The Ideas behind the Quasi Categories

Whoever it was who organised the residual miscellany into the two quasi categories, it is not easy to see what unities he had perceived or why he thought they could best be captured in this quasi terminology.

In relation to quasi-contract the first necessity is to eliminate a non­Roman distraction. Nowadays making our own exploration of the miscellany beyond contract and tort the unity which we think we can discover within it and thus remove from it, narrowing the uncharted residue, is the class of event generically describable as unjust enrich­ment at the expense of another. Mistaken payments are a species of that genus, which is recognisable whenever the law imposes an obligation, independent of the recipient's consent, to give up wealth received at the expense of another. This category of law, the law of unjust enrichment, is often hung on a Roman peg. For in D.50.17, which is �On various long-standing legal principles', there is this fragment from Pomponius, D.50.17.206 (Pomponius, 9 Various Readings):

Iure naturae aequum est neminem cum alterius detrimento et iniuriafieri locupletiorem By the law of nature it is fair that nobody should be enriched through loss and wrong to another.

Torn from its context, this is vague and meaningless except, and the exception is real and important, as a principle of the same open- textured kind as pacta sunt servanda (agreements should be kept) or sic utere tuo ut alieno non laedas (so use your own property as not to harm another's). Such principles provide a child's-eye view of areas of law. �Open-textured’ is the fashionable language for them, but �over­simplified’ might be better if the honesty did not, wrongly, suggest that their work was of no importance to grown-up lawyers. Masses of cases and statutes thrown together without attention to any child’s-eye view of what is going on lose their coherence. The first sign is that the area in question becomes very difficult to learn, except by heart. So the generality of Pomponius’s principle does not mean that it is useless. But from that alone we cannot infer that the Romans did use it. For different systems can put to conscious use different open-textured principles. And really there is no evidence that the Romans did create and organise a legal category under the principle against unjust enrich­ment. They had a great deal of law and litigation about the phenom­enon of enrichment-to-be-reversed. The condictio proves it. But in the divisions which they chose to make, �unjust enrichment’ did not find an independent place. Otherwise we might have expected to find some such statement as this: every obligation arises from a contract, from a wrong, from an unjust enrichment or from some other causal event. There is no such statement.

It is certain that the Roman category of quasi-contract included many figures which do belong in the category of unjust enrichment if that unity is once perceived and used. But it also includes other matter. For example the man who intervenes in the affairs of another (nego­tiorum gestor) is under a duty to perform his intervention carefully and must pay for loss which is attributable to want of that care. The actio negotiorum gestorum directa lies to enforce that duty.

So the Roman category of quasi-contract is not the category of unjust enrichment under another name. This is simple enough. But we make it very difficult for ourselves by using the term �quasi-contract’ as an unexamined synonym for unjust enrichment. In other words we borrow the Roman phrase to denominate a category of our own. But, worse still, all this is done without the clarity which comes from close attention. For the land beyond contract and delict is not much visited.

All this has been negative. Whatever our usage of the noun made from quasi ex contractu, the Romans did not see this category of obliga­tions as generated by unjust enrichment. How then did they see it?

The difference from contract is the absence of agreement. These obligations are not incurred voluntarily, by being undertaken. They are imposed willy-nilly when the event happens. In that point they resem­ble delictual obligations. But they differ from delictual obligations in that the events which trigger them are allowed or even encouraged, not forbidden. So, the quasi-contractual events are those which are lawful but bring into being involuntary obligations.

That is not quite all there is to it. It is also true that for each quasi- contractual event there is, in terms of its consequences a close analogy to one of the contracts in the Roman scheme. This analogy can in the majority of cases be identified as something more specific than an unanalysed impression or feel.

In the case of the other two, involuntary common ownership and legacy, it is impossible to assert that they share an action and an actional regime with a nominate contract. But the legatee's actio ex testamento (on a will) is a close cousin to the condictio. And common ownership, though giving rise to actions different in structure from the action on partnership, can claim an analogy to partnership of the less technical kind and one which is underpinned by Gaius's account of the history of societas as an imitation of joint heirship.^[99]

So the members of the category share the qualities of lawfulness and involuntariness with respect to the obligation. But to justify the name �quasi ex contractu there is also an analogy to a contract in the list of contracts. The analogy does not make the quasi-contractual events into �sort of contracts’. It is an analogy of consequence: although there is no contract at all, nevertheless once the thing is under way there is either or both a loose factual or behavioural similarity and, at least in three cases, a close actional similarity, or indeed identity, to a named contract.

As was said earlier, if it were not for the contrast between the two different quasi categories you probably would not look for these simi­larities. It is because some of these miscellaneous obligations are quasi ex delicto rather than quasi ex contractu that you are driven to justify the name. Otherwise quasi ex contractu, or quasi ex delicto, would suffice to explain all cases neither contractual nor delictual. Or, more accurately, suffice to conceal by means of a fiction the absence of explanation.

