Ernest Metzger

Roman Judges, Case Law, and Principles of Procedure

Most who study Roman law today do so as historians, not lawyers. History includes doctrine, but Roman legal doctrine is rarely used to solve modern problems. There are exceptions: Roman law helps to solve modern problems in certain jurisdictions¹ and academic writing sometimes gives a Roman solution to a modern problem.² But the time is past when Roman sources were routinely put to work in the world of affairs,³ and most would say codification is the main reason.⁴	1
It is not a cause for regret. During virtually all of its "second life" Roman law has avoided becoming a specialist subject for antiquaries. Roman lawyers of the past wanted the world's attention and they got it: they spread Roman learning widely and profoundly. Also, codification made permanent some very good Roman ideas, even if the effect on legal science was not entirely good.⁵ The problem now is that Roman law has not yet recovered from its success. In reading the work of their predecessors, historians often come across ideas that are ostensibly Roman but look very modern. These ideas have perhaps crept in when someone has used the sources for a worthy purpose, but has not read them in a properly historical spirit. This is not to say that historians of today are pitted against lawyers of yesterday. Anyone with an interest in the ideas expressed in Roman legal sources⁶ might commit this kind of error. If the blame lies anywhere, it lies with the quality of the ideas and the admiration they inspire. We therefore forgive the Commentators their generalizations, the natural lawyers their logic and geometry, and the pandectists their reliance on the very words of the Corpus Iuris Civilis.⁷ But we do feel a special kinship with those of our predecessors who took a critical attitude to the texts, even when we disagree with them.⁸	2
When we speak about modern ideas influencing the study of Roman law, we usually think of private law.⁹ Yet the study of Roman civil procedure, my subject here, has suffered the same influences. Roman civil procedure virtually invites the use of modern ideas, because our knowledge of it is based on the smallest evidence. We have a single example of a treatise on procedure in book four of Gaius's Institutes (though Gaius writes more on actions than on procedure proper), together with a body of expurgated classical texts, fragments of statutes, and some documents prepared for litigation. The scarcity of the evidence, and the gap-filling required to make it make sense, have made textbook discussions of procedure vulnerable to fashion. The problem is made worse by the fact that the evidence has tended to trickle in,¹⁰ so that an idea, though widely accepted, may be based on outdated evidence. Anyone who studies procedure should be wary of ideas that have passed unchanged from textbook to textbook and be alert to the possibility that these ideas are modern.	3
Where do these ideas come from? The easiest explanation is that when writers of the past found a gap in the evidence on procedure, they simply supplied an idea from the modern law. This would explain in particular why some Roman procedure seems to reflect the ideas of nineteenth-century German writers, who were seeking to reform their own laws and institutions.¹¹ The problem with this explanation is not that it is wrong, but that "anachronism" as such is not really the enemy. Ideas attributed to antiquity are not wrong because they happen to be modern, and in any event anachronism is extremely useful in the study of procedure. Without resorting to anachronism, for example, Maine could never have made his famous (wholly anachronistic) statement that, in early legal systems, the substantive law appears to be "gradually secreted in the interstices of procedure."¹² Hoetink, in his influential¹³ 1955 essay on legal historiography,¹⁴ draws a distinction between good and bad anachronism. In some respects, he says, we are able to know antiquity better than the ancients; we know how the story turned out, and our intellectual arsenal is bigger. To try to stand where the ancients stood and observe what the ancients observed means turning our back on these considerable advantages.¹⁵ On the other hand, if we set out to explain events by assigning modern motives or states of mind to the actors, then we have strayed into bad anachronism, because we are making spurious connections between human actions and the events being explained.¹⁶ Thus it is perfectly acceptable for Maine to say that early law did not clearly observe the modern distinction between substantive and procedural law, but it would not have been acceptable for him to assume that the ancients were capable of making the modern distinction but, e.g., lacked the desire or opportunity to do so.	4
My argument is that the Roman formulary procedure has, in the past, attracted the same admiration as Roman private law and that in particular its admirers have used the kind of anachronism Hoetink discouraged: they have assigned modern motives or states of mind to the actors. My chief illustration is the use of "principles of procedure," developed by nineteenth-century German writers and applied by some to Roman procedure. These principles provided a platform for observers to admire Roman procedure, but invited them also to admire what could not be seen very clearly. I discuss these principles together with the subject of precedent: certain principles of procedure, I argue, contributed to the common view that in Roman law a judgment could not be a precedent for a future judgment. But first a short example will illustrate the point.	5

Summons
In the classical law a summons (in ius vocatio) for a lawsuit between persons was performed privately: a person would find his opponent or his opponent's representative, and bring him before the magistrate.¹⁷ This was the law from the time of the Twelve Tables.¹⁸ But there is a large gap in the evidence; Cicero, an important source of procedural law for the late republic, makes only a single reference to in ius vocatio.¹⁹ At one time, the gap was easily explained by the rubric to Digest 2.6, which ostensibly preserves the praetor's edict on the matter: In ius vocati ut eant aut satis vel cautum dent That those who are called before the magistrate should go, or provide a surety or an undertaking If the edict gave the defendant the choice of following immediately or giving some sort of promise to appear later, then perhaps (some believed) a plaintiff-to-be would avoid using the summons, preferring to obtain a promise to appear from his opponent.²⁰ This would explain the almost total absence of in ius vocatio in Cicero.²¹ The discovery of Gaius's Institutes, however, forced writers to find a new explanation for the absence. Institutes 4.46²² revealed something of the original, uninterpolated condition of the Digest rubric: the edict gave an action against a person who, once summoned, neither followed nor put forward a representative. A consensual promise to appear was not an option offered in the edict. Therefore if, in practice, consensual promises did indeed push aside summonses, some other explanation would have to be found for that practice. It is here that certain writers brought forward a host of "psychological" assumptions to show that Roman litigants preferred to begin their lawsuits consensually, not by a summons:²³ the defendant would prefer to agree on his appearance date, rather than allow himself to be seized immediately in the street; the plaintiff would prefer to know when his opponent was available, rather than risk missing him; both parties would prefer to arrive before the magistrate in a good state of preparation. Some believed also that the educated classes would not have submitted to such an archaic procedure as in ius vocatio.²⁴	6
These motives or states of mind, though not modern per se, do reflect a very modern balance of interests on the litigants' part, a balance they achieve by ignoring certain daunting features of Roman procedure. Roman procedure was uniquely ruthless about (1) forcing the appearance of a civil defendant before the magistrate, and (2) the legal effect of "joinder of issue" (litis contestatio). A defendant who appeared once before the magistrate could be forced to appear again and again²⁵ on penalty of, perhaps, a summary trial by a bank of judges or the sequestration and sale of his property.²⁶ Moreover, his only likely exit from this cycle of appearances was joinder of issue, which subsumed his prior rights into a single action, limited his defenses, and precluded him from raising certain matters in any subsequent suit.²⁷ In short, the power lay so much on the plaintiff's side at this stage of the proceedings that one wonders why a defendant would necessarily put all worries aside and submit to these risks purely for the sake of orderliness. And yet the "consent theory" assumed that so many defendants felt this way that summons by in ius vocatio fell out of use.	7
The view that in ius vocatio fell out of use, however, is no longer part of the common opinion. The common opinion is that consensual first appearances, to the extent they were used, would have required the use of the summons as an accompaniment,²⁸ at least in some cases.²⁹ Therefore Cicero's virtual silence on in ius vocatio cannot be evidence that consensual first appearances replaced summonses. No doubt some defendants very willingly gave promises instead of facing immediate arrest,³⁰ but the idea that Roman defendants as a class checked their fears and "took the long view" has no support in this argument from silence.	8
The discarded view that in ius vocatio fell out of use was not prompted by carelessness or blindness to the evidence. Certain writers simply admired the system of Roman litigation and concluded that in ius vocatio was crude and archaic and did not really belong. They then tried to mend the incongruity by assigning to the actors motives and states of mind that in the long run could not be supported by the evidence. The result was a picture of Roman procedure that was better than it actually was. This admiration-and-mending I illustrate more fully below.	9

