1. Introduction

1The culture of adjudication, i.e. the culture of dispute resolution represents a considerable part of law and legal culture in whole and has specific importance in the history of law. Some of the earliest evidences on law specifically relate to dispute resolution process. According to G. Pfeifer, they date back to third millennium BC (Old Babylonian period).¹ No matter on the model of dispute resolution or whether it was judicial or extrajudicial, throughout the legal history it was a field of law with considerable significance. Dispute resolution has been a field of law cultivated and appreciated in various periods, different contexts and diverse legal cultures. In all legal cultures it was of the greatest importance to have the dispute finally resolved.² In the Western world Roman law first systematically dealt with the problem of settling disputes in a different way than going through usual controversial rulings constrained by judicial jurisdiction. In the West Roman law founded the culture of arbitral dispute resolution which have ever since (in Middle Ages, New Ages and nowadays) conceptually and terminologically been relying on the Roman legal tradition. Therefore, the Roman arbitration is an important part of the legal heritage pertinent to the culture of dispute resolution.

2Roman law took interest in arbitration because it was perceived as a means of dispute resolution which might have considerable advantages to civil litigation which was based on law. Arbitration diverted the disputing parties from excessive civil litigation into various modes of dispute resolution that were considered to be less confrontational. Roman law recognized variety of extra-judicial mechanisms of dispute resolution to ensure that the controversies and differences between the disputing parties were indeed ended. In most cases such different arbitral mechanisms should not be perceived as alternatives to going to court. Evidences on Roman arbitration indicate important reasons that stimulated the parties to go beyond constraints of ordinary jurisdiction for resolving their disputes, conflicts and other differences. Recourse to arbitration was merely a reaction to disadvantages of Roman civil litigation. In certain legal matter the disputing parties were reasoning whether to go to court and to take over many unpredictable risks of civil procedure or, optionally, to use and eventually benefit from practical or psychological advantages of available extra-judicial mechanisms that were adequate means of resolving their dispute. In cases where civil litigation was too complicated (because of formalities and procedural stages), slow, beyond financial power of the disputing parties, risky, considerably subjected to possible risks of excessive claims (pluspetitio) or when civil justice was de iure or de facto inaccessible or unreachable, arbitration is shown as a necessity.³ Furthermore, civil litigation was guided by formal procedures, while arbitration relied on informal ones that were in practice easier to achieve, which promoted its attractiveness and accessibility.

3P. Stein explained that in certain cases the Roman formula in procedure per formulas (especially within bonae fidei iudicia) gave wide discretion to the iudex to award the plaintiff whatever the iudex thought he should gain according to the dictates of the good faith. On the other hand, P. Stein noticed that formula was always more rigid in cases of valuation of the debts, which were obviously very important matters that frequently appeared in the legal practice. A problem could rise in debt matters if the plaintiff was doubtful about how much he could prove that he was owed⁴ or if he did not know exactly how much he was owed or when he was in search for a means to determine the amount of what was owed to him. Such procedural circumstances (rigidness) related to formula in financial disputes facilitated the use of arbitration which is excellently attested in Tabulae Pompeianae Sulpiciorum.

4When the disputing parties arranged arbitration, they gave their consent to whatever award an arbiter might render. By arranging it they partially or completely (depending upon the exact type of arbitration) took over the control over the dispute resolution process because the arbitral proceedings were generally private or at least in most of their content. This is the reason why the arbitration is often described as a flexible means of dispute resolution adjustable to the practical needs, desires and interests of the disputing parties.

5Final settlement of the dispute was the main purpose and the ultimate goal of arbitration. When Justinian’s jurist compiled 8th title of the 4th book of the Digest, which is the most important source for arbitrium ex compromisso, in the first place they put Paul’s fragment saying that the functions of compromissum and iudicium are substantially analogue because they should both ultimately lead ad finiendas lites⁵, i.e. to the final dispute resolution. The science of Roman law identified the expression ad finiendas lites pertinet not as a classical text, but as an interpolation of Byzantine jurists of the 6th century.⁶ By adding this interpolation, the Byzantine jurist explicitly clarified and supplemented Paul’s wordings (which even in modern science of Roman law are found to be worth of discussing⁷) that the arbitration is not sustainable unless it finally ends the dispute. Roman law took interest in arbitration to ensure that the litigation was indeed ended by arbiter’s award. The award of the arbiter was adjudication – final and binding determination of parties’ rights and obligations. It had to finally end all disputes, differences and uncertainties between the disputing parties that had been mentioned in the arbitral agreement.

