Chapter 4 ‘Actions'

Gaius, in his discussion of the law of actions, begins with an account of the different types of actions found in Roman law of the classical period. We have already come across many of these in our discussion of the various branches of Roman private law.

Inst.Gai.IV.1.... If we should ask how many classes (genera) of actions there are, the better view is that there are two, real (in rem) and personal (in personam)... 2 A personal action is one which we raise when we sue someone who is under an obligation to us, either contractual or delictual. That is when we claim in our principal pleading that ‘there is a duty to give, to do, to fulfil'. 3 A real action is where we claim either that some corporeal thing is ours, or that some right is available to us, for instance a right of use or of usufruct.... 6 Sometimes we raise an action solely for restoration, sometimes solely for penalty, at other times for both a thing and the penalty. [translation: Gordon/Robinson]

After setting out the different types of action, Gaius proceeds to tell us about their history:

Inst.Gai.IV.11. The actions used by the old lawyers were described as actions in the law, either because they were set out in statutes, since at the time the Praetor's edicts, which introduced numerous actions, were not yet in use - or because they were precisely adjusted to the words of statutes, and were accordingly observed as if they had been statutes.... 30. But gradually all these actions in the law fell into disfavour.... Therefore, by the Aebutian Act and the two Julian Acts these actions in the law were swept away and the system of litigating by means of specially drafted phrases, that is by formulas, was introduced. [translation: Gordon/Robinson]

To understand Roman litigation, one must first appreciate that all litigation, whether in Roman or in modern law, is based on ritual.

Inst.Gai.IV.30. But all these legis actiones gradually became unpopular. For the excessive technicality of the early lawyers was carried so far that a party who made the slightest mistake lost his case. So by the Lex Aebutia [second century BCE] and the two Leges Iuliae [17 or 16 BCE] they were abolished, and litigation by flexible pleadings, that is by formulae, was established.

This discussion will therefore focus primarily on the formulary procedure as it existed in the classical period.

It is important to remember that for much of the classical period, the formulary procedure existed alongside a newer bureaucratic procedure (cognitio) which had been introduced during the early years of Augustus' reign. This process was initially confined to certain areas of the law as decreed by the emperor, but in time it came to merge with the formulary procedure so that by the end of the period under discussion it had almost replaced it entirely. The official abolition of the formulary procedure in the first half of the fourth century CE must also have affected the way in which the idea of the formula was approached by the courts. Take the following statement:

C.2.57.1. Emperors Constantinus and Constans to Marcellinus. Legal formulas, with the snares of their syllables laying a trap for the acts of all shall be completely abolished. Given January 23 (342). [translation: Blume]

What does this mean? It is difficult to tell. It is interesting to note, however, that a similar justification was used for the abolition of the actions-at-law in favour of the formula during the reign of Augustus. Why did the formulary procedure go into decline? The reasons are twofold. First, the expansion of the imperial bureaucracy, which favoured this form of procedure, caused it to be used more widely and in more areas of Roman private law. In second place, the formulary procedure was never applied in the provinces of the Roman Empire. In most provinces which had been created after the fall of the Republic, the bureaucratic process was the standard form of civil procedure (except where the local legal culture was so sophisticated that the Romans left it intact). For the purposes of our discussion, therefore, attention will chiefly be paid to the formulary procedure with certain comments about the bureaucratic process.

Before looking at the procedure, let us first look at the court itself.

In order to start litigation in Roman law, the plaintiff (the person wishing to litigate) had to summon the defendant to appear in front of the praetor. This could be done in a variety of ways, but the underlying notion, inherited from the ritualistic procedure of the Twelve Tables, was that summons was essentially a private process.

Inst.Gai.IV.183. Finally note that someone who wishes to take proceedings against another person must summon him at law (in ius vocare) and that the person summoned incurs a penalty under the Praetor's Edict if he does not come [translation: Handouts]

In second place, and related to the first, the defendant's appearance could be secured using a formal promise backed up by security (usually a sum of money). This promise was called a vadimonium, and if the defendant contravened the promise he would forfeit the security. The formal promise was commonly used where a matter had been deferred from another court. It is important to remember that the formal promise did not replace the act of summons. The two procedures operated in tandem. The formal promise was merely a means to secure the presence of the defendant and we know from examples of it that the parties often agreed to appear in the vicinity of the court near a landmark, say a statue of the emperor in the forum. Once they had met at the appointed place and time, ritual summons could take place.