Case Law?
Whether the Romans had a system of case law is the kind of anachronistic but nonetheless worthwhile question mentioned above.³¹ The answer, however, is probably no. The formulary procedure itself, by anyone's account, would have made any system of case law extremely difficult.³² The formulary procedure lasted roughly from the end of the second century BC to the third century AC. Forms of action were described precisely, and this gave the law continuity, but in an individual case the forms could be altered and assembled in different ways to create a specific statement of issues. This altering and assembling was performed, however, not by a judge, but by a judicial magistrate. The magistrate drew up the statement of issues and passed them on, fully formed, to the judge for resolution. This division of labor alone must have put many opportunities for innovation beyond the judge's reach. A system of case law would be all the more difficult to realize because a judgment was oral³³ and gave no reasons, and therefore any innovations in the judgment would have to be inferred from the issues, the resolution, the legal advice taken by the judge, and the like. Accordingly the common opinion says that the decision of a Roman judge did not make law for future decisions³⁴ and the common opinion is probably right.	10
Kaser has reviewed the evidence thoroughly and given the strongest defense of the common opinion.³⁵ It is clear, he says, that the Romans had a different notion of law than our own: they accepted the jurists' law as "law" even though it was not a binding pronouncement of the state and was frequently ambiguous. They might have extended the same treatment to judges' decisions, but the evidence suggests they did not.³⁶ Various authorities describe the sources of law: the juristic writers³⁷ do not mention the decisions of judges at all, and the rhetorical writers, where they include the decisions of judges, mean only that a decision might be a source of a right (like a contract),³⁸ that it might "make law" for a later decision in the same dispute,³⁹ or that it might serve as evidence of the law for an unrelated but legally similar case.⁴⁰	11
These are the authorities that bear directly on the question, but a review of the history of Roman lawmaking, according to Kaser, gives the same answer: that the decisions of judges had no greater force than as evidence of the law.⁴¹ For a time the college of pontifices had a monopoly on declaring what the law was,⁴² a monopoly that ensured that the law was consistent and that the judge could not depart from the advice he was given.⁴³ When the competence to declare the law passed to the jurists, it was possible for differences of opinion about the law to arise. We might think that judges would then have had the freedom to contribute to the law through their decisions, but Kaser says they were no more capable of making law than before. Knowledge of the law nevertheless remained the preserve of the now freed community of jurists, because the judicial magistrates and judges alike lacked, as before, the specialist juristic training.⁴⁴ Juristic training, according to Kaser, elevated the value of expert advice. It meant that a jurist could be interested in a judgment only insofar as it favored or contradicted a professional legal opinion. The opinion of a lay judge might interest others, but to a jurist it meant nothing.⁴⁵ How the particular case turned out ... leaves him cold, because he knows that judgments are given by laymen whose sometimes irrelevant opinion he is indifferent to. The jurist writes for those who are like himself, and for his professional successors. In this respect Roman jurisprudence is an esoteric science, which decides for itself whose contributions to value, and which passes over lay opinion—indeed, even the opinion of an elementary teacher of law like Gaius—in silence.⁴⁶	12
Kaser's argument is a careful and balanced one. He does not ask whether a judge's decision was binding in the way a modern judge's decision is, which would hold decisions to a higher standard than jurists' law.⁴⁷ He also acknowledges the popular force of a judge's decision—a very public pronouncement of a state-appointed officer⁴⁸—and defends the case-based nature of jurists' law against the claim that the jurists were engaged in system-building.⁴⁹ But his argument is nevertheless a very particular one: he asks and answers the question of whether a Roman judge participated in Rechtsfindung,⁵⁰ that is, in the kind of intellectual activity remitted to the jurists from the pontifices. We might be satisfied with Kaser's argument that they did not, but still wonder whether, as a matter of empirical fact, judges ever relied on past decisions that innovated in some way and that were therefore more than "evidence of the law." We could call this "precedent," even if such decisions were never binding, were ignored by the experts, and left nothing obvious behind in the Digest.⁵¹	13
This is more or less the kind of precedent that Jolowicz argues did exist for a time in Rome.⁵² What he has in mind is not a body of case law with the same intellectual pretensions as jurists' law, nor of course binding law, but simply a practice among judges of following and citing prior decisions. This practice, he suggests, existed until the late republic, when jurists began truly to command the scene and could give judges better support than prior decisions could.⁵³ Before that time, "precedent ... undoubtedly played a part in the development of the law."⁵⁴	14
Jolowicz's argument is based almost wholly on inference. He relies on reports of judicial proceedings from Roman Egypt. The reports are extracts from official diaries, sometimes perhaps made as evidence of rights, but sometimes also as authorities for legal propositions.⁵⁵ The latter Jolowicz deduces from the fact that collections of reports on particular points of law have survived⁵⁶ and that certain petitions and judgments cited and relied on these reports as authority.⁵⁷ He sums up this part of his argument: One may say then that there existed [in Roman Egypt] a definite practice of citing decided cases as authority in courts of law, that judgments were sometimes expressly based on such authority, and that the practice was facilitated by the use of official diaries and collections made from the reports that they contained. But it is unlikely that there was much in the way of a theory of case law, though it was no doubt clear that judgments had to give way to "laws," including especially imperial enactments of any sort, and quite possibly some idea that a judge need not consider too carefully the decision of one whose rank was inferior to his own. In particular there is no reason to suppose that case law was justified as a species of customary law.⁵⁸ The nub of Jolowicz's argument, however, is the inference he asks us to draw from the dates.⁵⁹ The collections of decisions, and petitions and decisions that cite earlier decisions, date from the late first century to the middle of the third: "The period for which we have evidence," Jolowicz says, "thus corresponds not too badly to that of the classical Roman law,"⁶⁰ by which he means post-republican law. The common opinion regards the practice of citing and following cases to be provincial only,⁶¹ but Jolowicz does not believe the practice is Greek, nor is it attested in Ptolemaic Egypt, and this suggests to him that it originated in Rome.⁶² He finds support for this inference in a study that (he accepts) gives several examples in which judges from republican Rome established precedents.⁶³ He concludes: It is possible that the growth of [jurists'] responsa in the early Empire caused less attention to be paid to precedents at Rome, but the annexation of Egypt comes at the end of the Republic, and republican practice is thus sufficient to explain the development in that country of a system which appears to be thoroughly in keeping with the practical genius of Roman administration.⁶⁴ Accordingly what Jolowicz is suggesting is that to some unknown (and probably unknowable) extent, precedent existed in republican Rome. He does not suggest that anything like a system of case law existed, or that judges ever made law. The precedential force he attributes to the decisions of judges is similar to the precedential force Honoré attributes to jurists' law:⁶⁵ some judges found justification for their decisions in the examples which previous decisions furnished.	15
For present purposes I assume that Jolowicz's conclusion is at least plausible. What I wish to point out is that the common opinion, rejecting the existence of "case law," would not necessarily reject the existence of this kind of precedent. Kaser, defending the common opinion, denies the existence of case law because the sources deny its existence and because the judge's office was incapable of creating what the Roman jurists (and therefore we) would regard as law. Jolowicz's claim is too modest to worry the common opinion on these points.⁶⁶ He does not argue for the existence of "case law" and, to the contrary, makes no assumptions about the quality or content of any precedent. In fact nothing in his notion of precedent would bar a judge from following a prior decision for any foolish reason. He does broadly conclude that judges must have contributed to the development of the law more than is usually appreciated, but on his explanation these contributions are incidental (because there was no theory of case law) and limited in time (because the jurists eventually met the need). The hegemony of jurists' law is indeed something on which both Jolowicz and the common opinion agree, Jolowicz speculating only that there was a time when the hegemony did not exist and judges looked to one another for support.⁶⁷ This is the only concession his explanation requires of the common opinion.	16
Jolowicz's is among a handful of explanations that claim that judges contributed to the development of the law to some degree.⁶⁸ It is worth asking why these explanations have received very little attention.⁶⁹ This is the principal subject of a recent study by Tellegen-Couperus: she asks why the common opinion has neglected the role of judgments in the development of the law, and recites arguments (quite different from those of Jolowicz) in favor of such a role.⁷⁰ She suggests that the common opinion resisting such arguments is based on two assumptions that are mistaken.⁷¹	17
The first is the mistaken assumption that the judge decided on facts alone and that questions of law were left to magistrates and jurists.⁷² If this assumption had ever taken hold, it is easy to see how the common opinion would have passed over arguments like Jolowicz's: a decision on the facts leaves very little for a subsequent judge to rely on, even when the influence of jurists is absent. But did it take hold? There are echoes of it in Savigny's Historical School which, as Whitman points out, interpreted Roman sources in a way hostile to the idea of precedent.⁷³ In particular Puchta, Savigny's successor, played down the authority of precedent in Rome by emphasizing the different roles of judge and jurist.⁷⁴ Puchta made it clear, however, that the fact/law distinction was wrong: If one wanted to compare Roman judges with today's jury it would be a mistake to imagine that the former were concerned only with the investigation of facts, that the legal points had been determined for them during the proceedings in iure.... The magistrate indeed decided the general legal basis to the claim in the course of allowing the actions and exceptions and ordering the trial, but even in the most simple action, one for a specific sum of money owed, there could be occasion for legal questions to arise in the trial, and this was even more the case in actions in which the judge, instructed to enquire what one party ought to do for the other in good faith, is given wide scope for legal considerations, and this is the case even in real actions (e.g., rei vindicatio), where the judge is only generally instructed to enquire whether the plaintiff is the owner, and where therefore the entire body of law on individual modes of acquisition could come up for consideration.⁷⁵ Many others subsequently expressed the same view,⁷⁶ and in fact it is difficult to find writers who maintain the opposite.⁷⁷	18
The second mistaken assumption Tellegen-Couperus cites is that judges were lay persons without legal knowledge.⁷⁸ On this assumption, she argues, the common opinion has rejected judgments as irrelevant to the development of the law, even though there are reasons to believe some judges had good knowledge of the law.⁷⁹ This assumption is closer to the mark: we have seen it above. Kaser (representing the common opinion) does argue that judges lacked juristic training and that as a result jurists were not inclined to take their judgments seriously.⁸⁰ The problem here is less with the assumption than with the argument. By disproving the assumptions underlying the common opinion, Tellegen-Couperus hopes to disprove the common opinion. But disproving this assumption does not help the cause: individual judges might be well versed in the law, but as a rule jurists still cite other jurists, not judgments.⁸¹ More important for present purposes, this assumption would not explain why "precedent" in Jolowicz's very limited sense has gone mostly undiscussed: one might assume that all judges were laypersons, but still accept that a judge could rely on an earlier judgment, even if his understanding of it were superficial.	19
The common opinion, I have suggested, is essentially right, but the common opinion does not logically exclude a theory of precedent like Jolowicz's. That the common opinion does not acknowledge this kind of precedent can be attributed, not to the assumptions just discussed, but to certain ideas found in some of the older literature. There the judge is not limited to deciding facts or disparaged for his ignorance of law, but plays a genuinely important part in the formulary procedure. He does not make law, but his duties complement the law. His role is very special and admired, but it is a role in which paying any attention to other judgments would make no sense.	20

