Ancient Roman Law and Rhetoric

What was the role of rhetoric in all of their different courts and procedures?

By O.E. Tellegen


In his book on the art of rhetoric in the Roman world, George Kennedy states that, in Greek rhetoric, the judicial oratory had a position of leadership, and he assumes that this was also the case in Rome(1). Consequently, he dedicates a much larger number of pages to judicial oratory (1 1) in early Roman rhetoric than to deliberative (4) and epideictic oratory (2).

He begins by stating that the formulary procedure was the most characteristic feature of Roman procedure(2). I quote: ‘The magistrate before whom a case was initiated, usually a praetor, drew up a precisely worded formula or legal definition of the wrong alleged (“if so and so proves to be the case, then let the defendant be condemned”) and referred the case to a court whose function then became to decide whether or not the conditions of the formula were fulfilled.’ However, according to Kennedy, the formulary procedure did not encourage elaborate speech-making, and he hardly refers to it again in this chapter. As for the patrons or professional orators who pleaded other people’s cases: they determined the tactics and procedure of the case and, most important of all, its rhetorical presentation, but for technical legal advice, they depended on jurisconsults, professional legal authorities(3). In this description, Kennedy minimises the role of oratory in the formulary procedure, although that was the most important legal procedure in the later Roman republic and the early empire. Therefore, he undermines his own statement about the relevance of judicial oratory in Rome(4).

The reason why Kennedy minimises the role of oratory in the formulary procedure cannot be found in the sources but in the modern roman- ist literature to which he refers(5). For two centuries or more, Romanists (i.e., modern jurists studying ancient Roman law) have regarded law as an autonomous concept linked with the truth, and rhetoric as a way of escaping from, perverting, and concealing the truth. Following Plato, they think that jurists always try to find the absolute truth about law, whereas orators only go for probabilities by means of argumentation^’. They also applied this view to the two parts of the formulary procedure, in the following way. In the first part, which took place before the praetor, the formula had to be worded. This was an extremely technical process, which was dominated by the jurists. In the second part of the formulary procedure, which took place before the judge, the unus iudex, only the facts mentioned in the formula were at stake; it was dominated by the patrons or orators who would try with all means including lies, slander, and emotional appeals to win their client’s case. Therefore, in this view, only the first part of the procedure was important for the development of Roman law, and the role of oratory was hardly relevant at all.

To discuss the pros and cons of this view is beyond the scope of this paper*7). What I want to do here is point out that the sources provide us with much more complicated information about the role of the patrons in the formulary procedure. They show that the patrons were active in the first part of this procedure, particularly in the drawing up of the formula. Moreover, they show that, in their speeches before the judge, the patrons did not plead about facts only, but about the juridical interpretation of facts or words. In the light of these sources, therefore, it seems that the patrons did contribute to the development of Roman law. In this paper, I would like to assess the significance of these sources for the relevance of rhetoric in the formulary procedure.

Because Kennedy discusses judicial oratory according to its different aspects and not according to the different procedures, this chapter is sometimes rather confusing. Therefore, in order to make my point, I will first have to put this procedure in its proper context; I will give a general outline of the procedures that were used for civil and criminal cases, and assess what the role of rhetoric in these procedures was. Then I will concentrate on the formulary procedure, discuss the different parts, and show with the help of examples from Cicero and Justinian’s Digest that patrons played a vital role in the first and the second part of this procedure and therefore also in the development of Roman law. I hope it will then be clear that, in Rome, judicial oratory was as important in the formulary procedure as in the other procedures.

Legal Procedures and the Role of Rhetoric

Augustus as Pontifex Maximus (Via Labicana Augustus) / Photo by Marie-Lan Nguyen, Palazzo Massimo alle Terme, Wikimedia Commons

The oldest procedure for civil law cases was the trial by means of legis ac- tiones (literally, actions based on the law)(8). It was conducted orally and was divided into two phases: the first before the, pontifex who decided whether the litigants could take legal action, the second before a judge who was a private citizen and who, after hearing evidence, pronounced his judgment. In early Rome, this evidence was often obtained with the help of irrational, supernatural methods such as ordeal or the taking of auspicia by observing the flight of birds. The legis actio procedure was very formal, strictly Roman, and allowed litigation for a small number of legal problems only. In the third century BC, it was somewhat modernised, one of the changes being that a court of one hundred judges, the centumviri, was introduced to decide on claims concerning large inheritances and possibly also on disputes about the ownership of land and about guardianship. Emperor Augustus allowed the centumviri to continue even after 1 7 BC when he abolished the legis actiones as the regular procedure. In the 2nd century AD, they were replaced by the imperial courts.

Around 200 BC, a new way of civil litigation was created, the formulary procedure, which came to function side by side with the legis actiones. It also consisted of two phases, but was much less formal. The first phase took place before a pontiff or, by the end of the second century BC, before a magistrate, usually the praetor. It differed from the legis actiones in that the litigant parties no longer had to pronounce their claims orally but put them down in a written formula, and also in that litigation was no longer limited to a fixed number of legal problems. It was the praetor who decided whether a judge should be appointed to decide on a particular problem. The bulk of Roman private law has been created in this context, in the last two centuries of the republic; then, there was a formula for every case that could be brought to court. At the same time, legal experts had collected judgments and opinions on the interpretation of the formulas, and they continued to do so in the first two centuries of the empire. In the course of the 2nd century, the formulary procedure went out of use and the imperial courts took over.

e know very little about the way criminal cases were tried in the early republic. We do know that a distinction was made between serious offences that were regarded as harmful to the state (treason, murder) and lighter offences that were regarded as harmful to the individual citizen only (theft, defamation of character). For an offence of this second category, the injured party himself had to institute a civil procedure, i.e., at first by means of a legis actio and later per formulas . For an offence of the first category, every citizen was able to sue the perpetrator. There are various interpretations and reconstructions of the way in which these crimes were dealt with. According to Mommsen and Jones, the proceedings took place in the presence of a magistrate, and anyone who was condemned to death by him could appeal to the popular assembly(9). This view has been challenged in various ways. The most recent view is that of Santalucia, who maintains that only the popular assembly (in this case, the comitia centuriata) had the authority to condemn a person to death after he had been found guilty by a magistrate(10).

