We have spoken so far of ownership as something which explains itself, but in order to understand the next point it is necessary to enquire a little more closely into its nature. The Roman law of classical times is dominated by the absolute conception of ownership which it has evolved and by the action through which this right is asserted, the vindicatio. Ownership, in the developed law, may be defined as the unrestricted right of control over a physical thing, and whosoever has this right can claim the thing he owns wherever it is and no matter who possesses it.
If I possess a thing and you own it, then all you have to do is to prove your ownership and I must give it up; it is not necessary for you to allege that I have done you any wrong. I, on the other hand, the possessor, have nothing to do but to sit tight and wait for you to prove your right; if you do not succeed in proving that you are owner, I remain in possession. Now this very clear-cut conception is not to be found in all, or even most systems of law. English law, for instance, has never known an action corresponding to the vindicatio, at any rate with respect to movables. The action by which an owner recovers his thing which has got out of his possession all allege that the defendant "unjustly detains" the thing or that the plaintiff lost it, that it came into the hands of the defendant and that the defendant has converted it to his own use; to use technical language, the actions for the recovery of movables all "sound in tort". Further, the wrong alleged is, strictly, not one to ownership, but to possession, for, if the thing be "bailed", e.g. lent, by the owner to another, it is the bailee who can bring the action against the third party in whose hands the thing is found, the bailor being originally confined to his rights against the bailee. Hence it can be said that "although the bailor was the owner, the sum of his rights as owner was originally his better right as against the bailee to get possession; for this better right to get possession was the only form of ownership which the mediaeval common law recognised". This idea of a relative right to possession as contrasted with the absolute Roman dominium, is common to other Germanic systems of law besides the English, and it indeed appears that the Greeks also knew only of a similar relative right.
It must however be emphasised that the Roman conception, which simplifies matters very considerably, is an achievement of the developed law; in its origin Roman law was, in all probability, similar to other systems. In the developed law, as we have seen, the conception of absolute ownership applies to all sorts of things, res nec mancipi as well as res mancipi, the only difference being that res mancipi need mancipation (or in iure cessio) if they are to be transferred, whereas res nec mancipi do not, but it has been suggested that at one time the distinction went further, and that res nec mancipi originally did not need mancipation and indeed could not be mancipated, for the simple reason that they were not capable of being fully owned, and that therefore the assertion of full ownership ex iure Quiritium which is a necessary part of mancipation, as of in iure cessio would not be applicable to them. The possession a man acquired in them would, no doubt, be protected by the law of theft, but not by an action in rem. This idea, however, has been very vigorously attacked, and indeed it cannot be maintained, at any rate for historical times, in face of what Gaius tells us of the form of a vindicatio in the legis actio period. Assertion of Quiritarian ownership was an essential part of the proceedings and yet Gaius mentions a column, a ship and a flock of sheep when giving examples of things which, as they could not conveniently be brought into court, were represented by parts for the purposes of the ritual claim. All these are res nec mancipi, and so, at the time of which Gaius is speaking, it was evidently possible to assert full ownership even in such times. Of course it may have been different earlier, but we have no evidence of a time when res mancipi could be owned whereas res nec mancipi could not, nor is it certain that res nec mancipi could not be mancipated, though there would hardly ever be an occasion for mancipating them, seeing that they could be alienated equally well by mere delivery of possession (traditio).
In any case the evolution of the absolute right of ownership was no doubt gradual; we have seen that one of its characteristic features is the position of the defendant in a vindicatio; he himself need not show how he came to be in possession, need not, in technical language, show title; so long as the plaintiff does not prove title he can keep the thing. But in the earliest form of vindicatio known to us this is not yet the case. If we look at the form of the legis actio per sacramentum in rem as given by Gaius we see that not only the plaintiff asserts that the thing in question is his, but the defendant makes a similar assertion; then the plaintiff asks the defendant on what grounds he makes the claim. It is clear therefore that the defendant could not simply rely on his possession, but had to show some title, and it is further clear from what Gaius goes on to say that the defendant could not even rely on retaining the interim possession of the thing, i.e. until the action was decided, for the praetor gave this to "one or other of the parties", i.e. might give it to him or to the plaintiff. All this is quite different when we come to the procedure of classical times.
----------
Source:
Historical introduction to the study of Roman law, H. F. Jolowicz, pages 142 - 144.