Principles of Procedure
In the passage quoted above Puchta discourages his readers from confusing the ancient judge with the modern jury. For his time, this was useful advice. The Roman formulary procedure had disappeared by the third century AC and had no obvious historical continuity with modern German procedure. But the formulary procedure was attractive to some who contributed to the literature on procedure in the nineteenth century,⁸² and discussions of procedure therefore sometimes invited comparisons between the modern and the ancient judge. This attraction to the formulary procedure began with the Historical School: In a much greater degree [than by the natural law school] was procedural science forwarded by the historical school. Unlike the natural law school, this started with the rules of positive law and traced them back to their original sources. It showed that the law of Germany was compounded of Roman, canonical, and Germanic elements and sought to reconstruct these elements in all their distinctness.... A notable impetus was given to this movement by the discovery of the Gaian manuscript in 1816. Now, for the first time, was made possible a thorough understanding of the Roman civil procedure and, especially, the formulary system. And, contemporaneously, under the influence of the re-awakened national consciousness, there came about an understanding and correct appreciation of the original German law.⁸³ The discovery of Gaius's Institutes was fortunate. Even though the formulary procedure was not in full flower at the time he wrote, Gaius had a law teacher's interest in setting out its details, and the Institutes was rich in new information.⁸⁴In the decades following its discovery, writers on procedure beginning with Bethmann-Hollweg, a disciple of Savigny, undertook to study the relationship between the German common-law procedure and all its antecedents, including the formulary procedure.⁸⁵ The works produced by Bethmann-Hollweg, Keller, Heffter,⁸⁶ and others combined history and dogma, relying to a high degree on Roman law.⁸⁷	21
An important vehicle for comparisons between ancient and modern procedure were the so-called "principles of procedure," which, generally speaking, were a set of specific, observable qualities inherent in a given system of procedure. German legal science has been the most energetic in the development of these principles. Millar, writing in 1923, says: Not the least of the contributions for which theoretical study is indebted to German procedural science are certain generalizations which it has made concerning procedural method. By such generalizations it has identified and delimited the fundamental conceptions which consciously or unconsciously determine the form and character of systems of procedure.⁸⁸ For example, a given system of procedure may observe the principle that the parties and not the judge are responsible for determining what materials the judge will rely on in settling the controversy, the so-called Verhandlungsmaxime ("the principle of party presentation").⁸⁹ The system might also insist that both sides be heard (Grundsatz des beiderseitigen Gehörs: "principle of bilateral hearing") or that proceedings be open even to those who are not directly interested in the outcome (Grundsatz der Öffentlichkeit: "principle of publicity"). These principles are less often discussed among Anglo-American lawyers than continental lawyers,⁹⁰ but no Anglo-American lawyer should doubt how useful they are whenever civil procedure needs reform.⁹¹ In particular, when German-speaking peoples were trying to codify a common civil procedure out of the different practices of the territorial and imperial courts, the act of identifying and refining principles of procedure was essential. The drafters of the Deutsche Zivilprozessordnung (1877) were able to identify and reject a principle that required the parties to present all possible issues at the outset (Eventualmaxime), a principle that tended to inflate the number of issues in dispute. But they were also able to identify and embrace the principle of orality (Grundsatz der Mündlichkeit), which lessened the complexity and secrecy of a procedure dominated by writing.⁹²	22
Are the principles useful to historians of Roman law? There is of course nothing wrong in claiming that the Romans observed certain principles in their procedure, regardless of whether they actually discussed them. For example Kaser, in his textbook (the leading one) on Roman procedure, puts the principles together at the opening as a kind of summary.⁹³ Even if the principles are a purely modern creation, they provide a means of describing Roman procedure as a whole; Kaser of course does not claim that they are followed consistently. Wenger, in his own (earlier) leading textbook, also uses the principles as a means of description, though he has a different approach from Kaser, omitting a general summary of principles and instead citing the principles as the need arises, as in this example: The judge must hear both sides, if they or their representatives are ready to engage. Each side must therefore be given the opportunity to speak. That alone is signified by what today we call the principle of bilateral hearing, although this principle was never so distinct that the judge could give judgment only when both sides had actually spoken.⁹⁴ Wenger is speaking to readers already familiar with the principle and telling them that they may observe it (or some of it) in this aspect of Roman procedure.⁹⁵ Both Kaser and Wenger illustrate what Hoetink, discussed above,⁹⁶ approved as a good use of anachronism. That the Romans themselves did not speak of an idea does not mean we cannot profitably observe the idea in Roman procedure and give our own name to it.⁹⁷	23
What makes a historian hesitate of course is the overtly historicist claim that the study of Roman procedure is to some degree the study of the original German law,⁹⁸ coupled with the claim that certain principles "unconsciously" determined Roman procedure.⁹⁹ These claims threaten to introduce what Hoetink warned was the bad use of anachronism, at least so far as they purport to explain causally the existence of rules and institutions by modern motives or states of mind. An example arises under the principle of orality. This principle, as just mentioned, regards it as a virtue for certain aspects of a system of procedure—particularly the parties' presentations to a tribunal—to be oral rather than written.¹⁰⁰ Kaser says very generally that Roman procedure was characterized by the principle of orality during all its historical phases.¹⁰¹ But Seidl, in making the same point, goes on to describe the various advantages that modern doctrine attributes to the principle of orality and that the Romans supposedly gained by observing it.¹⁰² The reader is left thinking that Roman trials were oral because the Romans were mindful of this principle and its advantages, rather than, e.g., some combination of illiteracy, time limitations, available methods of writing and duplication, etc.¹⁰³ A second example arises under the "principle of publicity." Bethmann-Hollweg says that the principle governed both during the formulary procedure and the earlier legis actio procedure.¹⁰⁴ He then goes on to argue that civil trials before a single judge were always public.¹⁰⁵ The problem here is that a legal system which follows "the principle of publicity" is not simply holding trials in public, but observing a principle of due process. According to the principle, public participation acts as a restraint on abuses of procedure.¹⁰⁶ The converse of "publicity" is not "privacy" but "secrecy,"¹⁰⁷and in this respect Bethmman-Hollweg is assuming a great deal about the reasons for public trials in Rome. If he had said that Roman trials were characterized by "a desire not to be secret" his assumptions would be more obviously incorrect.¹⁰⁸	24

