Arra

Several times we have encountered this word. It means �deposit’, in the sense of �earnest’ as opposed to the sense of a deposit for safe-keeping. The account of Justinian’s legislation in relation to documents is incomplete without some description of his use of arra.

What is the matter in dispute? It is convenient first to be able to draw on a phrase which will still be unfamiliar: �the arra rule'. The arra rule, derived from Greek law, is that where a party deposits an arra and subsequently withdraws from the transaction which the deposit was to sanction, he forfeits all claim to its return; and, from the other side, where a party has received such an arra and subsequently withdraws, he must return the arra twice over, so as to put him in the same position as he would have been in had he been depositor rather than depositee. In short, the arra rule is: in case of withdrawal, depositor forfeits, depositee repays double. The question in dispute is whether Justinian intended to introduce the arra rule solely to cover the pre-contractual stage which he created, for written contracts, between agreement and final docu­mentation. Or did he mean the arra rule to displace the normal regime for sanctioning executory contracts even after the moment at which, according to the classical rules as modified by his reform in relation to writing, they acquired legal effect? The narrowest view: he said that, where the contract was to be in writing, if you wanted a sanction at the pre-contractual stage between agreement and writing, you could give and take arra and rely on the arra rule.

The cause of this dispute is a clash between what is said in the Codex and what is said in the Institutes in relation to the contract of sale. It arises in relation to the part of Justinian’s enactment which follows that which we have already looked at. So far as there is a clash, we ought not to forget that the Institutes is an elementary textbook. That is, we should not give it primacy in resolving the difficulty. Nevertheless I shall give the passage from the Institutes first. For it is that passage which seems to do something drastic, to support the larger of the two views set out earlier. I will put my own cards on the table. I think the right view is the narrow one. J.3.23 pr.:

... Donec enim aliquid ex his deest, et poenitentiae locus est et potest emptor vel venditor sine poena recedere ab emptione.

... For as long as any of these are lacking there is both a locus poenitentiae [time for changing one’s mind] and the possibility for either buyer or seller to withdraw from the purchase without sanction.

So far so good. There is no doubt at all that �any of these' which may be lacking are the steps required to make a written contract final, for this passage follows on from the description of the way in which a docu­ment is finalised. There is a bit of a puzzle as to why it was thought necessary to say �both locus poenitentiae and possibility of withdrawal sine poena' which repeats the same point twice. But this is hardly more than an infelicity. Perhaps it is due to the need to emphasise. For there is a problem in the offing. What if, while the documents are being drawn up, one party withdraws and then, after the notary has released them, the other says that they have retroactive effect? �Now we have final documents your attempt to withdraw is no help to you'.

Ita tamen impune recedere eis concedimus, nisi iam arrarum nomine aliquid fuerit datum: hoc enim subsecuto, sive in scriptis sive sine scriptis venditio celebrata est, is qui recusat adimplere contractum, si quidem emptor est, perdit quod dedit, si vero venditor, duplum restituere compellitur, licet nihil super arris expressum est.

However, we allow them to withdraw without sanction only if nothing has already been given by way of arra. If there is an arra-giving behind the transaction (hoc enim subsecuto), then, whether the sale has been celebrated with or without writing, the one who refuses to complete the contract loses what he gave if he is buyer and is obliged to restore double if he is seller. And this rule applies even if nothing is said about what is to happen about the arra.

There are no surprises in the first part. The absence of sanction for withdrawal before final documentation is qualified: there will be a sanction if there has been arra given. If the second part went on to give shape to that sanction and no more, there would be no problem. For contracts in scriptis, between agreement and final writing: with­drawal sine poena, subject to the arra rule if arra is given.

But the second part is not so simple. It undoubtedly says that the arra rule is to apply sive in scriptis sive sine scriptis venditio celebrata est, whether the sale is �celebrated' in writing or not. For example, the text does certainly contemplate that if arra is given under a straightforward sale by word of mouth (binding by the classical rules and unaffected by Justinian's intervention in written sales) the arra rule is to apply. Similarly in a written sale already finalised (binding under Justinian’s own rule).

That is all right. It is inelegant in that it suddenly reaches beyond the particular problem in the reader’s mind—the space before the final writing in contracts in scriptis—but it does not do any substantive damage to the larger picture to which the reader has to re-focus. For the arra rule can be grafted on to the normal regime for sanctionary contracts, an extra sanction if arra happens to have been given.