6Roman law did not define or recognize differences between categories of substantive and procedural law.⁸ This is the reason why the Roman jurists and the Roman sources in whole did not considerably discuss or describe the arbitral procedure, i.e. how arbitration operated in practice. Their primary concern was to discuss and identify perquisites and guarantees for arbitration to be practically operative. Therefore, not a single fragment is dealing with the arbitral procedure, decisive questions and facts of the procedure, the course of the proceedings and its handling. Most of evidences on the arbitral procedure derive from epigraphic sources (Corpus inscriptionum latinarum, Tabulae Herculanenses, Tabulae Pompeianae Sulpiciorum) which record actual arbitral cases, how they were procedurally operated and lead to final arbitral awards. These epigraphic evidences are primarily of documentary nature – they are meant to be permanent records of actual case and dispute resolution mechanism and therefore contain no jurists’ comments and discussions.

7This paper will briefly examine the key features and common procedural guidelines of the Roman arbitration. It will discuss arbitration from the procedural point of view, especially how different procedural mechanisms and techniques were implemented to achieve goals which made arbitration distinctive and attractive proceeding: accessibility, flexibility, low cost effectiveness, rapidity and efficiency. It will give an insight on how the Romans balanced and kept in permanent tension different procedural values and goals to have the dispute finally resolved. The Roman jurists had no interest in defining or explaining procedural phenomena, values and ideas or in giving name to them. This does not mean they did not recognize practical meaning and importance of the procedure and the procedural values. In this paper such practical perceptions of procedural values which are common to all types of Roman arbitration will be examined as well.

2. An overview of the main guidelines of the arbitral procedure

8The arbitral proceeding was initiated once the arbitral agreement achieved its full legal effect. If the effect of the agreement was postponed, depending upon the suspense condition, perquisites for an arbiter to handle the proceeding would be fulfilled when the conditional circumstance actually happened.⁹ The arbitral agreement could define the exact place of arbitration. If the disputing parties did not explicitly determine the location of arbitration, the place where the disputing parties reached the arbitral agreement was considered to be a place of the arbitration. If the disputing parties did not set the time limits (within what time the arbitration should be ended), an arbiter was empowered to make a choice with regard to all circumstances.¹⁰ An arbiter would set the period of time which was convenient (modicum), i.e. reasonable to handle the proceeding and render the arbitral award. The necessary period of time was not defined by hypothetical legal rule or determined in practice. An arbiter would set the period of time which was convenient enough for him to handle the proceeding and to render the award. Its determination should conform to general perceptions of what was practical, expedient and reasonable in actual case considering complexity of the legal matter, circumstances, position and interests of both disputing parties. Once a person accepted his arbitral position (receptum arbitri) and set the necessary period of time, he was bound by what he had undertaken, which was enforceable by public authorities, i.e. by praetor.¹¹ Texts of Tabulae Pompeianae Sulpiciorum record arbitrations in which parties had probably given instructions to an arbiter to resolve their dispute immediately – at one single hearing held in a same day.¹²

9Arbitration in Roman law gives opportunity to the parties to have their dispute resolved in a language they understand. On the question of language arbitration was fully adaptable and adapted to the actual needs and desires of the disputing parties.¹³ The parties could choose the language they found convenient and understandable for their mutual procedural communication. There were no constraints for the parties in provinces or in those territories where Roman government was not fully established to select Latin language and to resolve their dispute following Roman legal and terminological standards.¹⁴ There is substantial number of evidences on arbitrations conducted in Greece during the Roman rule that were exclusively recorded in Greek language.¹⁵ The arbitral files written exclusively in Greek show a general trend toward usage of Greek language in the arbitral proceeding as well. D. Roebuck and B. De Loynes de Fumichon stated that some papyri show that all stages of the arbitral process were recorded in Greek and some in Coptic and other provincial languages.¹⁶ With a high degree of certainty we can reach the conclusion that numerous arbitrations on boundary disputes in Roman province of Dalmatia between different Liburnic and Dalmatian communities that took place during the first half of 1st century were conducted at least partially in provincial languages of local tribal communities, though the inscriptions of final settlement are exclusively written in Latin.¹⁷ On the other hand, Tabula Contrebiensis (87 BC), a famous epigraphic record of the arbitration between tribal communities in Spain in 1st century BC, shows an opposite example: the disputing Celtiberian tribal communities agreed not to use their provincial languages in the dispute resolution process but rather to benefit from technical Roman legal language and clear concepts of Roman law.¹⁸