Immediacy and Orality
Two principles are especially relevant to trials before the judge. The principle of orality has been mentioned; closely related is the principle of immediacy, which tries to preserve the integrity of a judgment by ensuring that arguments and proof are put to the judge in the most direct manner possible. Under this principle it is a virtue for a judgment to be given by the very person who considers the proof and hears the parties' recitals. If either of these acts is performed by a deputy, then the procedure is "mediate."¹⁰⁹ Any delay between presentation and judgment is likewise contrary to the principle.¹¹⁰ The principle of orality is related to the principle of immediacy, in the sense that a system of procedure that encourages allegations, proof, or arguments to be put orally to the judge furthers the cause of immediacy.	25
Some works say that private lawsuits in Rome were characterized by orality and immediacy.¹¹¹ The evidence usually cited is a supposed rule governing the conduct of trials from the time of the Twelve Tables to at least the time of Cicero. The rule states that a civil trial¹¹² must not last more than one day: pleading, proof, argument, and judgment must take place on the same day, the judgment being given before sunset (the "one-day rule").¹¹³ If judgment cannot be given before sunset, then the case must begin anew on another day, with at least some of the events of the previous session being repeated.¹¹⁴	26
The virtues of immediacy and orality are perhaps more obvious here than in any other aspect of the Roman trial, and in discussing the rule the older procedural writers make their clearest statements about the judge and his office. The core of their view is that a judge who must make a decision within the limits of this rule will have a vivid picture of the case in mind and thereby be less liable to make a mistake. Bethmann-Hollweg, writing in 1864, says: The aim of this extremely narrow time limit lay ... mainly in the fact that out of a coherent and vivid development of the case, passing into the judge's understanding as, in one sitting, the hurried litigant furnishes it, the truest conviction on the part of the judge will develop. Indeed it is for this reason that our own jury courts follow nearly the same principle.¹¹⁵ The main virtue of the rule and the attendant principles is the accuracy of the judge's decision as Baron, writing in 1884, confirms: The entire proceeding must also be carried out to completion in one day so that, there being no written record, the judge can give judgment with a fresh impression of what he hears. If this is not possible, the judge announces an adjournment and the proceeding begins anew on a new date; factual material in particular is presented again.¹¹⁶ The repetition of events in the subsequent hearing is an important qualification to the rule; it helps to ensure the accuracy of the decision, if in fact the trial cannot finish in one day, as Wenger explained in 1925: [With the one-day rule] the judge could give his decision under the freshest impression of the proceedings taking place orally before him. Proceedings are often both oral and immediate.... [In bigger cases] the advocates' recitals, recapitulating all of the substance of the arguments and proof, easily preserved the immediacy of the proceedings.¹¹⁷	27
There are good reasons to believe, however, that the one-day rule did not exist at all.¹¹⁸ Its existence depends on a very narrow and unnatural reading of the sources. The main source of the rule is the following text from the Twelve Tables:	28
XII Tab. I, 9: si ambo praesentes, sol occasus suprema tempestas esto.¹¹⁹ If both [parties] are present, sunset shall be the latest time. The words of the Twelve Tables have been assembled piecemeal from sources much later than the law itself, and context is often difficult to determine. The modern editors of the Twelve Tables have suggested that this text refers to the hearing before the magistrate rather than trial before the judge, ¹²⁰ though conceivably the sunset rule could apply to both. Assuming it does apply to trials, the meaning of suprema tempestas is the main point of dispute. Proponents of the one-day rule assume that it means "the latest instant," so that the text as a whole announces "sunset shall be the latest instant [for trial]." The alternative is that suprema tempestas means "the latest time of day," which would suggest a different rule, that trials (or proceedings before the magistrate) could not take place at night. The modern editors would support the second alternative: "Note that a ban on proceedings at night need not preclude resumption the following day."¹²¹ In their favor is the fact that the sources for this text expressly treat suprema tempestas (and the shortened term suprema) as a time of day,¹²² and the lex Plaetoria (post 241 BC) is known to have clarified the time of day deemed to be suprema.¹²³ None of the sources for XII Tab. I, 9 suggests "instant," "point of demarcation," or "end point" for tempestas.¹²⁴ To find support for "latest instant" one must go outside the Twelve Tables and its sources: in Cicero's pro Tullio and Tacitus's Dialogus there are references to advocates "using up the day" (diem eximere) by speaking too long, and in the pro Quinctio Cicero charges his opponent with speaking so long he leaves no time for judgment.¹²⁵ In the first two passages, the proponents of the one-day rule must presumably take diem as something like a technical term for "the one allotted day" and in the last passage would infer that judgment must be given on the day Cicero is speaking. These interpretations are possible, but it is equally possible the three passages are reciting the ordinary remarks of an impatient advocate.¹²⁶ In any event, the modern editors of the Twelve Tables are unconvinced: "Cicero, Tull. 6, and Quinct. 34, manifestly do not prove that the hearing of cases was limited to a day...."¹²⁷	29
Here it is enough to say that the older procedural writers have read the sources in a very selective way. The excerpts above¹²⁸ suggest they have done so because they have in mind a particular kind of judge: one whose impressions of the case must be cultivated with the help of orality and immediacy. Persons familiar with German procedure will know that this kind of judge is a very modern one. A commentary to the German Code of Civil Procedure describes hearings before the judge in this way: The merits of an oral hearing are obvious. The parties' recitals give a vivid picture of the actual circumstances of the case and the matters in dispute. Gaps and uncertainties in their presentations can be easily taken care of, misunderstandings corrected.¹²⁹ Immediacy serves many of the same purposes as orality. A textbook on German law says, typically: "[T]he court which has to deliver the ruling must—due to [the principle of immediacy]—obtain the most direct impression of the case."¹³⁰ Similarly, a German textbook on procedure, speaking of a rule limiting the re-presentation of factual material before a second judge, says "Such a rule however withholds from the reviewing court the very essential impressions which immediate contact with the parties and evidence provides."¹³¹ Whether accurate impressions were similarly demanded in Roman trials is the question. That this view of the judge is modern does not mean it cannot be Roman. Yet this view first arose, so far as I can determine, only in the middle of the nineteenth century,¹³² and modern textbooks of Roman law tend not to describe the Roman judge in this way, even though the one-day rule still occasionally appears.¹³³ I suggest below that this view of the judge had its origin in the program of the Historical School.	30

Decisions as Data
It was the general aim of the Historical School, first, to uncover by historical research the actual, subsisting law, and second, to find, realize, and refine the system that lies within that law.¹³⁴ Because the raw "data" for the law was properly found only by historical research, Savigny saw no reason to confuse the historical method by distinguishing theory from practice: The convergence of theory and practice is the starting point for genuine improvement in the administration of justice, and something which we must learn from the Romans in particular: our theory must become more practical, and our practice more scientific than it has been up to now.¹³⁵ For present purposes, the convergence of theory and practice had two important consequences, somewhat opposed to each other. On the one hand, making practice more scientific limits the significance of a single case. A person can learn from a single case, Savigny says, only if he also keeps before him a picture of the whole of the law.¹³⁶ On the other hand, the products of practice—including the decisions of courts—can have scientific worth as historical data, and therefore can be the object of investigation and research.¹³⁷ (Mohnhaupt charts the development of Savigny's thinking on this very point, showing that by the 1840s Savigny's work was relying to a much greater degree on the citation of court decisions, though of course Savigny continued to reject the idea that a decision could have any precedential force.¹³⁸) To fulfill this goal of combining theory and practice, it was important that the judge's office not be reduced to the mechanical application of texts; the bench should be staffed by "intellectually charged persons" (geistreiche Menschen) with exposure to developing theory.¹³⁹ The reform of procedure would help to bring about this desired state of affairs.¹⁴⁰	31
The procedural writers who followed Savigny, I suggest, accepted the scientific worth of a judge's decision: it is for that reason that the Roman judge must be furnished with procedures that will give his decisions value as historical data. The Romans, on this view, have consciously or unconsciously introduced the principles of immediacy and orality in order to allow the judge to gain the clearest impression of the case. It is essential that the judge gain the clearest impression, because only then will his decision be the "true" one,¹⁴¹ expressing the sense of right felt by his fellow Romans.¹⁴² The judge, one should note, is neither mechanically applying rules, nor expressing his own opinion (as Ihering says, the judge's decision might go against his own convictions¹⁴³). The value of his decision as historical data arises from his authenticity as a voice for this shared sense of right, as this passage from Baron makes clear: The essence of the institution of the [Roman] judge lies in the fact that judgments in legal disputes take place, not through government officials, but through men: men who spring directly from The People. This arrangement has two advantages: (1) certain citizens ... are summoned to take part in public affairs, and not just for voting, but for a profound intellectual activity; (2) the law assumes a quality of nationhood; this is because the judge, in giving judgment, automatically uses the notions of Right that live in The People.... Here it should be emphasized that the institution of the judge in Rome knew nothing of the modern distinction between "factual" and "legal" questions; the Roman judge ruled on both alike; ...¹⁴⁴ An authentic decision, of course, is not law, but only the data for law,¹⁴⁵ and even as data it is not necessarily valuable to a jurist. A jurist, like any scientist, is not interested in every event that corroborates what he already knows. But any data that deserves the name must be produced under the proper conditions: procedural principles are an essential part of those conditions, because they help to ensure the accuracy of the data.	32
In short, the procedural writers admired Roman procedure as a model for the modern law. But in their enthusiasm to complete the model they assigned to the Romans their own contemporary desire for authentic decisions and then equipped the judge with the procedural principles needed to achieve them.	33