e are much better informed about the new criminal courts which were introduced for serious crimes in the 2nd century BC. Initially, these courts were appointed occasionally, but in 149 BC, the lex Calpurnia established the first court for a particular type of offence, namely the quaestio de repetundis; its task was to deal with cases of extortion by Roman governors in Italy or the provinces. A specially appointed praetor was in charge of the proceedings. In the following fifty years, other quaestiones perpetuae were established: for murder and poisoning, for forgery of wills and other documents, for high treason, for bribery during elections, and for embezzlement of public funds. As from 100 BC, new magistrates (tresviri capitales) were appointed to deal with less serious crimes committed by slaves, foreigners and possibly also Roman citizens from the lower social strata.

We know something about the procedure in these quaestiones because some of the speeches made by Cicero when he acted as advocate or prosecutor in such lawsuits have survived. Every respectable citizen, not only the injured party, could lodge a complaint against someone. If the praetor allowed the complaint, a jury was set up by drawing lots. The course of the lawsuit was determined by the parties and their advocates. The plaintiff had to prove that the defendant had committed the alleged crime. If he was unsuccessful he ran the risk of being accused himself because he had knowingly lodged a false complaint. If he was successful he received a considerable reward; for instance, if the defendant was sentenced to death and his property was confiscated, then the accusor was entitled to a share of this property. It is not really surprising that in time the system was abused and some people even made a profession of accusing rich fellow-citizens. These people became known as delatores. The defendant, on the other hand, had to prove that he was innocent. Finally, it was the jury who decided whether the accused was guilty or not and what the penalty would be, if it was not evident from the law.

Like the formulary procedure, the quaestiones perpetuae continued to exist in the early empire, but as from the time of Augustus, their work was first supplemented and soon taken over by the new criminal court of the emperor and his delegates. By the year 200, all quaestiones perpetuae had disappeared.

These new courts of Emperor Augustus were not introduced as such in Rome, but they originated in those provinces that, in 27 BC, had come directly under his control. There, the administration of justice was delegated to officials in the emperor’s service. In the course of the 1st and 2nd centuries, this example was followed in the senatorial provinces, then also in Italy and finally in Rome. A characteristic of this imperial administration of justice, commonly called cognitio extraordinaria, was that the trial consisted of one phase only and that judgment was given by the emperor or by an official acting on behalf of the emperor. The disputes that were settled by means of the cognition procedure could be very different in nature: not only could they be about matters concerning private law and criminal law, but they could also be disputes between citizens and government officials. Another difference is that the judge now played a much more active role; he could, for instance, conduct investigations on his own initiative and by any means at his disposal. Although this new system may have worked for the common Roman citizens, it was not acceptable for the senators: therefore, Emperor Tiberius allowed the senate to develop into a forum privilegiatum for crimes committed by persons from their own class.

What was the role of rhetoric in all these different courts and procedures? It will be clear that, in the primitive legis actio procedure, there was no room for rhetoric. It was only after the procedure had been modernized and the special court of the centumviri had been established that the art of oratory became useful in this type of procedure. The oldest and at the same time most famous example of such a trial is that of the causa Curiana which took place in about 92 BC. Cicero mentions it in several of his rhetorical works because of the exemplary pleas of the advocates, L. Licinius Crassus and Q. Mucius Scaevola, interpreting the testament in question according to the intent and the letter, respectively. In the criminal proceedings before the popular assembly, it may have been relevant for a person to be able to persuade the audience, but so little is known about this trial that it is impossible to draw conclusions with regard to rhetoric.

In the second century BC, two new procedures were established: the formulary procedure for civil law cases and the quaestiones perpetuae for criminal cases. It is obvious that the newly introduced art of rhetoric was useful when a Roman governor was sued for extortion by the inhabitants of his former province. Cicero’s speech against Verres may be the most famous proof of the importance of rhetoric for such criminal law trials. In the other cases that were tried in a quaestio perpétua, members of the upper classes were typical plaintiffs and defendants; it was essential for them to be able to speak well in public. I will deal with the relevance of rhetoric for the formulary procedure in the next section.

It is difficult to tell whether the cognitio extraordinaria offered possibilities for orators to show their talents, because most of the examples of or references to speeches that we have from the early empire were held in the old- fashioned courts of the centumviri and the quaestiones perpetuae, and in the senate. However, we know from the works of Tacitus, Fronto, Aulus Gellius, and others that, in this time, legal advocacy was booming. According to Parks, the very fact that free political oratory was suppressed by the imperial system meant that free oratory flourished the more in the forensic context, where the emperors were less likely to impede it(11). Therefore, the art of rhetoric must have been practised in the imperial courts as well.

Rhetoric in the Formulary Procedure


Kennedy, and many authors with him, thinks that, in the formulary procedure, only the second phase, before the judge, was relevant for rhetoric. It is usually assumed that the only thing advocates did was to plead their client’s case before the judge. However, in my view, the advocates were involved in the first part of the procedure as well. Therefore I will begin with what happened before the praetor and try to demonstrate with the help of a number of passages in Cicero that the advocates had to use their knowledge of rhetoric (and law, for that matter) from the very start of the trial. Then I will discuss the second part of the trial, before the judge, and try to show that rhetoric was very important not only for the advocates to help win their cases, but also, in general, for the development of Roman law.