Once this had been done, the parties could then proceed to the court of the praetor to commence the first part of the lawsuit.

Assuming that the praetor could see the parties on the day, a number of legal issues had to be resolved in his court before the matter could progress to a full civil trial in front of the judge. Some of these issues were procedural, others substantive, but to the Roman legal mind these two things were often the same. The first thing to consider was whether the parties could in fact appear before the praetor. The Digest title De postulando (concerning appearances in court) (D.2.4) contains an account of the section of the Praetorian Edict which effectively barred some persons either from being representatives for others or from appearing as litigants. Many of these restrictions related to the status of the person, namely either because they were tainted with infamy on account of a previous judgement or because they were alieni iuris. It is also interesting to note that in the Lex Irnitana (chapter 84), restrictions were placed on the type of lawsuit that could be heard at a local level. If, upon investigation by the local court, it was found that the matter had to be deferred to a different court, then the case was at an end and the parties made a vadimonium (a formal promise) to appear, say in Rome, to take the matter to the appropriate court.

Once it had been established that the parties could appear in front of the praetor, the legal basis of the claim then had to be investigated.

To sue under Roman law of the classical period, one had to show that one had a legal claim to do so. This legal claim had to be founded either in the original ius civile or in the more recent ius honorarium developed through praetorian intervention. By the advent of the classical period, one way of demonstrating that one had a legal claim was with reference to the Praetorian Edict:

D.2.13.1pr (Ulpian, Edict, book 4). Whatever action a person wants to bring, he must state its nature (edere actionem).... 1 The word 'state' (edere) includes the making of a copy [by the defender], or expressing the whole case in a written statement (Hbelius) and giving him it, or dictating it. Labeo says that a person states his case when he takes his opponent up to the praetor's album and points out what he is going to dictate, or tells him the one he wants to use. [translation: Handouts]

We have already seen that the Praetorian Edict contained stock 'formulae' which set out the skeleton of a legal claim and which could then be filled in by the parties with reference to the facts of the case. It was the main task of the praetor to investigate the matter to see whether the plaintiff did, in fact, have a legal claim based either on the civil law or on the Praetorian Edict. In more complex cases where the parties did not base their claim on existing law, the praetor also had to decide whether he was prepared to extend the existing law by analogy (thereby granting an actio utiHs) or by granting legal relief based on the specific set of facts presented to him (by granting an actio in factum). The formula contained all of the essential elements of the lawsuit. Let us look at its parts:

Inst.Gai.IV.39. These are the parts (partes) of a formula: a demonstratio, an intentio, an adiudicatio and a condemnatio. 40 A demonstratio is the part of a formula inserted at the beginning to show the subject-matter of the action, as in this part of a formula: 'Whereas (quod) Aulus Agerius sold the slave to Numerius Negidius'... 41 An intentio is the part of a formula in which the pursuer defines his claim, as in this part of a formula: 'If it appears (si pa ret) that N.N. ought to pay (dare oportere) A.A. 10,000 sesterces'... or again; 'If it appears that the slave belongs to A.A. by Quiritary right (exiure Quiritum)'... 43 A condemnatio is the part of a formula which gives the iudex power to condemn or absolve the defender, as in this part of a formula:... 'Condemn, iudex, N.N. to A.A. for a sum not exceeding (dumtaxat) 10,000 sesterces. If it does not so appear, absolve him' [translation: Handouts; A.A. = Aulus Agerius, is the stock name used by the Roman jurists to describe the plaintiff and N.N. = Numerius Negidius, the stock name for the defendant]

Having described the various parts of the formula and their functions, let us look at two examples:

Let Titius be iudex. If it appears that N.N. ought to pay A.A. 10,000 sesterces, if there was and is no fraud on A.A's part involved in the matter, condemn, iudex, N.N. to A.A. for 10,000 sesterces. If it does not so appear, absolve him. [translation: Handouts]