Precedent
From this view of the Roman judge, certain ideas naturally followed. Case law of course is impossible; it mistakes the data for the scientific finding. Another idea, particularly important here, is that no judge could take any interest in what another judge had done. Each judge takes an impression of the rights and wrongs in the specific case before him and from that impression gives a yes-or-no decision.¹⁴⁶ That decision, on this view, is useless to a subsequent judge. He has no reason to import a distillation of the impressions of another judge, taken from other facts, into his own case. To use an example from the physical sciences: a falling body does not accelerate at a certain rate because another falling body accelerates at that rate, but because it obeys a law on the acceleration of falling bodies. In the same way a judge obeys the law, not single manifestations of the law.	34
This view therefore leaves no room for a theory of precedent like Jolowicz's. Judges would not rely on past court decisions, because there is nothing to learn from them. Those decisions remain discrete expressions of what is right in a specific case. This is true even if, as Jolowicz suggests, the judges of republican Rome did not always have jurists to advise them. Nothing in this view, however, belittles the judge or his office: his decisions lack the ability to instruct other judges, not because he is untrained or only permitted to decide questions of fact, but because he fulfills the necessary, admired, and almost oracular role of announcing what is true for the facts before him.	35
This view of the Roman judge is not a modern one, and the question is whether with its disappearance the common opinion could acknowledge a practice of judges following judges. I suggested above that it could. The common opinion rejects the idea that judges made law, but that is not inconsistent with the idea that some judges relied on previous decisions. This is an empirical question, different from the question of lawmaking. Jolowicz has suggested that ideas were indeed exchanged among judges in this way, before jurists dominated legal advice. Even if this exchange of ideas left nothing behind in the sources, the practice alone would be significant to historians. It would affect their understanding of the judge's office, the conduct of trials, and the development of professional advice. Most important, the silence of the common opinion on this notion of precedent should not be taken as a rejection. The silence, I have suggested, is due to a discarded, historical theory of lawmaking that borrows substantially from its own time.	36
The second part of Jolowicz's argument is on a different footing: he argues that through their decisions judges contributed to the development of the law. Dawson, Collinet, and Tellegen-Couperus have also suggested that judges contributed to the development of the law. The irony here is that the discarded, historical theory accommodates this view better than the common opinion. On the one hand, the procedural writers discussed above would presumably argue that a Roman judge's decision, though authentic, would rarely provide anything new for a jurist to ponder; a jurist, after all, would probably have advised the judge in the first place. On the other hand, they might admit that in the rare case a judge's decision could reveal a genuinely new piece of information. To borrow Dawson's example: a judge might make a decision on whether a certain act constituted good faith.¹⁴⁷ The procedural writers would regard this decision as new data, awaiting the examination of the jurist. Jolowicz would say that this decision contributes to the development of the law. These are not so very different.	37
The common opinion, however, does not easily accommodate the idea that judges contributed to the development of the law. This is because it regards judicial decisions as no more than evidence for the law.¹⁴⁸ But here the common opinion is at its weakest, at least so far as the late republic is concerned. Special juristic training may have created some manner of division between "law" and "evidence for the law," but Jolowicz is suggesting very plausibly that this division was not so clear until "the growth of responsa in the early Empire caused less attention to be paid to precedents at Rome."¹⁴⁹ This is exactly where Kaser's account of the ascendancy of jurists should give way, in particular his assumption that the authority to interpret the law passed seamlessly from the pontifices to a ready fraternity of jurists.¹⁵⁰ This account is impatient to see jurists off the marks and giving advice to judges, and passes over the irregular development of the jurists' profession.¹⁵¹ If precedents were observed in the late republic, then lawmaking followed a far less orderly division of labor than the common opinion acknowledges.	38
The more general point is that Roman law has been admired for a long time, but sometimes for qualities it does not have. Procedure, like private law, has been a victim of this. Principles of procedure have sometimes played an unhelpful role. Writers of the past have discovered ideas in the sources that were not really there. They did so, not because they were careless, but because Roman law is inherently rich. But a historian, like a husband, must beware of old admirers.	39
Ernest Metzger is a Senior Lecturer in the School of Law, University of Aberdeen, Scotland, and editor of IusCivile.com. This article is based on the author's Oldfather Lecture at the University of Illinois at Urbana-Champaign on November 9, 2001. He wishes to thank William M. Calder III and the Department of Classics at Illinois for the invitation. He also wishes to thank Bruce Frier of the University of Michigan Law School for his comments and suggestions. All translations are by the author. Notes ^1. There is a short discussion, with literature, of Roman law as a living source of authority in Andrew Lewis, "Roman Law in the Middle of Its Third Millennium," Current Legal Problems 50 (1997): 414–17. In the United Kingdom, Roman law has been used conspicuously in Indian Oil Corp. v. Greenstone Shipping SA (Panama), [1988] Q.B. 345; Shilliday v. Smith, 1998 Sess. Cas. 725; and most recently in Scotland: McDyer v. The Celtic Football and Athletic Co., 2000 Sess. Cas. 379, where a football spectator was injured by a falling object. On McDyer and the use of Roman law in Scotland, see Tammo Wallinga, "Effusa vel deiecta in Rome and Glasgow," Edinburgh Law Review 6 (2002): 117–23. ^2. The best examples of this are not blind applications of Roman rules, but demonstrations of how Roman law achieves something the modern law for some reason does not. See Alan Rodger, "Mrs. Donoghue and Alfenus Varus," Current Legal Problems (1988): 1–22; Peter Birks, "Harassment and Hubris. The Right to an Equality of Respect," The Irish Jurist (n.s.) 32 (1997): 1–45. ^3. The view that the study of Roman law today is largely historical is discussed by Lewis, "Roman Law," 414–19. But others wish the situation were different. Wieacker had hopes that legal history could be conceived in such a way that it remained part of legal science. He had in mind a mostly doctrinally oriented historical study. See Franz Wieacker, A History of Private Law in Europe, trans. T. Weir (Oxford: Clarendon Press, 1995), 336–40. Zimmermann's view is similar to Wieacker's. See, e.g., Reinhard Zimmermann, "Savigny's Legacy: Legal History, Comparative Law, and the Emergence of a European Legal Science," Law Quarterly Review 112 (1996): 576–605. Zimmermann regrets that "A neo-humanistic approach to legal history has superseded the historical approach to legal science" (ibid., 598) and suggests that legal history can help to find common doctrinal features underlying European legal systems (ibid., 600–601). Similarly: Reinhard Zimmermann, "Roman and Comparative Law: The European Perspective (Some Remarks apropos a Recent Controversy)," Journal of Legal History 16 (1995): 26. Jolowicz would be sympathetic to Zimmermann's view. H. F. Jolowicz, "Utility and Elegance in Civil Law Systems," Law Quarterly Review 65 (1949): 322–36. ^4. Zimmermann discusses the point at length, but does not attribute the return of historical Roman law solely or even predominantly to codification. Reinhard Zimmermann, Roman Law, Contemporary Law, European Law: The Civilian Tradition Today (Oxford: Oxford University Press, 2001), 46–52. ^5. Zimmermann, Roman Law, 1, notes that codification fragmented the European legal tradition, but it is very much part of his argument that the tradition was continuous notwithstanding codification. ^6. As opposed to, for example, the history of the ideas or the sources themselves. ^7. See Peter Stein, Roman Law in European History (Cambridge: Cambridge University Press, 1999), 73; M. H. Hoeflich, "Law and Geometry: Legal Science from Leibniz to Langdell," American Journal of Legal History 30 (1986): 96–109; Zimmermann, Roman Law, 18–19. A common illustration of this is the use of "maxims," which, in their original context, were narrow explanations of rulings, not generalized principles. See Peter Stein, "Civil Law Maxims in Moral Philosophy," Tulane Law Review 48 (1974): 1076. Buckland made pandectist scholarship a special target: "German writers often seem to attribute to Roman law rules and modes of thought which are the product of later ages." W. W. Buckland, "Wardour Street Roman Law," Law Quarterly Review 17 (1901): 179. ^8. E.g., the French humanists and the Dutch elegant school. See especially Lewis, "Roman Law," 403–8, who suggests that, instead of placing ourselves in the tradition of a revived Roman law beginning around AD 1000, we should acknowledge the historical nature of the present discipline and place ourselves 500 years into a historical tradition, beginning with humanist scholarship. ^9. Several examples are given in Buckland, "Wardour