. (Gaius, 2 Nuggets). Obligations are born either of contract or of wrongdoing or, by virtue of some particular law, from a variety of types of causes.

That last bit, not so easy to translate, is really just adding �other miscellaneous events' as an exhaustive third category after contract and wrongs. And that is not only wonderfully honest but also won­derfully useful. it takes the pressure off the two nominate terms and allows them to be defended even though they do not cover between them all the causative events from which obligations arise. The ground gained by the dichotomy in the institutes is thus not lost. But nor is it retained at the cost of sacrificing truth, running away from known error. That never does any good. Notice that the word for wrong has changed too, no longer delictum (from delinquere) but maleficium (from malum + facere, bad doing).

We are not going to chase on through D.44.7. If we did have time we would find that Gaius seems to do to the residual miscellany something which is done to it in Justinian's Institutes. It may have been done by Gaius or it may have been attributed to him by inter­polation. We will go straight on to Justinian's statement.

J.3.13 is Justinian's introduction to obligations. We have seen it begin with the vinculum iuris definition. Then it says, at J.3.13.1-2:

1. 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. 2. Sequens divisio in quattuor species deducitur: aut enim ex contractu sunt aut quasi ex contractu aut ex maleficio aut quasi ex maleficio. Prius est, ut de his quae ex contractu sunt dispiciamus...

1. The main division between all obligations puts them into two genera: for they are either civil or praetorian. The �civil' ones are those set up by statutes or indeed recognised by the ius civile.

2. The next division makes four species: for they are either from contract or as though from contract or from wrongdoing or as though from wrong­doing. And we must look first at those which arise from contract...

This passage does two things. First, it asserts that there is such a thing as a praetorian obligation. Gaius had simply assumed that obligatio was a ius civile phenomenon with which praetorian forms of action had nothing to do. Without wanting to push the analogy too far, this is rather like, in England, aligning the linguistic usage of common law and equity, driven apart by centuries of separate development. Then it turns the threefold classification with which we left Gaius into a division by four.

It resolves the miscellany into two quasi categories. As we shall see the obligation to repay an indebitum becomes a central example in the category quasi ex contractu. But I am not going to discuss the rest of the content of these quasi categories here. They will have a section of their own right at the end. We can do contract and delict safely so long as we remember that they do not exhaust the whole category of obligations. We do not need to think about ordering the miscellany, if miscellaneous the residue really is.

ii. The classification of obligations in Stair

I want to finish by looking at Stair's partial departure from the Roman scheme. Stair based himself on the pattern of Justinian's Institutes, but not in the sense of borrowing it lock, stock and barrel. He accepted it with a characteristically critical and rational independence. In the result the pattern of the Institutes is often departed from, for good reasons given. This is the subject of Professor A. Campbell's David Murray Lecture for 1954, �The Structure of Stair's Institutions'.^[12] This is not the place to look at all the differences.

1. The same right, as it is in the creditor, it is called a personal right, but as it is in the debtor, it is called an obligation, debt or duty, which is retained as the more proper name, Inst. de obligationibus inprin. [Author's note: This is Stair's mode of citing J.3.13 pr. and is, I think, forward looking to the following sentence, notwithstanding the punctuation.] Obligation is a legal tie by which we may be necessitate or constrained to pay, or perform something. This tie lieth upon the debtor; and the power of making use of it in the creditor is the personal right itself, which is the power given by the law, to exact from persons that which they are due.

2. obligations by the Romans are distinguished in four kinds: in obligations ex contractu, vel quasi ex contractu, ex maleficio, vel quasi ex maleficio. Which distinction insinuates no reason of the cause or rise of these distinct obliga­tions as is requisite in a good distinct division; and therefore they may be more appositely divided, according to the principle or original from whence they flow, as in obligations obediential, and by engagement, or natural and conventional.

Pausing there we can see that Stair's main classification, his summa divisio, is between obediential and conventional. �Natural' is offered as a synonym for �obediential' as is �conventional' for �by engagement'. The contrast is between obligations imposed willy-nilly and obliga­tions incurred because willingly accepted.

Whether Stair's criticism of the four Roman categories is thought fair depends to some extent on whether it is understood as focusing on the quasi categories or on all. It is certainly true that the two quasi categories have no explanatory force in the names which they bear. The same is not true of contract and wrongdoing.