Let Titius be iudex. Whereas A.A. sold N.N. the slave in question, which is the subject of this action, whatever N.N. on that account ought to convey or do for A.A. in good faith (dare facere oportere ex fide bona) condemn, iudex, N.N. to A.A. for that sum. If it does not so appear, absolve him. [translation: Handouts]

Notice that these two formulae are different. The one starts with 'if it appears that...' and the other begins with 'whereas...'. This is because the nature of the claim is different. In the first case it is for a fixed sum, whereas in the second case it is left to the discretion of the judge to decide the amount of the condemnation. The text above describes the stock elements of the formula, but it should not be forgotten that the defendant could add his counterclaims or defences to the formula as well during the first stage of the lawsuit in front of the praetor (such as the defence of fraud which appears in the first example of the formula above 'if there was no fraud Once the praetor was satisfied that the plaintiff had a legal claim that could be framed in terms of an actio and both parties were satisfied with the claims and counterclaims or defences inserted in the document, a judge was chosen.

Notice how in the formula the name of the judge is inserted at the start. This is an indication that the formula as fleshed out by the parties in front of the praetor was in essence an instruction to the judge to decide the matter according to the formula. To be a judge in Roman law one needed to fulfil a number of requirements (mostly of property and status). Judges as a rule did not have expert legal training, but they were sometimes assisted by a counsel of experts and could also, as a passage from Aulus Gellius (Attic Nights 14.2) shows, take advice from philosophers and lawyers on a point of law. In the case of Rome, those who were eligible to become judges were fi rst examined by the praetor. If they were found suitable, they had to swear an oath that they would fulfil their duties to the best of their abilities. Once this had been completed, their names were entered into an album which contained lists of names. It was the prerogative of the parties to choose their own judge (the plaintiff starting first and the defendant having the right of veto) until the parties had settled on a candidate. If they could not, one was assigned to them by the praetor. Once a judge had been chosen, his name was entered at the top of the expanded formula. This then concluded the first stage of the lawsuit and it was said that 'joinder of issue' had occurred. The consequences of this moment for the civil suit are explained as follows:

C.3.9.1. The Emperors Severus and Antoninus to Valens An issue has not been brought to trial (in iudicum deducta) if there has only been an initial presentation of it or a preliminary hearing on the type of action involved. For there is a great difference between joinder or issue (litis contestatio) and the stating of an action (actio edita). Issue has been joined only when the iudex begins to hear the case as it is presented in the statement of the claim. (narratio negotii) [202 CE] [translation: Handouts]

That meant that the parties were now bound to take the matter to the second stage and were bound by the judgement of the court. After the lapse of a period of time, the parties would reconvene with the judge to begin the second stage of the lawsuit.

At this point, a few observations about representation are required. One feature of the formulary procedure is that the parties could be represented by others and therefore did not always have to be present in court. This could be useful, especially in cases where a case had to be deferred to Rome for a hearing. Representation was possible in both stages of the civil suit under the formulary procedure. The normal rules regarding the status of the representative applied, but specific attention has to be paid to the category of persons under the Edict de postulando who could not act as representatives in a court of law. Roman legal sources speak of two types of legal representative. The procurator (or general agent) was appointed by the party only informally. He had to provide security at the start of his appearance and, if the party whom he represented lost the case and was condemned by the court to pay an amount of money, the procurator had to pay it. The cognitor (or litigation agent) was appointed formally using the contract of mandate in the presence of another person. If a cognitor lost the case, the party who had instructed him was liable for the payment of the court judgement. It has been suggested that the distinction between these two types of legal representatives became blurred during the classical period.