Street Roman Law," 179–92, and W. W. Buckland, "More Wardour Street Roman Law: The actio de in rem verso,"Law Quarterly Review 31 (1915): 193–216. See also Alan Watson, "Illogicality and Roman Law," Israel Law Review 7 (1972): 14 (= Alan Watson, Legal Origins and Legal Change [London: Hambledon Press, 1991], 251) ("[T]he usefulness of Roman law for later ages, coupled with its enforced isolation from other systems of antiquity, has often led to an exaggerated respect for it, and to its being regarded as well-nigh perfect, immutable, fit for all people"); Francis de Zulueta, The Roman Law of Sale (Oxford: Clarendon Press, 1945; reprinted 1957), 25 ("So long as the Corpus Iuris was in force as actual law, a harmonious doctrine had to be extracted from the texts, even at the cost of forced interpretations ..."); H. J. Wolff, Roman Law: An Historical Introduction (Norman: University of Oklahoma Press, 1951), 219–20 ("[That excerpts from the Digest] had to be taken for authoritative statements of valid law compelled the Pandectists to assume too conservative an attitude regarding the sources"). On the gradual break of legal history from legal science, see Mathias Reimann, "Nineteenth-Century German Legal Science," Boston College Law Review 31 (1990): 871–73; Wieacker, History, 330–34. ^10. The discovery of Gaius's Institutes in 1816 is the most obvious example, but there are many other examples of discoveries that have significantly added to our knowledge of Roman procedure: lex de Gallia Cisalpina (Roman Statutes, ed. Michael Crawford [London: Institute of Classical Studies, 1996], vol. 1, no. 28), discovered in 1760; the Fragmenta Vaticana (P. E. Huschke, E. Seckel, and B. Kuebler, Iurisprudentiae Anteiustinianae Reliquias [Leipzig: Teubner, 1927], 2/2:191–324), discovered in 1821; lex Coloniae Genetivae (Roman Statutes, ed. M. Crawford, vol. 1, no. 25), discovered in 1870; the Este Fragment (Roman Statutes, ed. M. Crawford, vol. 1, no. 16), discovered in 1880; collections of waxed tablets from Pompeii, Puteoli, and Herculaneum, discovered in the last one hundred years, for which see Peter Gröschler, Die tabellae-Urkunden aus den pompejanischen und herkulanensischen Urkundenfunden (Berlin: Duncker & Humblot, 1997); and the lex Irnitana (Julian González, "The Lex Irnitana: A New Copy of the Flavian Municipal Law," Journal of Roman Studies 76 [1986]: 147–243), discovered in 1981. ^11. The most well-known example of this, which considerably affects procedure, is the supposed "intuition" of the Roman jurists. The idea that a Roman jurist, as a member of a professional class, had special powers of discerning the law was put forward by Savigny and followed by some even in modern times. See Max Kaser, "Zur Methode der römischen Rechtsfindung," in Ausgewählte Schriften (Camerino: Jovene, 1976), 1:10–14; A. Arthur Schiller, "Jurists' Law," in An American Experience in Roman Law (Göttingen: Vandenhoeck & Ruprecht, 1971), 159, and the critical comments by Laurens Winkel, "The Role of General Principles in Roman Law," Fundamina 2 (1996): 108–9, and Wolfgang Waldstein, "Topik und Intuition in der römischen Rechtswissenschaft," Festgabe für Arnold Herdlitcza, ed. Franz Horak and Wolfgang Waldstein (Munich: Wilhelm Fink, 1972), 248–49. Another example, argued persuasively by William Turpin in a recent study, is the idea that the cognitio procedure was introduced as part of a coherent program of law reform undertaken in early imperial Rome, an idea that Turpin attributes in part to Savigny. William Turpin, "Formula, cognitio, and Proceedings extra ordinem,"Revue Internationale des Droits de l'Antiquité, 3rd ser., 46 (1999): 501–2. ^12. Henry Sumner Maine, Dissertations on Early Law and Custom (London: John Murray, 1883), 389. And without resorting to anachronism, Watson could not have argued that Maine was wrong. Alan Watson, "The Law of Actions and the Development of Substantive Law in the Early Roman Republic," Law Quarterly Review 89 (1973): 387–92. ^13. Winkel, "The Role of General Principles," 108: "Since Hoetink it has generally been accepted that the use of anachronistic concepts is inevitable, but they have to be realised and justified." ^14. H. R. Hoetink, "Les Notions Anachroniques dans l'Historiographie du Droit," Tijdschrift voor Rechtsgeschiedenis 23 (1955): 1–20. ^15. Ibid., 7–8. ^16. See ibid., 14, 15–16, and especially 10: "Je crois que pour poser les problèmes les notions soi-disant anachroniques sont absolument admissibles, tandis qu'elles ne sont certainement pas admissibles quand il s'agit d'expliquer de manière psychologique les actions et la conduite des hommes d'autrefois." ^17. See Max Kaser, Das römische Zivilprozessrecht, 2nd ed. rev. Karl Hackl (Munich: Beck, 1996), § 30 II. ^18.XII Tab. I, 1–3 (Roman Statutes, ed. M. Crawford, 2:584–88). ^19. Cicero, pro Quinctio 61. ^20. See the authorities cited in Ernest Metzger, "The Current View of the Extra-Judicial vadimonium," Zeitschrift der Savigny-Stiftung für Rechtsgeschichte (rom. Abt.) 117 (2000): 140–41 nn.23–25. ^21. M. A. von Bethmann-Hollweg, Handbuch des Civilprozesses (Bonn: A. Marcus, 1834), 1:247; idem, Der römische Civilprozeß (Bonn: A. Marcus, 1865), 2:199. Much of the subsequent literature followed Bethmann-Hollweg's lead and declared that lawsuits, at least up to Cicero's time, were begun consensually rather than by summons. The literature is cited in Metzger, "The Current View," 142–43 n.29, but Kelly can be quoted as an example: "Towards the end of the Republic actual in ius vocatio came to be generally replaced, as a means of initiating litigation, by vadimonium ... ; this is the procedure found, for example, in all the speeches of Cicero." J. M. Kelly, Roman Litigation (Oxford: Clarendon Press, 1966), 6–7. ^22. Gaius, Institutes 4.46: Ceterae quoque formulae quae sub titulo DE IN IUS VOCAN-DO propositae sunt, in factum conceptae sunt, velut adversus eum qui in ius vocatus neque venerit neque vindicem dederit. ^23. See Giovanni Pugliese, Il Processo Civile Romano (Milan: Giuffrè, 1963), 2:401; André Fliniaux, Le Vadimonium (Paris: Arthur Rousseau, 1908), 105; Bethmann-Hollweg, Der römische Civilprozeß, 2:199. ^24. Fliniaux, Le Vadimonium, 104–5 ("[L']usage se répandit d'abondonner le procédé brutal et archaïque de l'in jus vocatio pour assurer la première comparution du défendeur in jure à l'aide d'un vadimonium ..."), 105 ("Nous savons ... qu[e le vadimonium] devint le mode de citation usité entre gens de bonne société"); Bethmann-Hollweg, Der römische Civilprozeß, 2:199 ("[E]ntsprach [diese neue Einleitungsform des prozesses] mehr als jenes überraschende, unhöfliche Antreten auf offener Straße dem Anstandsgefühl der gebildeten Classen"). In the note immediately following, ibid., 199 n.15, Bethmann-Hollweg explains the basis of his opinion: "Darüber läßt sich freilich nicht streiten; daß aber die Römer so fühlten, beweißt m.E. die Ausschließung der in ius vocatio gegen Respectspersonen schon im älteren Recht." Bethmann-Hollweg has in mind the persons named in Digest 2.4.2, 4 (Ulpian 5 ed.) and Digest 2.4.3 (Callistratus 1 cog.), e.g., magistrates with imperium, priests performing sacred rites, and judges hearing cases. This is some support for his opinion, though it is a leap to infer from this brief list that the better classes of Romans rejected in ius vocatio. ^25. The best example is in Cicero, pro Quinctio 22, where the plaintiff Naevius repeatedly forces the defendant Quinctius to appear. ^26. That a magistrate might order the parties' reappearance if the matter were not ready for his decision is well attested: see Gaius, Institutes 4.184 and the authorities cited in Ernest Metzger, "The Case of Petronia Iusta," Revue Internationale des Droits de l'Antiquité 47 (2000): 159 n.30, 160 n.34. The penalty for refusing to make the promise, however, is uncertain. A magistrate in Cisalpine Gaul had the power to order a summary trial against a person who refused to promise to reappear in Rome, according to a statute from the first century BC (lex de Gallia Cisalpina, col. 2, ll. 21–24; see Roman Statutes, ed. M. Crawford, 1:466). Lenel believes it is possible that, in Rome, the penalty for refusing to promise to reappear was the sale of the refusing party's goods. Otto Lenel, Das Edictum Perpetuum, 3rd ed. (Leipzig: B. Tauchnitz, 1927; reprinted Aalen: Scientia, 1956), 80–81 n.11. The possible contents of the edict are discussed by David Johnston, "Vadimonium, the lex Irnitana, and the edictal commentaries," in Quaestiones Iuris, ed. Ulrich Manthe and Christoph Krampe (Berlin: Duncker & Humblot, 2001), 119–20. ^27. See Kaser/Hackl, Zivilprozessrecht, § 42 II. ^28. The change of view was prompted by the discovery of memoranda that record promises to appear in a place other than the magistrate's court. See J. G. Wolf, "Das sogenannte Ladungsvadimonium," in Satura Roberto Feenstra, ed. J. A. Ankum et al. (Fribourg: Éditions Universitaires Fribourg Suisse, 1985), 63–65. Wolf says that in ius vocatio probably continued to be used to bring the defendant from the meeting place to the magistrate. Wolf's suggestion is accepted in the newest edition of Kaser: Kaser/Hackl, Zivilprozessrecht, 231. Another possibility is that in ius vocatio was used as a threat to induce the making of the promise. Teresa Giménez-Candela, "Notas en torno al 'vadimonium,'" Studia et Documenta Historiae et Iuris 48 (1982): 135, 165. ^29. Duncan Cloud has argued recently that at least some promises were for appearance at the court; these would not require the use of in ius vocatio. Duncan Cloud, "Some Thoughts on vadimonium," Zeitschrift der Savigny-Stiftung für Rechtsgeschichte (rom. Abt.) 119 (2002): 159. ^30. See Metzger, "The Current View," 133–78, arguing that this practice is poorly supported in the evidence. ^31. When preparing this material the following work was not available to me: Umberto Vincenti, ed., Il Valore