10D. Roebuck and B. De Loynes de Fumichon argued whether the disputing parties could arrange the rules of the procedure by themselves.¹⁹ This was the most important means by which they voluntarily controlled the arbitration in whole and the procedural peculiarities as well. The disputing parties defined the scope of arbiter’s powers in the arbitral agreement. If the arbitration was primarily private (arbitrium boni viri and arbitrium ex compromisso), they could control an arbiter (scope and exercise of his powers) and the procedure in whole. Though, there are no records in arbitral practice of the agreements explicitly defining general rules of the arbitral procedure or prescribing the procedural techniques that an arbiter was obliged to follow. The choice of the procedural rules was theoretically possible, but it seems that the procedure and the modus operandi were in main principles modelled similarly to ordinary judicial process.

11Exclusion of publicity was a distinctive attractiveness of Roman arbitration which facilitated protection of secrecy, confidentiality and an intention to have the dispute settled sine strepitu forensi. This procedural value is well recorded by Pliny the Younger on the example of the hereditary dispute of Asudius Curianus decided by three arbiters who acted as boni viri. The hearing and interrogations were conducted in private premises and the arbitral award was rendered and proclaimed to the party with legal interest²⁰. The same ratio of arbitration is expressed in ex compromisso proceedings. Arbitration ex compromisso intended to avoid publicity which regularly attended public hearings in ordinary judicial proceeding.²¹ Its purpose, among others, was to protect two major interests: 1) parties’ intention to have their dispute finally resolved; 2) confidentiality. Ulpianus, one of the most appreciated Roman classical jurist, in his commentaries to praetor’s edict said the arbitration ex compromisso needed not to expose secrets of the business transactions (secreta negotii) and of the personal identities of the disputing parties (intima).²² Arbitration ex compromisso could not result with infamy (infamia) of any disputing parties. The perquisite for the judicial process to result with infamy was the publicity of proceedings, i.e. that all the procedural actions were taken in public and that the award was publicly proclaimed. Exclusion of public in arbitration ex compromisso and denial of infamy were inevitably connected.

12The arbitral procedure was handled orally. E. Metzger noticed that orality generally “lessened the complexity and secrecy of a procedure dominated by writing”.²³ Orality made the procedure accessible, understandable and adaptable to needs and desires of the disputing parties and to circumstances that could eventually change. It promoted the use of mother tongue, i.e. spoken language. It affirmed the active role of the disputing parties and arbiters in the proceedings. Epigraphic content of Tabulae Pomeianae Sulpiciorum concerning iudicium arbitrale ex compromisso suggests that audire (ad audiendum) was the leading operative principle. It suggests that arbiters scheduled hearing with greatest care and attention in order to provide the arbitral process with all the necessary perquisites to promote orality as a major procedural value. Orality promoted hearing of all arguments and claims and comprehensive discussion of the legal matter. It allowed the disputing parties asking questions (interrogatio), making statements and comments on what arbiter(s) or counterparty had previously said, making counterclaims, demands for explanations, mutual interrogations of the disputing parties.²⁴