The Advocates before the Praetor

4th century Praetor throwing the mappa to open the races / VRoma

According to the modern books on Roman law, the procedure by formula began when the parties that had a dispute put their problem to the praetor (or, in certain cases, to the aedile) requesting him to appoint a judge to settle their dispute. The praetor first had to check whether the parties were qualified to be a plaintiff or a defendant in a legal action on the basis of the facts they had submitted. If they did not qualify, the praetor refused the action; if they qualified, he consulted the edict. In this edict, which he had issued on the basis of his ius edicendi at the beginning of his term of office and which, in principle, was applicable for the whole year, he had collected all the legal remedies that he was prepared to grant together with the text of the associated formulae. If the edict contained no action for the dispute in question, the praetor could either reject the parties’ request or make a new action based on the facts. If the edict did contain an action which was already known from the ius civile or had been created by a previous praetor or by himself, he gave permission to initiate the proceedings. The praetor and the opposing parties chose a judge and prepared a formula. In the second part of the formulary procedure, the judge heard the evidence and arguments of the advocates, and gave a decision.

This description, variations of which can be found in the modern books on Roman law, suggests that the parties themselves had to do all the work in the part of the trial that took place before the praetor. They could ask a jurist for advice, but they had to do the talking with the praetor themselves: they had to convince him that their case deserved to be put before a judge, and to determine which elements were to be included into the formula. If they were successful, then the advocates would come in and do the talking, and they would present the case before the judge.

On closer examination, however, this description leaves the reader with the following question. If the advocate had to present the case before the judge, would he not want to be involved in the drawing up of the formula because that was to be the basis of his case? It is difficult to find information on this matter in Justinian’s Digest since that work contains texts about the results of trials, not about the preliminaries. However, I did find some passages in the works of Cicero that deal with the preparation of trials: one in the letters to his friends, and two in his De oratore. I will discuss them in that order.

Letters to Friends

First-century AD bust of Cicero in the Capitoline Museums, Rome / Photo by Liam Clarkson-Holborn, Wikimedia Commons

In the summer of the year 44 BC, Cicero wrote a letter to his friend, the jurist Trebatius, asking him to help another friend of his, a certain P. Silius, who was about to be sued for an inheritance he had recently acquired. The letter, Ad familiäres 7.21, runs as follows:

Cicero S. D. Trebatio.
Sili causam te docui. Is pos tea fuit apud me. Cum ei dicerem, tibi videri sponsi- onem Ulam nos sine periculo facere posse, ‘Si bonorum Turpiliae possessionem Q. Caepio praetor ex edicto suo mihi dederit’, negare aiebat Servium, tabulas testamenti esse eas, quas instituisset is, qui factionem testamenti non habuerit; hoc idem Ofilium dicere; tecum se locutum negabat, meque rogavit, ut se et causam suam tibi commendarum. Nee vir melior, mi Testa, nee mihi amicior P. Silio quisquam est, te tarnen excepto. Gratissimum igitur mihi feceris, si ad eum ultro veneris eique pollicitus erit; sed, si me amas, quam primum. Hoc te vehementer etiam atque etiam rogo.

Cicero greets his friend Trebatius.
I have explained Silius’s case to you. Well, since then he has called upon me. When I told him that in your view we might safely make that stipulation ‘If the praetor, Q. Caepio, in accordance with his edict, has granted me the possession of Turpilia’s estate, etc.’, he said that Servius declared that this was not a will, that had been drawn up by someone who was not able to draw up a will; and that Ofilius had said the same; Silius said he had not talked it over with you, and asked me to commend him and his cause to you. There is no better man, my dear Testa, and I have no better friend than P. Silius, with the exception indeed of yourself. You will therefore greatly oblige me by going to him without an invitation and promising to do what he wants. But, as you love me, do so as soon as possible. This I earnestly beg of you to do, and I reiterate my request(12).

n this letter, Cicero makes it clear that his friend P. Silius is about to be drawn into a lawsuit. From the stipulation mentioned by Cicero, it can be inferred that Silius had, as heir upon intestacy, acquired the estate of a certain Turpilia, probably a relative, but that someone else had claimed the estate on the basis of a will that he says the deceased had made. Apparently, Silius had refused to hand over the inheritance because, in his view, this will was invalid. Turpilia had not had testamenti factio, i.e., as a woman, she could not make a valid will without the assistance of a guardian. It seems that Silius himself, as her nearest male relative, had been her guardian and that he had not assisted Turpilia in making a will. Therefore, he called the will in question invalid.

So far, the praetor had helped the testamentary heir by granting him the interdictum quorum bonorum, i.e., an order that Silius should hand over the estate to him. When Silius still refused to hand over the estate, the heir wanted him to cooperate in a procedure involving two sponsiones or stipulations. This stipulation consisted of a question put to Silius, i.e., whether he would pay a certain amount of money if it turned out that the praetor had rightfully given the interdictum quorum bonorum to the testamentary heir. Silius had to answer that he would. The claimant would have to promise the reverse. Then a judge would be appointed who was to decide whether the praetor had rightfully given the interdictum quorum bonorum to the claimant, and, in a more general sense, whether a woman could make a will without the assistance of a guardian.The answer to that question would decide who was entitled to the inheritance, Silius or the testamentary heir.

What is important in this connection is, first, that Silius had turned to Cicero for advice before the lawsuit about the inheritance had actually begun. He wanted to be sure that he could safely make this promise. Cicero had discussed the question of the stipulations with his friend, the legal expert Trebatius, and they had come to the conclusion that the will could be considered invalid and that Silius could make the stipulation. Since the lawsuit was now about to begin, Cicero asked Trebatius to hurry and help Silius. Apparently, Cicero did not want to act as an advocate himself for Silius – he was already in his sixties – and left it to his younger friend Trebatius to do so. In the late republic, it was normal practice for senators who were famous for their legal knowledge, to act as advocates for friends or clients.

A second conclusion to be drawn from this letter is that Silius or his opponent had also turned for advice to two jurists, namely Servius Sulpicius Rufus and Aulus Ofilius. Their sneaky answer with its double negation shows that these jurists were not impartial, as is often maintained in Romanist literature, but that they supported the claim of the testamentary heir by stating that the will could be considered as valid. It is not clear whether both jurists or one of them would act as advocate. However, it is a fact that Servius Sulpicius Rufus was famous for being an excellent orator as well as an excellent jurist(l3).