dei Precedenti Giudiziali nella Tradizione Europea (Padova: Cedam, 1998). However, based on the summary in Labeo 47 (2001): 451–67, my arguments would not change. ^32. There are useful, general discussions of the formulary procedure in David Johnston, Roman Law in Context (Cambridge: Cambridge University Press, 1999), 112–18; Ditlev Tamm, Roman Law and European Legal History (Copenhagen: DJØF, 1997), 53–64; Francis de Zulueta, The Institutes of Gaius (Oxford: Clarendon Press, 1953), 2:250–54. ^33. But see Bruce Frier, The Rise of the Roman Jurists (Princeton: Princeton University Press, 1985), 227 ("[Judges] probably also supplied written copies to the litigants"). Recorded judgments that survive are rare. The most outstanding example of such a record under the formulary procedure is probably the Tabula Contrebiensis (87 BC) from Botorrita in Spain, which recites two formulae and adds a judgment at the end. See J. S. Richardson, "The Tabula Contrebiensis: Roman Law in Spain in the Early First Century B.C.," Journal of Roman Studies 73 (1983): 34–41; Peter Birks, Alan Rodger, and J. S. Richardson, "Further Aspects of the Tabula Contrebiensis," Journal of Roman Studies 74 (1984): 45–73. A second outstanding example is a decision on the inheritance of a Roman soldier in Egypt, from perhaps A.D. 41 or 42, preserved on papyrus (P. Mich. III 159 = Fontes Iuris Romani Antejustiniani, ed. Vincenzo Arangio-Ruiz [Florence: S. A. G. Barbèra, 1969], vol. 3, no. 64). This decision does not actually preserve the aspect of a Roman formula (though the beginning mimics a demonstratio; see Gaius, Institutes 4.40), but it does use some of the technical language of the formulary procedure. See Paul M. Meyer and Ernst Levy, "Sententia des iudex datus in einem Erbrechtsprozeß unter Claudius," Zeitschrift der Savigny-Stiftung für Rechtsgeschichte (rom. Abt.) 46 (1926): 283–85, who also suggest the dates, ibid., 278. ^34. See, e.g., Mario Talamanca, Istituzioni di Diritto Romano (Milan: Giuffrè, 1990), 23; Frier, Roman Jurists, 229–31; Alan Watson, Law Making in the Later Roman Republic (Oxford: Clarendon Press, 1974), 171–72; John P. Dawson, The Oracles of the Law (Ann Arbor: University of Michigan Law School, 1968), 100–107; W. W. Buckland and Arnold D. McNair, Roman Law and Common Law: A Comparison in Outline, 2nd ed. rev. F. H. Law-son (Cambridge: Cambridge University Press, 1952), 6–10; Fritz Schulz, History of Roman Legal Science (Oxford: Clarendon Press, 1946), 92. ^35. Max Kaser, "Das Urteil als Rechtsquelle im römischen Recht," in Festschrift Fritz Schwind, ed. Rudolph Strasser et al. (Vienna: Manzsche Verlags- und Universitätsbuchhandlung, 1978), 115–30 (= Max Kaser, Römische Rechtsquellen und angewandte Juristenmethode [Vienna: Böhlau, 1986], 42–64). ^36. Ibid., 118–19. ^37. Gaius, Institutes 1.2: Constant autem iura populi Romani ex legibus, plebiscitis, senatusconsultis, constitutionibus principum, edictis eorum qui ius edicendi habent, responsis prudentium. A similar list is given in Digest 1.1.7 (Papinian 2 def.) and Justinian, Institutes 1.2.3. ^38. Cicero, de Inventione 2.22.68; Quintilian, Institutio Oratoria 7.4.6. ^39. Quintilian, Institutio Oratoria 5.2.1. ^40. Kaser, "Das Urteil als Rechtsquelle," 118–21. On the last item perhaps the clearest sources are Rhetorica ad Herennium 2.13.19 (which notes that decisions inevitably conflict and suggests how advocates can make the best of the decisions they have) and Quintilian, Institutio Oratoria 5.2.1 (who includes in his definition of praeiudicium judgments that are said to serve as exempla in other cases). Kaser says this use of judgments as "evidence for the state of the law" might be regarded as a "source of law," but notes that no jurist makes this characterization. Kaser, "Das Urteil als Rechtsquelle," 121. Other rhetorical sources that allude to judgments as evidence of the law are cited in Frier, Roman Jurists, 129 n.102. Advocates themselves may have regarded prior cases as something more than evidence of the law: see ibid., 229. On this point see also notes 148 to 150 below and accompanying text. ^41. Kaser, "Das Urteil als Rechtsquelle," 124–28. ^42. See Digest 1.2.2.6 (Pomponius ench.). ^43. Kaser, "Das Urteil als Rechtsquelle," 125–26. ^44. Ibid., 126: "Dennoch bleibt die Rechtskennerschaft das Reservat der nunmehr freien Juristenzunft, weil die Gerichtsmagistrate ebenso wie die Urteilsrichter nach wie vor der juristischen Fachbildung entbehren." ^45. Ibid., 126–27. Similarly, Schulz, Roman Legal Science, 24. ^46. Kaser, "Das Urteil als Rechtsquelle," 127: Wie der konkrete Prozeß ausgegangen ist ... läßt ihn kalt; denn er weiß, daß die Urteile von Laien gefällt werden, deren zuweilen unsachgemäße Meinung ihm gleichgultig ist. Der Jurist schreibt für seinesgleichen und für den Nachwuchs in seiner Zunft. Damit bleibt die römische Jurisprudenz eine esoterische Wissenschaft, die selbst entscheidet, wessen Leistung sie gelten läßt, und die die Laienmeinung, ja sogar die eines juristischen Elementarlehrers wie Gaius, schweigend übergeht. ^47. For a summary of views on the ius respondendi, which may have given special force to the opinions of some jurists, see Geoffrey MacCormack, "Sources," in A Companion to Justinian's Institutes, ed. Ernest Metzger (Ithaca: Cornell University Press, 1998), 11–14. ^48. Kaser, "Das Urteil als Rechtsquelle," 127. ^49. Ibid., 122–24, contra Okko Behrends, "Die Causae Coniectio der Zwölftafeln und die Tatbestandsdisposition der Gerichtsrhetorik," Zeitschrift der Savigny-Stiftung für Rechtsgeschichte (rom. Abt.) 92 (1975): 171. See also Schulz, Roman Legal Science, 24 ("[The jurists'] principle was to wait till the case occurred, and to feel their way from case to case"). Cf. O. E. Tellegen-Couperus, "The Role of the Judge in the Formulary Procedure," Journal of Legal History 22/2 (2001): 2 ("According to prevailing doctrine ... Justinian's Digest does not reflect the legal practice of the classical period but a scientific, theoretical kind of literature that was at most inspired in terms of style and content by the questions that were submitted to the jurists in practice"). ^50. Kaser, "Das Urteil als Rechtsquelle," 115: "[D]rängt sich ... dem angelsächsischen Romanisten die Frage auf, ob nicht auch die Römer, mindestens in gewissen Grenzen, das Urteil als Mittel der Rechtsfindung anerkannt haben, und gegebenfalls, weshalb sie dabei so zurückhaltend verfahren sind. Diesen bisher nur wenig untersuchten Fragen wollen wir im folgenden nachgehen." ^51. Honoré, writing on jurists' law: The doctrine that precedents are binding is not an essential feature of a system based on precedent.... What is necessary for a system of precedent is that arguments from example should be admissible in the sense that an appeal to a previous instance or example is an adequate justification for a decision, not necessarily that it compels decision. A. M. Honoré, "Legal Reasoning: Rome and Today," South African Law Journal 91 (1974): 89–90. ^52. H. F. Jolowicz, "Case Law in Roman Egypt," Journal of the Society of Public Teachers of Law 14 (1937): 1–15. Jolowicz reviews some of his arguments in idem, "Precedent in Greek and Roman Law," Bullettino dell'Istituto di Diritto Romano 46 (1939): 394–405; idem, Lectures on Jurisprudence, ed. J. A. Jolowicz (London: Athlone, 1963), 220–23; H. F. Jolowicz and Barry Nicholas, Historical Introduction to the Study of Roman Law, 3rd ed. (Cambridge: Cambridge University Press, 1972), 354 n.4. ^53. Jolowicz, "Case Law," 15. ^54. Jolowicz and Nicholas, Historical Introduction, 354 n.4. Jolowicz later used stronger language for this important conclusion. In the second edition of the Historical Introduction, he said somewhat equivocally that precedent played "some" part. H. F. Jolowicz, Historical Introduction to the Study of Roman Law, 2nd ed. (Cambridge: Cambridge University Press, 1954), 569 (appendix to 365 n.2). The edition that followed appeared after Jolowicz's death and was edited by Barry Nicholas. Nicholas possibly amended "some part" to "a part" in light of Jolowicz's very strong statement, published in 1963: "[I]t is clear that the influence of actual decisions in the development of the law was at all times considerable." Jolowicz, Lectures on Jurisprudence, 223. ^55. Jolowicz, "Case Law," 5. ^56. Ibid., 5–7. ^57. Ibid., 7–10. ^58. Ibid., 12. Jolowicz's interpretation of the papyrological evidence is bolder than that of Weiss, who had treated most of the same material twenty-five years earlier. Egon Weiss, "Recitatio und Responsum im römischen Provinzialprozeß, ein Beitrag zum Gerichtsgebrauch," Zeitschrift der Savigny-Stiftung für Rechtsgeschichte (rom. Abt.) 33 (1912): 212–39. Weiss saw the papyri as confirmation that, in provincial practice, a series of decisions might reveal local customary law, as suggested in Digest 1.3.34 (Ulpian 4 off. pro.) and Code 8.52.1 (AD 224). Ibid., 227–32. Jolowicz goes further than this, arguing that decisions were made on the authority of prior judgments—even single judgments—and not necessarily on the authority of any custom revealed in the judgments. Katzoff has treated the same material more recently and agrees with Jolowicz that decisions were sometimes based on prior judgments. Ranon Katzoff, "Precedents in the Courts of Roman Egypt," Zeitschrift der Savigny-Stiftung für Rechtsgeschichte (rom. Abt.) 89 (1972): 290 ("We can definitely assert that judges did occasionally base their decisions on the precedents cited to them"). On the other hand, Katzoff does not assert, as Jolowicz does, that decisions in Egypt were a source of law, but says only that they were "acceptable evidence of the law." Ibid., 291. Kaser cites Katzoff and Weiß with approval, but one wonders whether Jolowicz's article was available to him: "[Katzoff] nimmt mit Jolowicz an, daß die ägyptischen Precedents nicht so sehr Rechtsentstehungs- als vielmehr