On the appointed day, the parties and their representatives convened together with the judge to conclude the second stage of the civil suit. The aim of this stage of the lawsuit was to convince the judge, using the facts of the case, that he had to find either in favour of the plaintiff or the defendant. As one can see from the formulae described above, the judge had a rather narrow scope, he could only condemn or absolve (a third option - 'not proven' was available, but this is not mentioned in the formula). Since the judge did not have expert legal training, it is commonly said that this stage of the trial did not proceed using technical legal arguments (but this remains disputed). Rather, the representatives of the parties used rhetoric (the Roman equivalent of the law of evidence) to prove their client's case. Once both parties had brought evidence and had put forward their carefully constructed arguments using the conventions of rhetoric, the judge had to make a decision. Under the formulary procedure the court's judgement was always expressed in terms of money:

Inst.Gai.IV.48. The condemnatio in all formulae which have one is framed in terms of a money valuation. So even where the claim is for corporeal thing, such as land, a slave, a garment, gold or silver, the iudex condemns the defender not for the actual thing, as was the practice in early days, but for the amount of money he values it at. [translation: Handouts]

Where the lawsuit concerned the recovery of a specific object, a judgement in money would obviously not be preferable. Thus, to circumvent this problem, it was agreed that the parties could insert an 'arbitration clause' into the formula whereby the money judgement of the court would only be invoked if the party who had lost the lawsuit did not return the object.

Once the judge had fulfilled his task to deliver a judgement, he was no longer obligated to the parties. Take the following text:

D.42.1.55 (Ulpian, Sabinus, book 51). Once a iudex has given his decision (sententia) he ceases to be a iudex, and the law is that once a iudex has condemned someone for too much or too little, he cannot correct his decision. He has done his duty, for better or for worse, once and for all. [translation: Handouts]

The judge's decision was final and the only grounds on which it could be revisited were procedural as in the quasi-delict of the ‘erring judge' which, as we have seen above, most probably referred to procedural irregularities.

Once the judgement had been delivered, the losing party had a set time in which to comply with the judgement. If he disputed the judgement, the winning party could approach the praetor once more to obtain an action on the judgement (actio iudicati). The losing party would then be subjected to another lawsuit based on the non-fulfilment of the original judgement. If found guilty, he would be condemned to pay twice the value of the original judgement.

Inst.Gai.IV.9. We sue for what is ours and a penalty in those cases, for example, in which we claim for double damages against a defendant who denies liability. This happens in the case of a judgment debt (actio iudicati) [translation: Handouts]

If the party did not dispute the original judgement, the winning party could proceed with the enforcement. In most cases this would involve a seizure of the belongings of the losing party to recover the value of the judgement of the court. Such a seizure of assets did not give ownership of the goods to the winning party. It merely secured the assets until such time as a curator could be appointed to oversee the selling of the goods at public auction with prior notification of other creditors so that all of the debts could be recovered.

One final point needs to be made about the decline of the formulary procedure in favour of the cognitio. As mentioned above, the cognitio was a process which evolved out of the emperor's bureaucracy. Being a product of imperial bureaucracy, it comes as little surprise that it was from this procedure that a system of appeals arose during the classical period. Under the formulary process, the decision of the judge was final and could only be revisited on the grounds of procedural irregularity. Our sources do not show the existence of any system of appeal from local or regional courts to the court of the praetor in Rome, apart from the provisions contained in the Lex Irnitana whereby some actions (on account of the amount of money involved or the nature of the penalty) had to be heard in Rome. Under the cognitio, a bureaucratic process, the idea of an appeal to a higher authority was more compatible with the nature of the process and therefore it comes as little surprise that by the end of the classical period it had become established that decisions of local courts could be appealed either to the court of the governor of a province or to the court of the praetor in Rome. The final form of appeal was directly to the emperor, who by the middle of the classical period had become an important legislator through his imperial bureaucracy.

Suggested further reading

The following books contain a more comprehensive overview of the Roman law of procedure:

Bablitz, L. Actors and Audience in the Roman Courtroom (Oxford 2007)

Greenidge, A. The Legal Procedure of Cicero's Time (Oxford 1901) Harries, J. Cicero and the Jurists (London 2006)

Kelly, J.M. Roman Litigation (Oxford 1966)

Metzger, E. A New Outline of the Roman Civil Trial (Oxford 1997) Metzger, E. Litigation in Roman Law (Oxford 2005)

Powell, J. and Paterson, J. (eds) Cicero the Advocate (Oxford 2004)