From the De oratore

First page of a miniature of Cicero’s De oratore, 15th century, Northern Italy, now at the British Museum / Wikimedia Commons

The second source which shows that the advocates were involved in the beginning of the lawsuit is Cicero’s De oratore, dating from approximately 50 BC. It contains a discussion a few leading Roman senators were said to have held some forty years earlier, in 91 BC, but which actually had been made up by Cicero. The discussion was about the question of what makes a good orator. In the first part of this book, Cicero introduces L. Licinius Crassus, M. Antonius and Q. Mucius Scaevola the Augur as the main participants in the discussion.

Crassus, who reflects Cicero’s own views, begins. He states that a good orator knows not only everything about oratory, but also about law, history, and philosophy. Antonius does not agree with him. He thinks it is impossible and unnecessary for an orator to know everything; basic knowledge is enough. When specialised knowledge is needed, for instance, on a legal matter, information should be obtained from an expert like Scaevola. Then Crassus is asked to expound his views on the art of rhetoric more fully, and he reluctantly consents to do so. He starts by dealing with what makes a good orator. In section 166, he mentions the importance of legal knowledge for an orator and he supports his argument by telling two stories. The first story, De oratore, I 166, runs as follows.

Potes igitur, inquit Crassus – ut alia omittam innumerabilia, et immensa, et ad ipsum tuum ius civile veniam -, oratores putare eos, quos ,multas horas exspectavit, cum in campum properaret, et ridens et stomachans scaevola, cum Hypsaeus maxima voce, plurimis verbis, a M. Crasso praetore contenderet, ut ei, quern defendebat, causa cadere liceret, Cn. autem Octavius, homo con- sularis, non minus longa oratione recusaret, ne adversarius causa caderet, ac ne is, pro quo ipse diceret, turpi tutelae iudicio, atque omni molestia, stultitia adversarii, liberatur?

‘Can you then,’ said Crassus, ‘(to pass over other matters innumerable and of vast importance and coming to your favourite ius civile itself), can you consider those men to be orators, for whom Scaevola, half laughing and half enraged, waited many hours, though in a hurry to start for the Playing Field, while Hypsaeus, at the top of his voice and with most exuberant verbosity, was struggling to procure from Marcus Crassus, the praetor, the non-suiting of the party for whom he himself was appearing, and Gnaeus Octavius, though a man of consular rank, was objecting, in a speech every bit as long, to having his opponent lose his suit, and to having his own client released, by the folly of his opponent, from a degrading verdict of dishonest guardianship and from all trouble wha- tever'<14)?

Crassus’ story is about two advocates. One is trying to obtain a formula for bad guardianship against the defendant, but he is arguing in such a clumsy way that he seems to try and get his request for a formula rejected. The other advocate, however, resents the idea that the plaintiff will succeed in his efforts and he is doing his best to make the praetor grant the formula. In this way, he is really hurting his client’s interests because if his client is found guilty of bad guardianship, he will be declared infamus. Crassus then asks his audience: ‘Can you consider such advocates who ruin their clients’ cases through lack of legal knowledge, to be orators?’ It is obvious that he thinks they cannot. One of the partners in the discussion, Q. Mucius Scaevola Augur, agrees with him and adds that he has heard the story from Mucius himself (cf. section 167). This Mucius probably was his cousin Q. Mucius Scaevola who, at the time of this story, was only 14 years old and who later was to become a famous jurist himself, known as Q. Mucius Scaevola Pontifex(15).

From the way Crassus tells this story, it is clear that, in his day, it was perfectly normal for advocates to speak for their clients in the first phase of the lawsuit, before the praetor. It also appears to be normal that the praetor did not stop the advocates making fools of themselves. The only unusual aspect of their behaviour was their apparent lack of legal knowledge, which made them act contrary to their clients’ interests.

Roman bust of Crassus in the Ny Carlsberg Glyptotek, Copenhagen / Photo by Diagram Lajard, Wikimedia Commons

The second story Crassus tells his audience to illustrate the importance of legal knowledge for an orator is to be found in De oratore, I 168:

Quid? His paucis diebus, nonne, nobis in tribunali Q. Pompeii, praetoris ur- bani, familiaris nostri, sedentibus, homo ex numero disertorum postulabat, ut Uli, unde peteretur, vetus atque usitata exceptio daretur, CUIUS PECUNIAE DIES FUISSET? Quod petitoris causa comparatum essem non intellegebat: ut, si Me infitiator probasset iudici ante petitam esse pecuniam quam esset coepta deberi, petitor, rursus cum peteret, ne exceptione excluderetur QUOD EA RES IN IUDICIUM ANTE A VENISSET.

Again, a few days ago, when I was sitting on the bench of our friend Quin- tus Pompeius, the city praetor, did not one of our accomplished advocates apply, on behalf of the defendant in an action of debt, for the insertion of the ancient and familiar defence, ‘AS REGARDS SUCH MONEYS AS HAVE ALREADY BECOME DUE’? He did not understand that this clause had been introduced for the benefit of the plaintiff, to the end that, if that defaulter should satisfy the judge that money had been claimed before it had become payable, the plaintiff could not be barred, on bringing a fresh action, by the special plea ‘THAT THIS MATTER HAS ALREADY BEEN LITIGATED’.

In section 169, Crassus qualifies the mistake made by the advocate as a blundering ‘in the most trifling and insignificant technicalities’. Therefore the mistake must have been easy to recognise. Nowadays, however, it is no longer easy to understand what exactly went wrong. It may help if I first give a short survey of the content of the formula in a case like this.