Two small points are worth adding. First, a guardian or intervener incurs liability for bad management. Hence his case clearly offers the classifier a choice. Shall he be classified on the basis of the lawful act of becoming a guardian or intervening in another’s affairs or on the basis of the wrong of bad management or bad intervention? The choice of the former is not quite an accident. We know that in classical law the model formulae of these actions recited the entry on task, not the default after entry: �whereas he intervened’, not �whereas he managed badly’. According to the structure of the formulae the obligations were con­templated as flowing from the primary, not the secondary event. The choice was made, therefore, when the formula settled down, not when the business of classifying was done.

Secondly, it ought not to escape notice that among the non­contractual condictio events in the family behind mistaken payment there are some which encounter difficulty with the word �lawful’. The condictio given after theft and that given on account of disgraceful behaviour by the recipient are both difficult to explain as quasi- contractual if lawfulness is essential, as it seems to be. There are two ways out. Either to say that it is after all no accident that the list of obligations quasi ex contractu does not expressly consider every event in the condictio family. Or to say, by a refined but not impossible argument, that these apparently unlawful events can be admitted on the ground that, though they do involve unlawfulness, it is not qua forbidden acts that they trigger the condictio.

That leaves quasi-delict. Nobody has ever come up with a view which commanded universal acceptance. There is a difference from quasi-contract. All the quasi-delicts are events which the law forbids: no litem suam facere, no pouring or throwing from houses, no placing or hanging above public ways, no stealing in ships, inns and stables. But, if this suffices to make the line between the quasi categories, the next question is, Why are these figures not delicts? What is �quasi’ about them?

The judge tends to spoil almost every attempt to answer this ques­tion. The other three all show a number of features which could provide the distinction. They are all cases in which the liability is vicarious, the occupier, innkeeper and so on being made to answer even for the acts of others. They are all cases ofliability without personal misfeasance, liability which can extend to pure omissions. They are probably all cases of liability without proof of any fault in the defendant himself. Attempts have been made to fit the iudex, at the head of the list, into conformity with one or other of these features. It is difficult to make his liability look vicarious, except where he wrongly absolves. In that case, but not in wrongful condemnation, he can at a stretch be seen as substituted for another, the party whom he let off. Again, if you look hard you can find evidence that the iudex could be liable for not doing. That is, one means of litem suamfacere was failing to turn up.

Stein has argued that strict liability does provide the answer, even for the iudex.8 This is in one way much the best suggestion. It gives a convincing picture of the line between quasi-delict and delict. The two together form one super-category of unlawful events triggering invol­untary obligations. Within that there are events involving fault, or blameworthiness, and events which if they happen impose liability independently of fault. Delicts involve fault, quasi-delicts entail strict liability. That is attractive. Because there really is some difficulty in treating events of strict liability as �wrongs’.

⁸ Peter Stein, â€?The Nature of Quasi-delictal Obligations in Roman Law’, 5 RIDA (1958), 563, 563ff-

But this probably will not work. Partly because Justinian seems to lean over backwards to find and point to traces of blame even in the three figures other than the judge. He would hardly do that if the whole point of the category was that it was based on no-fault liability. And partly because it is difficult to restrict the content of the judge's wrong in such a way as to make a no-fault liability plausible. Strict liability for wrong judgements is unimaginable. Nobody would do the job.

One feature which is common to the quasi-delicts is that they all involve defendants in special positions. You have to become a iudex before you can litem suamfacere, and so on. They all involve breaches of duties imposed not just on people generally but specifically on judges, occupiers, innkeepers. At first sight this seems to have no jurispruden­tial weight. In short, so what?

However, I think the answer does lie in this direction. The business of elegant classification is delicate. The classifier may have been worried about the special positions as having some potential for de-stabilising the category of delict. An argument could be made that they could and should be classified according to the first, lawful event (becoming a judge, becoming an innkeeper), not according to the second event, the litem suam facere or the inn-theft. A separate category for special-position-wrongs would effectively isolate them and antici­pate all possible attempts at such destabilisation.

If this is right the law of obligations was intended to be divided according to obligation-creating events, in this way. On the one hand lawful events, themselves divided between undertakings (contracts) and other events not connoting voluntary acceptance of the obligation (quasi-contracts). On the other hand, unlawful events. These might have been left in one class or might sensibly have been divided between events involving moral blame and events of strict liability. But in fact they seem to have been divided between forbidden events which anyone could commit (delicts) and forbidden events which could be committed only within the context of a special relationship (quasi­delicts). This classification has influenced both civilian systems and common law systems, though Austin did not like it and warned the common law against it.^[100] Contract and delict remain, under some critical pressure, dominant categories of our legal thought. The miscel­laneous residue is still troublesome, still not much understood. The quasi categories have certainly not settled the way in which it should be seen.