Back at the beginning, in the section on the conceptual map, we noticed how Gaius divided obligations into two categories.1

They arose either from contract or from delict. And then he found that this simple dichotomy would not quite do. There were obligations which arose from neither of these two events.

Residual miscellanies of the kind which Gaius uses here are devices to protect categories which have been successfully identified. There is not necessarily any need to give up contracts and wrongs as useful, coherent and manageable classes just because, contrary to first impres­sions, it turns out that there are obligation-triggering events other than these two. �Others' allows the ground which has been gained to be held despite the set-back. It admits that there is some unmapped territory beyond what has been reduced to order. But it also constitutes a challenge. Nobody is going to be happy for long with leaving territory unmapped. Attempts will be made to map it. That is, people will look into the miscellany to see whether another useful category of events can be taken out of it, so as to leave a smaller unruly residue. Or, even better if it can be done, they will try to divide the whole miscellany into named classes, so as to make the map complete. For those with the right cast of mind the challenge is irresistible. And there is a temptation to cheat.

The next division puts obligations into four species: aut enim ex contractu sunt aut quasi ex contractu aut ex maleficio aut quasi ex maleficio (for they are either from contract or quasi from contract or from wrongdoing or quasi from wrongdoing).

This statement certainly purports to eliminate the miscellany altogether. It divides the entire category of �others' between quasi ex contractu and quasi ex maleficio. But at first sight this does not look very promising. Because the names are not informative. They do not tell us anything positive about the unities which are supposed to have been identified. We usually anglicise the quasi categories to �quasi-contract' and, ignoring the change from delictum to maleficium, �quasi-delict'. But these terms should not be allowed to encourage the belief that the Latin words mean anything as positive as �sort of contract' and �sort of wrong'. Their message is almost entirely negative. Quasi ex contractu means �as though from contract (scilicet when in reality there is no contract)' and quasi ex delicto means �as though from delict (scilicet when in reality there is no delict)'. So the first message in this terminology is the negative one, that the named event (contract, delict) is not present. Beneath that, there comes an implication which is positive but uncer­tain in content: although there is no contract (or delict), yet there is something unidentified which makes �as though from contract' suitable for some and �as though from delict' suitable for others. This suggestion of an unidentified resemblance to one or other of the main categories has its source not in the quasi phrases individually but solely in the fact that there are two quasi categories. If the whole miscellany had been renamed quasi ex contractu, there would have been no suggestion of resemblance to contract but only a negative assertion that the category- three obligations arose as though from a contract, though no contract was made.

some to the other quasi category, we have to infer that there is some similarity to justify that distribution.

We cannot really go further without knowing what specific events were placed in each quasi category. But there is one question which ought to be put first. Was it Justinian’s commissioners who invented the quasi categories, or was it Gaius himself ? It is a question which cannot be finally answered. In the Digest it appears from excerpts from the Nuggets that Gaius not only modified the dichotomy by introducing the third miscellaneous category, but that he also described some members of the miscellany as arising quasi ex contractu and others as arising quasi ex maleficio (D.44.7.5 = Gaius, 3 Nuggets). It is almost certain that this passage has to some extent been tampered with. And it is possible that the interpolations include the insertion of the quasi labels. My own view is that that is unlikely. There is considerable evidence that Gaius enjoyed and attached importance to accurate classification. It is consist­ent with this that he would himself have tried to do something with the miscellany. But it is not possible to be sure.