The case was about a debt. The formula for an action of debt was traditionally a very simple one. It generally consisted of three parts: the appointment of the judge, a description of the conflict formulated by the plaintiff (intentio) and the authorising of the judge to pass judgment (condemnatio). The intentio could be simple, for instance: ‘If it appears that the defendant ought to pay 1000 sesterces to the plaintiff, . . . ‘. In this type of cases, the judge was allowed to investigate only those aspects of the case that had been put forward in the formula. Therefore, the formula was all right if the defendant simply wanted to deny the claim. However, it would not do if the defendant wanted to put up another defence, for instance, that the money was owing but that it was not due yet, or that it had been agreed by the plaintiff and the defendant that the money should not be sued for. Then the defendant had to insert a defence in the formula to this effect. Of course, the plaintiff could answer this defence with a reply (replicatio) and the defendant could reply to this replicatio with another reply (duplicatio). It will be clear that it was essential for the advocate to include in the formula those elements which were favourable for his client, so that he would be able to make a proper defence before the judge(16).

In this case, the advocate for the defendant had insisted on inserting a wrong defence in the formula, because, as Crassus explains, this defence was used for the benefit of the plaintiff. He thereby made it difficult for himself to win his case. For what was the function of this particular defence? In Roman law, there was a rule that the same thing could not be claimed twice from the same defendant. When the parties had agreed on a particular formula and the praetor had authorised a judge to give a binding verdict, the plaintiff could not sue the defendant again for the same debt. If the plaintiff sued again for the same debt, the praetor could refuse the action or he could allow the defendant to insert in the formula the defence that ‘this matter has already been litigated’. Of course the plaintiff could then try and convince the judge that this was a different claim, but he would have a hard time of it. However, if the defendant succeeded in convincing the judge that it was the same claim, the plaintiff would lose the case. Consequently, he would want to protect himself against this risk, and include the replicatio ‘as regards such money as have already become due’ in the formula. The fact that, in this case, the advocate for the defendant wanted to insert these words in the formula, proved that he did not understand the function of this defence. It is hard to believe that, on the basis of this formula, he could properly defend his client and convince the judge to absolve him(17). He would certainly lose his case and thereby harm his own client.

Three conclusions can be drawn from this story. The first is that, as in the first story, it was perfectly normal for an advocate to speak for his client before the praetor. The second is that it was also perfectly normal for the advocate to be involved in drawing up the formula. In modern literature, it is always considered the task of the jurists to advise on the wording of the formula since that is considered a typically juridical activity. Of course, legal experts will occasionally have given advice to litigants, their advocates, or both, on how to formulate a particular defence or reply, particularly when these elements were new. However, as soon as such a formulation had turned out to be useful, the praetors included them in their edict so that it would become common knowledge. Then the advocates themselves could consult the edict and decide which elements they wanted to use. However, as Crassus/Cicero pointed out, they needed to have a basic knowledge of Roman law in order to properly use the devices promised by the praetor. The third conclusion is that it was not the praetor who ‘drew up a precisely worded formula’, as Kennedy et al. maintain, but the advocates. The only thing that the praetor had to do was to check whether he could give his approval to the wording of the formula and to the choice of the judge(18).

The Advocates, the Judge, and the Development of Roman Law

A judge makes a ruling / Wikimedia Commons

When the proceedings before the praetor were concluded and a citizen had been authorised to act as a judge and pass judgment on the basis of the formula in question, the second part of the trial or rather, the real trial could begin. According to Kennedy c.s., ‘the judge had to decide whether or not the conditions of the formula were fulfilled. He heard the case, sometimes with some advisers, and rendered his decision, from which there was no appeal.’ It was the advocate’s task to convince the judge of the righteousness of his cause. According to Kennedy, the procedure before the judge did not encourage elaborate speech-making and there was probably much interruption and questioning. It was only by Cicero’s time, when set speeches were conventional in other kinds of courts, that they had come to be used in private cases also.

I have two objections to this description, a major one and a minor one. I will begin with the minor one. It concerns the suggestion that it was only in Cicero’s day that set speeches came to be used in the private courts. I think we have to distinguish between a set speech, in the sense of a speech prepared by the advocate before the trial, and a published speech in the sense of a speech that was published by the advocate after the trial. Of course, the earliest speeches held in private courts that have come down to us are those held by Cicero, e.g., his Pro Quinctio, Pro Caecina and Pro Tullio. In his Brutus, Cicero relates that Marcus Porcius Cato (consul in 194 Β C) was the first orator to write down his speeches; from then on, many orators followed his example(19). From another source, we know that this Cato was an expert in civil law,(20) so some of Cato’s 150 speeches which Cicero says he has read must have dealt with civil law cases. Furthermore, there is no reason to assume that an advocate only prepared a speech in writing when he knew that he was going to publish it. On the contrary, in his De oratore, Cicero makes Lucius Crassus say that the best way to prepare a speech is to write it down(21). The creative period of Roman law began already in the late 3rd century BC, when, for instance, the notions of bona fides and aequitas were introduced in the law of contract. When, in the early 2nd century BC, rhetoric began to provide the advocates with the means of preparing their speeches in a much better way than they had been able to do before, they will certainly have used those means. Therefore, set speeches in private law courts must have been common practice long before the time of Cicero.

My major objection concerns Kennedy’s remark that the judge had to decide whether the conditions of the formula were fulfilled. He thereby suggests that the judge had to decide about facts only. Consequently, in their pleas, the advocates also had to deal with facts only. The same point of view can be found in Romanist literature. There, generally, a contrast is made between the praetor who – with the help of the jurists – had to decide on the law, and the judge who only decided on facts, so that it is the praetor who is considered as having been responsible for the development of Roman law.(22) Consequently, the relevance of rhetoric for the development of Roman law is denied.

I think this view is incorrect. In the Digest, that anthology of the writings of the classical jurists that was compiled in the 6th century at the order of Emperor Justinian, there are many texts to prove my point. However, there is one text which is particularly appropriate. It contains the judgment of someone who was a famous jurist in his time, the late 2nd century BC, and it shows that several other jurists recorded this judgment because they considered it important for the development of Roman law. Moreover – and that makes the text extra interesting – the legal problem in that case can be historically identified. The text in question, D.24.3.66, runs as follows:

Iavolenus libro sexto ex posterioribus Labeonis. In his rebus, quas praeter nu- meratam pecuniam doti vir habet, dolum malum et culpam eum praestare opor- tere Servius ait. Ea sententia Publii Mucii est: nam is in Licinnia Gracchi uxore statuit, quod res dotales in ea seditione, qua Gracchus occisus erat, périssent, ait, quia Gracchi culpa ea seditio facta esset, Licinniae praestari oportere.