13Concentrated proceeding was distinctive feature of Roman arbitration which affirmed its low cost effectiveness, efficiency and rapidity. Arbitral files documenting its operation in practice indicate arbiter’s duty to schedule hearings frequently and continuously. Such trend was an intentional procedural method which supported an integrated and concentrated debate, permanent awareness of decisive facts in dispute and of the evidences presented. Different mechanisms were used to achieve concentrated arbitral proceeding. Tabulae Pompeianae Sulpiciorum indicate that scheduling the hearings (dies ad audiendum datio) was arbiter’s important procedural task. Hearings were scheduled cautiously with all the necessary information proclaimed to the disputing parties in their presence.²⁵ No hearing was sustainable unless all procedural subjects were present, which explains the need of the accuracy of schedule. The rule was softened by providing the disputing parties with possibility to be legally represented. Awards were always rendered after the last hearing had been completed and generally no delays were tolerated. Classical jurists paid considerable attention to defining the time limits of arbiter’s competences and arbitration in whole. The clauses in compromissum could allow an arbiter to extend the time or to make proper postponements.²⁶ Though, these were exceptions, which means that an award rendered out of the time was invalid and invalidity could not be avoided by an unauthorized postponement.²⁷ An arbiter could do nothing outside powers given to him by compromissum, i.e. outside the time limits set by it.

14Efficiency, low cost effectiveness and procedural rapidity were promoted by immediacy. Immediacy is one of the major guidelines and working principles pertinent to all types of the Roman arbitration. It was facilitated by several rules which were the determinants and essentials of the arbitral procedure. Subjects in the proceeding were bound to be present at each procedural act and hearings.²⁸ Absence of 1) arbiter, 2) single or more arbiters of the arbitral commission 3) any or both disputing parties unconditionally stopped the proceedings and sequence of procedural phases. This rule was softened with a provision that the parties may be legally represented (procuratores, patroni causarum) in arbitral proceedings²⁹ or, exceptionally, that they were allowed make observations and claims in a letter.³⁰ Subsequently, no absence excuses were sustainable. Subjects in the proceeding were bound to attend inspection which was handled on disputed place.³¹ Disputing parties were allowed to be legally represented in inspection.³² The arbiter’s position was strictly personal – he could not be substituted.³³ If there were multiple arbiters, none of them could be excluded or substituted by another. Parties could add additional arbiter(s) neither or make reselections. Only those arbiters who participated at hearing could render the award.³⁴ Under this rule it is a virtue for a decision to be made by the very person who considers the proof and hears the parties. The award had to be rendered by an arbiter himself or by the person whom he had previously delegated for this purpose. It had to be proclaimed orally in presence of both parties or, if they were absent, in presence of their legal representatives. Otherwise the decision had no legal effect.

15Succession into the position of the disputing parties was possible only if it had been previously explicitly foreseen and determined in arbitral agreement. Procedural succession in arbitration was very rare, which supports the views on immediacy regarding to disputing parties as well. Though, procedural succession is uniformly documented both in Digest and in epigraphic sources. This type of succession is documented when the disputant in compromisso anticipated that he should be succeeded by his heir (heredemve). The succession of heir was not a succession mortis causa, but a conventional procedural succession which was foreseen and arranged in the arbitral agreement. Identification of the exact heir who succeeded the disputant was a matter of succession mortis causa.³⁵

3. Arbiter’s role in handling the procedure

16Roman arbitration handled by boni viri and arbiter compromissarius was an adversarial proceeding with an active role of an arbiter. Once the procedure was agreed, an arbiter had wide competences in handling the procedure, which is, amongst other features, recognized as a distinction between him and iudex.³⁶ There were no procedural determinations of what he was supposed to do or fixed constraints in handling the procedure unless the parties themselves had arranged certain clauses in the arbitral agreement by which he was bound. In the proceedings an arbiter was allowed to do what he thought was appropriate, convenient and practical, as well as what was objectively just and legal for arbitration to reach the goal of rendering the final award. Selection of different modes, methods and mechanisms of dispute resolution was left to his discretion. An arbiter was a person (private individual) not a function which allowed him to exercise his discretion without being bound with ius civile. The idea of arbiter’s discretion, which is in literature often proclaimed as a recognizable feature of his activity in rendering the award and applying the substantive law, should be perceived as a working method in exercising his procedural competences. He could choose whether to conciliate, mediate, adjudicate, advise, determine legal matter as an expert etc. Postclassical imperial constitutions and letters of Cassiodorus regarding arbitrium boni viri document inspections on the disputed place where an arbiter introduced the disputing parties with how he will act, what his modus operandi will be, on what they could expect from him and remarks on interpretations he will give.³⁷ The constitutions of the emperor Constantius indicate arbiter’s authority to encounter procedural abuses in inspections. Arbiter’s duty was to schedule the inspection and to call the disputing parties to be present. If one of the disputing parties was intentionally absent with aim to prevent arbiter and other party in taking the procedural actions, an arbiter could continue the proceeding without stopping the inspection.³⁸