Then the bombshell. Taken as a whole, does not the passage say that withdrawal impune is always possible so long as arra has not been given? Does it not say that in sales written and unwritten the only sanction is the arra rule if arra has been given? If so, that is something quite incompatible with the â€?normal regime’. Suppose an unwritten sale. We meet in town. We agree that I will sell and you will buy the 10,000 litres of wine in my vats for £5,000. The normal regime says that we are now bound and can only â€?withdraw’ on pain of paying damages. Does this text say that we can impune recedere (withdraw without sanction) if only I have not insisted on your paying over an arra of, say, £2,000 so as to attract the sanction of the arra rule?

It all depends on that wretched colon, which in the English I changed into a full-stop. The punctuation of the Latin is modern. It is not to be relied on. What is the effect of the colon? It makes everything which comes afterwards, hence the very wide statement of the arra rule for all kinds of contracts in scriptis and sine scriptis, seem to be an expansion of what comes before. And what comes before is �We allow withdrawal without sanction if arra is not given.’ So the expan­sion is all about the system which can provide a sanction, against the background of impune recedere. The effect of the colon is to make free withdrawal the general rule.

Without the colon the impune recedere just looks back to the pre- contractual space before the finalisation of documents.

If we now turn back to the enactment in the Codex we will not find anything which supports the wider interpretation of what is said in the Institutes. We have already seen Justinian providing that, when the contract is to be in writing, nobody is to claim any right from any preliminary document and then going on to rub in the change in huiusmodi venditionibus (in sales of this—i.e. written—type). He goes on (C.4.21.17, continued):

Illud etiam adicientes ut et in posterum si quae arrae super facienda emptione cuiuscumque rei datae sunt sive in scriptis sive sine scriptis, licet non sit specialiter adiectum, quid super isdem arris non procedente contractu fieri oporteat, tamen et qui vendere pollicitus est, venditionem recusans, in duplum eas reddere cogatur, et qui emere pactus est, ab emptione recedens, datis a se arris cadat, repetitione earum deneganda.

And we also add this, that hereafter if, in relation to a purchase which is to be made of any res whatsoever, some arra has been given whether in scriptis or sine scriptis, then, even if it has not been expressly said what should happen in relation to the arra if the contract fails to go forward, yet he who has promised to sell shall on repudiating the sale be compelled to restore double, and he who has agreed to buy shall on withdrawing from the purchase forfeit the arra given by him, its recovering being refused.

The last part of this just lays down the arra rule. It raises no problems. The first part expresses the range of the provision. It does contain some puzzles. The order is odd. What does �whether in writing or not in writing' go with? It might be linked to the facienda emptio cuiuscumque rei or to the arrarum datio. It is closer to the latter, except that the notion of dare in scriptis is a bit strained.

If that is right, the provision is about the case in which arra is given in emptio-venditio (written and unwritten) and it says that the arra rule is to apply. It does not say that without arra there is no sanctioning regime at all. On the other hand, if I am wrong and the clause goes with the arra giving, then the provision is about �purchases to be made'. Nothing turns on the use of �purchase' instead of �sale'. If facienda’ is taken in a loose lay sense, the scope is simply �all and any case of emptio-venditio in which an arra is given'. And then the effect of the provision is very much as before, with the addition of the careful statement that once the arra is given it does not matter whether it is given in writing or without writing. However, lastly, if �facienda’ is taken technically so as to mean strictly that the emptio is still imperfect, still to be made binding, then the scope of the provision is only �contracts of sale about to be made' and that would certainly suggest that it was intended to focus only on those which, though to be made in writing, are, pending finalisation, still in their pre-contractual phase. So on this view the scope of the provision would be narrower than appears from its re-telling in the Institutes. But I think this third possibility is the least likely. �Emptio facienda' seems to me to indicate a purchase �to be performed' rather than one �to be concluded'. However, even if this third view were right, what would be the result? It would open a gap between Codex and Institutes, but it would do nothing to support the view that the sanction for an executory contract had become arra or nothing.

The simplest summary seems to be this. Justinian introduced a rule which said that if a contract was to be in writing it had no force till finally written. And he said what �finally' meant. To cover the period up to that finalisation you could rely on arra if you chose to. Also, you could rely on arra after the finalisation, or indeed in unwritten con­tracts. If you did, the arra rule would apply to you. If you did not, the normal regime of damages would sanction the contract. Presumably, in a case in which arra had been given under a binding contract, breach of which would give damages, you could only recover damages if and so far as they exceeded what you obtained by virtue of the arra rule.

4.