Iavolenus in the sixth book of his commentary on the posthumous works of Labeo. Servius says that a husband is responsible for fraud and negligence with regard to all the property belonging to the dowry except for ready money. So ran the judgment of Publius Mucius. For he decided in the case of Licinnia, the wife of Gracchus, that matters belonging to the dowry that were lost in the revolt in which Gracchus was killed, should be restored to Licinnia because, he says, the revolt had broken out through the fault of Gracchus.

Publius Mucius Scaevola / Wikimedia Commons

The compilers of the Digest took this text from a collection of cases by Iavolenus Priscus, a famous jurist who lived in the first century AD and who had a career in the army (he was commanding officer in various legions), in politics (he was consul in 86), in administration (he was governor of three provinces), and in law. This text is drawn from a commentary Iavolenus wrote on the works of Antistius Labeo, a jurist from the time of Augustus. In the text, Iavolenus quotes an even earlier jurist, Cicero’s friend Servius Sulpicius Rufus who, in his turn, refers to Publius Mucius Scaevola, who is known as a consul, pontifex maximus, and jurist from the second century BC.

The text is about one Licinia (the spelling of Licinnia being otherwise unknown, I will abandon it). She was the wife of a certain Gracchus. On the basis of other sources, it is certain that this must have been Gaius Sempronius Gracchus. It is a well-known fact that he was one of the tribunes of the people who, in 121 BC, following his brother Tiberius Gracchus, proposed a number of reforms that were not well received by the conservative section of the Senate, the Optimates. Tiberius had made the same proposals ten years before, and this, partly due to his self-willed behaviour, had led to his murder. The same fate was to befall Gaius Gracchus. In the very same year, 121 BC, he met his end together with a large number – some say 3000 people – of political allies. His widow Licinia now had to fend for herself, so it was very important for her to get her dowry back.

In Roman law, a dowry was basically a contribution of the wife’s family to the expenses of the husband’s household. The dowry, which could consist of money and/or tangibles like land, silver tableware, and slaves, became the full legal property of the husband. In the late republic, a second purpose came to the fore, i.e., the idea that the dowry should also serve the maintenance of the wife after the marriage had been dissolved. Therefore, the husband then had to take care that the value of the dowry remained intact. He could sell dotal property, but at the end of the marriage, he or his heirs could be obliged to make repayment.

From Iavolenus’ text, it can be deduced that Licinia ‘s dowry consisted partly of money and partly of a number of specific objects. She may initially have asked her husband’s heirs to return the dowry to her. The problem was, however, that the part of her dowry that consisted of individual objects had been wholly or partly lost. This had probably happened when, after his death, the home of Gaius Gracchus was demolished(23). Apparently, the heirs had agreed to pay her the money but they had refused to pay her the value of the lost objects. The question then was whether the heirs of Gaius Gracchus could be held liable for the damage caused by the loss of these objects. Licinia took the matter to court.

The trial about Licinia’s dowry took place in a formulary procedure. This can be deduced from the title of Book 24.3 in the Digest: ‘How after the marriage has been dissolved the dowry can be claimed'(24). The texts in D.24.3 all deal with problems concerning the actio rei uxoriae, i.e., the action which the praetor could grant to a woman who – after a divorce or after the death of her husband – wanted to get her dowry (the res uxoria) back. Consequently, this must also have been the action granted by the praetor to Licinia against the heirs of her husband. Cicero, in his rhetorical works, refers twice to the formula of this action as traditional and well-known(25). However, we do not know whether, around 120 BC, it had already become standard. Therefore, it is not clear how much room there was for the advocates to negotiate about the wording of the formula. About one thing negotiation was possible, i.e., about the name of the judge that had to be stated in the formula. In this case, the praetor and the parties agreed to have Publius Mucius Scaevola for a judge(26). This is an interesting choice because, on the one hand, Scaevola was Licinia’s uncle but, on the other hand, he had been a supporter of the Gracchan faction(27). The formula of the action went more or less as follows(28):

Publius Mucius Scaevola shall be judge. If it appears that the heirs of Gaius Gracchus should return to Licinia the dowry or a part of it, then you, judge, must condemn the heirs to whatever will be better of more fair towards her, and if it does not appear, you must absolve them.

The crucial words in this formula are ‘whatever will be better or more fair towards her’, or, in Latin, ‘quod eius melius aequius erif‘ . They indicate that he judge did not have to base his decision on a law or a contract, but only on bona fides and aequitas, equity. There was no need for the litigating parties to insert defences. The words ‘quod eius melius aequius erif are so general that both parties could interpret them to suit their own interest.

Denarius of Publius Licinius Crassus / Wikimedia Commons

Unfortunately, we do not know who the advocates were, but we can try to reconstruct their pleas. From Iavolenus’ text, it becomes clear that the major point decided in this trial was that a husband was not only liable for damage to dotal property that he had inflicted with fraudulent intent but also for damage caused by his negligence. It seems that, until 121 BC, the husband was only liable for fraud. Therefore, it seems that the advocate of Gracchus’ heirs will have argued on the basis of the formula that the items in question had not been lost as a consequence of bad faith on the part of Gaius Gracchus but as a consequence of circumstances beyond his control, viz., the riots at the time of his death. Therefore, the heirs could not be compelled to pay the value of these objects to Licinia.

Licinia’s advocate evidently replied that this damage was most certainly to be compensated because Gaius Gracchus himself had given cause for the occurrence of the riots, as a consequence of which the goods had been lost. Gaius could have anticipated that his course of action would have far-reaching consequences and so, by persisting in this course of action, had acted imprudently with respect to the dowry. It would be unfair to let her suffer the consequences of his imprudent conduct.