The summa divisio does not indicate the extent of Stair's departure from the Roman scheme. it is slipped in at a high level. Below it comes a line of sub-categories represented by Titles 4-9 and Titles 11-17 of Book I. Titles 4-9 are the species of obediential obligations: �4. Con­jugal Obligations; 5. Obligations between Parents and Children; 6. Obligations between Tutors and Curators, and their Pupils and Minors, between Persons Interdicted and their Interdictors; 7. Resti­tution; 8. Recompense or Remuneration; 9. Reparation, where of Delinquence, and Damages thence arising'. Then follows Title 10, which is a general introduction to conventional obligations, called �Obligations Conventional, by Promise, Paction and Contract'. Titles 11—17 then run through the species of conventional obligations: �11. Loan, or Mutuum and Commodatum, where of Bills of Exchange;

12. Mandate or Commission, where of Trust, etc.; 13. Custody or Depositum, where also of Pledge and Hypothecation; 14. Permutation and Sale, or Emption and Vendition; 15. Location and Conduction, where of Annualrent and Usury; 16. Society, where of Co-partnery; 17. Accessory Obligations, where of Transaction, Caution, Oaths and other Accessories.'

All this looks very different from the Roman classification, but what is really different is exaggerated by the inversion of the Roman order. If one goes backwards the Roman picture comes back into focus. Thus Title 10, from which hang Titles 11—17, is, with its dependencies, the category of obligations ex contractu. Title 9, the last species within obediential obligations, is, in its second name, �of Delinquence'. And its content shows that it is indeed the category ex delicto. So Stair's scheme brings together, albeit in inverse order, the two main Roman categories, contract and delict. Then, proceeding further from back to front, Titles 8 to 4 contain much of the matter which the Roman scheme put into the miscellaneous residue, later broken into quasi­contract and quasi-delict. So it is not unfair to say that beneath a different summa divisio Stair uses a scheme which goes (i) miscellaneous, (ii) delict, (iii) contract, exactly the Gaian order in reverse. And one way of summing up his intention is to say that he was aiming at a better, more rational treatment of the miscellaneous group. Better, that is, than Justinian managed with his two quasi categories.

This short comment on Stair's scheme requires two modifications to save it from having misrepresented the picture. First, his five heads of the miscellany are not identical in content to the Roman matter. The chief difference is that he brings in obligations based on marriage and parenthood, matters which the Roman scheme dealt with in the law of persons. Stair dispenses with �persons' as a separate division of the law. Secondly, Stair's category of delict is so named only secondarily. Its full name is �Reparation, where of Delinquence, and Damages thence arising'. It is reparation first and then delict (delinquence). And repar­ation forms the last of a run of three—restitution, recompense (or remuneration), reparation—in which the unity is made to depend on the content of the obligations: obligations to make restitution, to make recompense, to make reparation. But in this run reparation at least turns out to be a category with two unities. It is not all obligations to make reparation however arising, but obligations arising from delict which concurrently are obligations to make reparation.

We saw that Stair adapted the Roman scheme to try to improve it and we began with criticisms made by him. At this point we may venture two criticisms of his attempted improvement. First, he allows himself to do something which a perfect classification should always avoid: he varies the basis of his categories, some contentual (restitution, recompense), some contextual (between parents and children, between tutors and pupils), some causal (contract). There are dangers in this. Just as there would be dangers for a naturalist who started by classifying animals as mammals, reptiles, birds and carnivores. That is an obvious case. It serves to show what disasters are likely to happen when a classification turns a corner. All the worse when it is not obvious. The other criticism is of categories with double unities, like delict/repar- ation. They can lock you into a view which may be wrong. The category seems to affirm that necessarily at all times and places delicts give rise to reparatory obligations. That particular affirmation is certainly incorrect at least in the sense that it is not universal. It may have been true for Stair's own time. But the fact that Roman delicts give rise to obligations variously measured shows that it is unsafe (at all times and places) to assume a bond between one kind of event and one measure of liability.

We can end with a criticism of the Roman scheme which Stair did notice and which was picked up more seriously by Austin, the first professor of jurisprudence at London University. He held the chair from 1826 to 1835. In his Lectures, he observed that the dichotomy between contract and delict was defective.^[13] Quite apart from the fact that it fails to exhaust the category of obligations as a whole, it is unsatisfactory because it fails to cope with breach of contract, an obligation-creating event quite distinct from contract (i.e. the making of a contract). Stair had encountered this problem in Institutions 1.9.3. Austin's professed scheme was to make a distinction between �primary' and �secondary' (or �remedial') obligations. In this scheme breach of contract and delict stand together on the secondary tier because they trigger obligations to remedy wrongs, the wrongs being breaches of anterior primary obligations. contract and, for example, receipt of a mistaken payment belong up on the primary level, since they are events which trigger obligations without themselves being breaches of anterior obligations. If you promise to pay me money or receive a mistaken payment from me your obligation cannot be seen as a remedy or sanction for any other pre-existing obligation, though in its turn it may need such a remedy or sanction itself. That is why it is called primary as opposed to secondary. This Austinian distinction has lasting value. I think Gaius would have said that on the Roman material itself the primary/secondary distinction was not strictly necessary. But I think he would have agreed that a classification should, ideally, have validity independent of the accidents of time and place. And on that basis he would perhaps have incorporated a summa divisio between primary and secondary before coming to contract (primary) and delict (secondary).