Rechtserkenntnisquellen waren, also Informationsmittel von der Art, wie sie von der römischen Rhetorik verstanden wurden." Kaser, "Das Urteil als Rechtsquelle," 128 n.63. This is not Jolowicz's view, as the quotation above makes clear. ^59. On what follows, see Jolowicz, "Case Law," 12–15. ^60. Ibid., 12. ^61. See note 58 above. ^62. Jolowicz, "Case Law," 12–15. Similarly, in his address to the Riccobono Seminar: "It is difficult to believe that a system of quoting precedents would have arisen in Egypt if it had been contrary to Roman ideas about the administration of justice. There was, as I have tried to show, nothing in classical Greece out of which such a system could have arisen. There is no reason to believe that the native Egyptians had anything of the sort." Jolowicz, "Precedent," 404. ^63. Jolowicz, "Case Law," 15. See Paul Collinet, "Le rôle des juges dans la formation du droit romain classique," in Recueil d'études sur les sources du droit: en l'honneur de François Gény (Paris: Librairie du Recueil Sirey, 1934), 23–31. ^64. Jolowicz, "Case Law," 15. ^65. Note 51 above. ^66. Kaser unfortunately does not give his opinion of Jolowicz's conclusion. See note 58 above. ^67. Jolowicz's suggestion is therefore significant primarily for what it suggests about the development of jurists' law and less significant for anything it might say about the character of that law. Cf. the conclusion of Tellegen-Couperus, "The Role of the Judge," 5, quoted below, note 70. ^68. I would include here Collinet, "Le rôle des juges"; Tellegen-Couperus, "The Role of the Judge" (whose argument is described below, note 70); Dawson, The Oracles of the Law, 104. ^69. Of course, the explanations may simply be wrong, whence the lack of attention. Yet Schiller called Jolowicz's study "brilliant" and was generally enthusiastic about its conclusions; he noted that it had not provoked any discussion. A. Arthur Schiller, Roman Law: Mechanisms of Development (The Hague: Mouton, 1978), 267–68. He also seems to cite part of Collinet's study with approval. Ibid., 267 n.11. Katzoff, writing after Schiller and with Schiller's encouragement, accepts Jolowicz's general conclusion, but so far as I am aware he is the only one to do so. Katzoff, "Precedents," 291–92. ^70. Tellegen-Couperus, "The Role of the Judge," 1–13. Her thesis is not that a judge's decision was ever a formal source of law, but only that judges contributed to the development of the law to a greater degree than is commonly appreciated. They did so, she says, through the jurists, who would use judicial decisions that they thought were important for the development of the law, and in which they themselves had played a role (either as advisers or indeed as judges), and incorporate them into their collections of responsa. Therefore, "the texts of the Roman jurists, which have been compiled in Justinian's Digest, must to a large extent consist of responsa which are closely linked to legal practice in general and to case law in particular." Ibid., 5. Tellegen-Couperus's argument is bolder than Jolowicz's because it links decisions directly to the writings of the jurists. She therefore confronts the common opinion on its own terms, asserting that decisions were indeed contributing content to jurists' law. Jolowicz, in contrast, more or less leaves it up to the reader to speculate by what means precedent played a role in the development of the law. What distinguishes Tellegen-Couperus's view from the usual view of responsa is her suggestion that the jurists (to some degree) took their cue from the judges, not the other way around. Compare Schulz, Roman Legal Science, 224–25; J. M. Kelly, Studies in the Civil Judicature of the Roman Republic (Oxford: Clarendon Press, 1976), 75–76. ^71. Tellegen-Couperus, "The Role of the Judge," 1–2, writes of "three assumptions," but the third assumption (that judgments were irrelevant to the development of the law) is intended to follow from the first two. ^72. Ibid., 2–3. ^73. J. Q. Whitman, The Legacy of Roman Law in the German Romantic Era (Princeton: Princeton University Press, 1990), 130. ^74. Ibid. ^75. G. F. Puchta, Cursus der Institutionen [Geschichte des Rechts bei dem römischen Volk, vol. 1], 9th ed. (Leipzig: Breitkopf & Härtel, 1881), 435: Es wäre ... ein Irrthum, wenn man die römischen Judices mit den heutigen Geschwornen vergleichen, und sich vorstellen wollte, sie seien bloß mit der Untersuchung des Factischen beschäftig gewesen, die Rechtssätze seien ihnen durch das Verfahren in iure vorgezeichnet worden.... Der Magistrat entschied allerdings über die allgemeine rechtliche Begründung des Anspruchs, indem er die Klagen und Einreden zuließ, und das Iudicium ordnete, aber selbst bei der allereinfachsten Klage, der auf eine bestimmte geschuldete Geldsumme, konnte sich noch mancher Anlaß zu rechtlichen Fragen im Iudicium finden, noch mehr war dieß der Fall bei den Klagen, in welchen durch die Anweisung, zu untersuchen, was eine Partei der andern ex fide bona zu leisten habe, dem Richter ein weites Feld rechtlicher Erwägungen geöffnet war, und eben so bei dinglichen Klagen, z. B. der rei vindicatio, wo der Richter nur im allgemeinen angewiesen war, zu untersuchen, ob der Kläger Eigenthümer sei, wo also die ganze Rechtstheorie der einzelnen Erwerbsarten in Frage kommen konnte. ^76. See Franz Wieacker, Römische Rechtsgeschichte (Munich: Beck, 1988), 1:667; Dawson, The Oracles of the Law, 104; Buckland and McNair, Roman Law, 402; Édouard Cuq, Les Institutions Juridique des Romains (Paris: Plon, 1902), 1:758–59; Julius Baron, Institutionen und Civilprozeß [Geschichte des römischen Rechts, vol. 1] (Berlin: Leonhard Simion, 1884), 354; Lord Mackenzie [Thomas MacKenzie], Studies in Roman Law, 4th ed. (Edinburgh: William Blackwood and Sons, 1876), 340. Cf. Tellegen-Couperus, "The Role of the Judge," 11 n.2. ^77. See Patrick Colquhoun, A Summary of the Roman Civil Law (London: William Benning, 1849), 1:34 (The judge "judged principally of facts," but points of law must have arisen also). ^78. Tellegen-Couperus, "The Role of the Judge," 1, 3–4. A fairly clear statement of this assumption is in Buckland and McNair: "In a system in which the iudex was not a lawyer, but a private citizen, little more than an arbitrator, it would be impossible for his judgements to bind." Buckland and McNair, Roman Law, 6. Compare J. A. Crook, Legal Advocacy in the Roman World (Ithaca: Cornell University Press, 1995), 175 n.18 : "The claim ... that unus iudex was not a lawyer needs re-phrasing: say, rather, he did not have to be, and there were no career judges." ^79. Tellegen-Couperus, "The Role of the Judge," 3–4. ^80. Note 46 above and accompanying text. ^81. Kaser, "Das Urteil als Rechtsquelle," 127. ^82. Arthur Engelmann, "Modern Continental Procedure," in A History of Continental Civil Procedure, ed. Arthur Engelmann et al. (Boston: Little, Brown, 1927; reprinted New York: Augustus M. Kelley, 1969), 598, 606. As in private law, both romanists and germanists were among the writers on procedure, and a rivalry existed. See K. N. Nörr, "Wissenschaft und Schriftum zum deutschen Zivilprozeß im 19. Jahrhundert," in Iudicium est actus trium personarum: Beiträge zur Geschichte des Zivilprozeßrechts in Europa (Goldbach: Keip, 1993), 147–49. ^83. Engelmann, "Modern Continental Procedure," 543. ^84. Book four of Gaius's Institutes was particularly helpful on the archaic legis actio procedure (4.11–29), the contents of the formula (4.39–68), and the bringing of a lawsuit (4.183–187). ^85. Giuseppe Chiovenda, "Roman and Germanic Elements in Continental Civil Procedure," in A History of Continental Civil Procedure, 78–79. Bethmann-Hollweg's research program is summarized in Nörr, "Wissenschaft und Schriftum," 144–46. ^86. The most famous are A. W. Heffter, Institutionen des römischen und teutschen Civil-Processes (Bonn: Adolph Marcus, 1825); F. von Keller, Der römische Civilprocess und die Actionen (Leipzig: Bernhard Tauchnitz, 1852); and M. von Bethmann-Hollweg, Der Civil-prozeß des gemeinen Rechts in geschichtlicher Entwicklung (Bonn: Adolph Marcus, 1864–74), 6 vols. ^87. Nörr, "Wissenschaft und Schriftum," 146, 148–49. ^88. R. W. Millar, "The Formative Principles of Civil Procedure," in A History of Continental Civil Procedure, 5–6 (first published in 1923). ^89. Millar, "Formative Principles," 11–21; R. C. van Caenegem, History of European Civil Procedure [International Encyclopedia of Comparative Law, vol. 16, ch. 2] (Tübingen: Mohr, 1973), 14; Anke Freckmann and Thomas Wegerich, The German Legal System (London: Sweet & Maxwell, 1999), 142. ^90. Millar, "Formative Principles," 5. But see William B. Fisch, "The Influence of German Civil Procedural Thinking and of the ZPO in the United States," in Das Deutsche Zivilprozessrecht und seine Ausstrahlung auf andere Rechtsordnungen, ed. W. Habscheid et al. (Bielefeld: Gieseking, 1991), 400–415. ^91. A good illustration of this is a joint project between UNIDROIT in Rome and the American Law Institute to harmonize procedural law at an international level, principally for trade matters. See Conference on the ALI-UNIDROIT Principles and Rules of Transnational Civil Procedure (London: Sweet & Maxwell, 2002). ^92. Van Caenegem, History, 93. Another example, from several decades earlier: when leading a ministry for the reform of Prussian laws in 1845, Savigny had prepared a set of discussion points for the reform of Prussian civil procedure. The "eighteen questions of principles" all draw on well-developed ideas. K. W. Nörr, "Die 18 Prinzipienfragen des Ministers Savigny zur Reform des preussischen Zivilprozesses," in Iudicium est actus trium personarum: Beiträge zur Geschichte des Zivilprozeßrechts in Europa (Goldbach: Keip, 1993), 181R

Roman Judges, Case Law, and Principles of Procedure

Ernest Metzger

Summons

Case Law?

Principles of Procedure

Immediacy and Orality

Decisions as Data

Precedent

Notes