How can these pleas be interpreted according to rhetorical theory? Since no laws were involved, it seems that the question has a status rationalis. The heirs recognised that they owed the dowry to Licinia, but the question was how this obligation was to be interpreted. Since their plea will have turned on the interpretation of the words quod eius melius aequius erit in the formula, I tend to identify it as a question of definitio. The advocate of the heirs could choose the topic of aequitas: it would be fair if the heirs were treated in the same way as earlier husbands and their heirs. The dotal objects had belonged to Gaius Gracchus’ property. He had treated them the same way as he had treated his other property. Now that his house with everything in it had been destroyed, he could not be held responsible for the loss of the objects belonging to the dowry. Therefore, his heirs should not have to pay their value to his widow either. Another topic the heirs’ advocate could choose was that of distinction. He could argue that the rules for a dowry consisting of money are different from those for specific objects. Of course, a husband was always liable for money and other fungibles, but his liability for specific objects was limited to fraud.

Licinia’s advocate could also refer to equity. The advocate may have pointed out that the purpose of the dowry was not only to contribute to the maintenance of the wife during marriage, but also to provide her with means to support herself after the marriage had been dissolved. Gaius Gracchus had been the owner of her dowry but, during the marriage, he was obliged to take care of the dotal property in such a way that he or his heirs would be able to return it to her at the end of the marriage. This obligation implied not only liability for fraud but also for negligence. Therefore, it was fair that the heirs would pay her the value of the objects that had been lost through Gracchus’ negligence. Finally, Licinia’s advocate could also deny that the rules for dotal money differed from the rules for other dotal property, and argue that a dowry is a dowry.

Publius Mucius Scaevola gave judgment in favour of Licinia. As usual, the text does not mention the reason behind this decision. I think Scaevola has been swayed by the relatively new idea that a husband, although being owner of the dowry, was in fact no more than a steward who had to look after the dotal property as if it belonged to his wife. With this judgment, he changed the legal position of husbands with regard to their wives’ dowries. From then on, their ownership was gradually reduced to a right of usufruct, and their freedom to dispose turned into a duty of stewardship.

Scaevola gave his judgment orally in the presence of both parties. He himself will have recorded it in writing. Since the administration of justice was not organised by the state, there were no official collections of judgments. Therefore, legal experts like Scaevola collected their own judgments or those of other judges, particularly if they themselves had given advice in the case in question.

In his turn, Servius Sulpicius Rufus took over Scaevola”s judgment as well, and so did, probably, Labeo and then again Iavolenus Priscus. It follows that all those jurists considered the judgment as an important contribution to the development of Roman law. Since it was based on the pleas of the advocates, trained in rhetoric, it is clear that rhetoric has been important for the development of Roman law in the formulary procedure.


In Roman rhetoric, judicial oratory definitely occupied a leading position. However, this was not only used in the criminal courts and, for private law cases, in the courts of the centumviri, but also in that of the unus index. The introduction of the formulary procedure removed the traditional limitations and created a need for interpreting existing formulae and making new ones. The art of oratory offered the means of doing so effectively.

Cicero, in his De oratore, makes Crassus say that, in Rome, knowledge of civil law was considered as much more important than in Greece, because there men of the lowest social station acted as assistants to orators in trials, whereas in Rome the most eminent and distinguished appeared in this capacity(29). This difference may also explain why it was Roman law and not Greek law that has conquered the Western world. Still, it was Greek rhetoric that made Roman law strong enough to do so.


  1. Cf. G. Kennedy, The Art of Rhetoric in the Roman World, Princeton, 1972, p. 7.
  2. G. Kennedy, 1972, p. 9.
  3. G.Kennedy, 1972, p. 14.
  4. In the same vein, for instance, P. Gerbrandy in his recent Dutch translation of Quin- tilian’s Institutio oratoria: De opleiding tot redenaar, Groningen, 2000, p. 17. In his A New History of Classical Rhetoric, Princeton 1994, p. 115, Kennedy is more positive: ‘Roman law was heavily dependent on precedent and what an orator successfully argued in a particular case could influence the development of the law.’ However, he gives no further explanation.
  5. Of the books mentioned by G. Kennedy, 1972, p. 9, note 9, the most influential still is the one by F. Schulz, History of Roman Legal Science, Oxford, 1946. See, in this work, particularly, p. 50-55.
  6. Already in Antiquity, Plato’s belief in absolute truth was controversial, but in the Middle Ages, it became popular again. Rhetoric has been taken seriously again only recently. Cf. Β. VicKERS, In Defence of Rhetoric, 2nd edition, Oxford, 1990.
  7. In recent years, there has been a tendency to modify this traditional view. See J.A. Crook, Legal Advocacy in the Roman World, London, 1995. In the same vein, J.W. Tellegen and O.E. Tellegen-Couperus, Law and Rhetoric in the causa Curiana, in Orbis Iuris Romani, Journal of Ancient Law Studies, 6, 2000, p. 171-203.
  8. For the civil procedures, see M. Kaser-K. Hackl, Das römische Zivilprozessrecht, 2nd edition, München, 1996.
  9. Th. Mommsen, Römisches Staatsrecht, 3rd edition, Leipzig, 1887 (rpt. Graz 1952), p. 163-164, and Id., Römisches Strafrecht, Leipzig, 1899 (rpt. Graz 1955 and Darmstadt 1961), p. 56-57. A.H.M Jones, The Criminal Courts of the Roman Republic and Principate, Oxford, 1972, p. 1-39.
  10. B. Santalucia, Diritto e processo penale nell’antica Roma, Milano, 1989, p. 31-89.
  11. E.P. Parks, The Roman Rhetorical Schools as a Preparation for the Courts under the Early Empire (John Hopkins University Studies in Historical and Political Science Ser. LXIII, no. 2), Baltimore, 1945, Ch. II.
  12. 1 use the text and translation by W. Glynn Williams for Loeb Classical Library, Cambridge (Mass.)-London, 1929 (rpt. 1983), II, p. 62-65; I only give a different translation of the words ‘negare aiebat Servium, tabulas testamenti esse eas, quas instituisset is, qui factionem testamenti non habuerit; hoc idem Ofilium dicere’. Glynn Williams has here: ‘he said that Servius maintained that a will, made by one who had not the legal right to make a will, was no will at all, and that Ofilius concurred’. Servius and Ofilius were two legal experts who had been consulted by the testamentary heir. For a proper understanding of the case, it is essential to translate Servius’ answer literally. He did not say that the will in question was invalid, but he said that it was not a will made by someone who could not make a will, in other words, that the will in question could be considered as valid. It is likely that the woman had made the will without the authority of a guardian. Traditionally, women could only make a will with the assistance of their guardian. However, in the 1st century BC, this tradition had fallen into disuse and women, perhaps with the exception of those belonging to the upper class, freely made wills themselves. The question of whether such wills were legally valid had not been answered yet. On this interpretation of Silius’s case, see J.W. Trllegen and O.E. Tellegen-Couperus, Vrouwenvoogdij: Nihil hoc ad ius, ad Ciceronem in B.C.M. Jacobs and E.C. Coppens (eds), Een Rijk Gerecht. Opstellen aangeboden aan prof. mr. RL. Nève, Nijmegen, 1998, p. 43 1-456, particularly, p. 450-454.
  13. Cf. Cicero, Brutus, 150-157. See also Pomponius in his well-known monograph, part of which has been included in D. 1.2, and particularly D.
  14. For both texts from De oratore, I use the edition and (slightly modified) translation by E.W. Sutton and H. Rackham, in Loeb Classical Library, Cambridge (Mass.)-London 1942 (rpt 1979), p. 114-115.
  15. For a more elaborate discussion of this text, see Olga E. Tellegen-Couperus, The So-Called Consilium of the Praetor and the Development of Roman Law in Tijdschrift voor rechtsgeschiedenis 69, 2001, p. 14-16.
  16. An important innovation of the formulary system was the introduction of the actions with the phrase ‘ex fide bona’ in the intentio. The judge was then definitely instructed to take ‘good faith’ into account, and condemn for an amount representing what the defendant ought to do in accordance with good faith. In such a formula, defences and replies were no longer necessary. However, this new formula was at first applicable only in a limited number of contracts (sale, hire, partnership, and mandate) and so the old-fashioned formula for claiming debts continued to be applied all through the late republic and early empire.
  17. According to K. Hackl, Praeiudicium im klassischen römischen Recht, Salzburg, 1976, p. 41, the advocate may have confused a praescriptio pro actore with a praescriptio pro reo but he does not give a further explanation. I agree with Hackl that the advocate may have requested a praescriptio pro actore, but I do not think he confused this device with a praescriptio pro reo because they are too different to be confused.
  18. A conclusion which cannot be drawn from this text is that the praetor used to sit with a consilium, as is maintained in Romanist literature. See my article mentioned in note 15.
  19. On Cato, see Cicero, Brutus, 61-76. Quintilian, Institutio oratoria, X.3, spends a whole chapter on writing as an important means of improving one’s oratorical qualities.
  20. Pomp.D.
  21. Cicero, De oratore, 1.150-153.
  22. See F. Wieacker, Römische Rechtsgeschichte, I, München, 1988, p. 447ff., and M. Kaser-K. Hackl, Das römische Zivilprozessrecht, 2nd edition, München, 1996. Earlier, in the same vein, F. Schulz, History of Roman Legal Science, Oxford, 1946, rpt. 1967, p. 49ff., followed by, for instance, A. D’Ors, Derecho privado romano, 4th edition, Pamplona, 1981, p. 57 nt. 1, p. 68-69.
  23. ccording to Appianus, Historia romana I 3,26, the houses of Gaius and his ally Fulvius Flaccus were demolished. This observation is partly confirmed by Cicero, who, in his speech De domo sua, (38. 102), mentions that Flaccus’ home was demolished because he had been an ally of Gaius Gracchus.
  24. D. 24.3 Soluto matrimonio dos quemadmodum petatur. Also O. Lenel, Palingenesia iuris civilis, I, Leipzig, 1889, rpt. Graz 1960, p. 313 includes this text in the chapter de iure dotium in Iavolenus’ Book 6 on Labeo.
  25. Topica, 66 and De officiis, III 61 .
  26. I conclude from the word ‘sententia ‘that P. Mucius acted as judge. In Romanist literature, however, the nature of the problem and the role of Mucius are interpreted differently: it is generally assumed that Licinia’s dowry had been confiscated and that P. Mucius, her uncle, had pleaded in the Senate for the entire dowry to be returned to Licinia. The assumed confiscation is based on a text by Plutarch, i.e., on the conclusion of his biography of Gaius Gracchus, and the word sententa is taken to mean Opinion’. For my arguments against this view, see The Role of the Judge in the Formulary Procedure, in The Journal of Legal History 22, 2001, p. 1-13.
  27. Her father was Q. Mucius Scaevola who, after his adoption, became known as P. Li- cinius Crassus Dives Mucianus. Cf. R.A. Baumann, Lawyers in Roman Republican Politics, München, 1983, p. 286-289.
  28. . Lenel, Das Edictum Perpetuum, 3rd edition, Leipzig, 1927, rpt. Aalen, 1956, p. 305, reconstructs the ado rei uxoriae as follows: Si paret Numerium Negidium Aulae ageri- ae dotem partemve eius reddere oportere, quod eius melius aequius erit, eius iudex Numerium Negidium Aulae Ageriae condemnato, si non paret absolvito. Of course, this reconstruction officially reflects the praetorian edict of 130 AD, when Emperor Hadrian ordered Salvius lulianus, the most famous jurist of his time, to draw up a definitive version of the edict. In fact, however, the text of the edict had not really been changed after 50 BC, when the actio de dolo was inserted.
  29. Cicero, De oratore, I 198.

Originally published by Revue belge de Philologie et d’Histoire 84:1 (January 2006, 59-75) under the terms of a Creative Commons license.