17In handling the proceeding an arbiter made procedural interlocutory decisions (de preparatione causae).³⁹ K.-H. Ziegler, whose views were followed by R. Knütel, differentiates arbiter’s final awards which he describes as sententiae quae arbitrium finiunt from procedural arbitral decisions (sententiae de preparatione causae) by which the disputing parties were given command or imposed a penalty (poena compromissa)⁴⁰. In arbitrations ex compromisso an arbiter could impose a penalty (poena compromissa) to a party liable for not complying with arbitration, arbiter’s interlocutory decision or not fulfilling his final award.⁴¹ Within the limits of compromissum and for purpose of achieving its goals an arbiter could make the decisions, issue warnings, commands and interdictions.⁴² These procedural decisions could be enforced by imposing the poena compromissa.⁴³ The Digest records a penalty imposed for inexcusable absence⁴⁴ and for disregarding the deadline set by an arbiter.⁴⁵ An arbiter was expected to handle the proceeding, schedule the hearings and render the award in those days when identical or similar proceedings were handled by the state courts. An arbiter would not be obliged to act outside these time determinants even if such clause was inserted into the text of compromissum. Though, this was left to his absolute discretion⁴⁶ and if he wanted, he could act outside these time determinants.

18In handling the procedure an arbiter should be concerned with ne propagentur arbitria⁴⁷. Classical jurist Paulus proclaims that an arbiter should render the award having in mind that arbitrations should not take too long to resolve the dispute. Though, ne propagentur arbitria should be understood as a general procedural value and guideline of arbiter’s activity in time which facilitated rapidity, efficiency and low cost effectiveness. The idea ne propagentur arbitria must be observed together with Ulpian’s fragment mentioning modicum tempus because they both set the criterion for the arbitration to be ended within convenient, i.e. reasonable time. Ulpianus claimed that the disputing parties should execute arbitral award within the reasonable time (modicum tempus).⁴⁸ The idea of modicum tempus should be understood not as a chronological constraint of award's execution only, but also as a guideline for arbiter’s procedural activity in arbitration. Among other values, an arbiter should appreciate what was practical, economical, easy to apply, low cost effective, rapid and convenient for both parties in dispute and for the actual arbitration in whole.

19The adversarial proceeding was promoted by hearings which substantially became operative through contradictory debate handled and controlled by an arbiter. The Herculanean tablet recording an arbitral hearing held on 26th January 69 AD documents procedural peculiarities of the arbitration ex compromisso:

„Interogationis L. Appuleium Proculum haberente palos CCC sex
qui fuerant depositi aput Nonium Primigenium“.
„L. Cominius Primus interogavit L. Appuleium Proculum haberente a se palos CCC sex ex sententia Ti. Crassi Firmi arbitri in controversia, quae fuit de finibus fundi Numidiani L. Comini Primi et fundi Stalsaniciniani L. Apulei Proculi, quos deposuissent apud M. Nonium Primigenium. Ibi L. Appuleius Proculus respondit: Ego meos palos CCC sex caesos a te, qui fuerant depositi aput M. Nonium Primigenium, recepi ex sententia Ti. Crassi Firmi arbitri“.⁴⁹

20This epigraphic evidence on Roman arbitration ex compromisso contains reference to the procedural value which even today the legal science identifies as audiatur et altera pars (Ger. rechtliches Gehör; Fre. le droit de la défense). Romans did not recognize it as theoretical principle, but they surely felt its practical meaning in judicial proceeding and in arbitration as well⁵⁰. Its practical content manifested in different types of Roman arbitral proceedings, which is probably best documented in aforementioned Herculanean inscription. In handling the procedure an arbiter had to stimulate comprehensive discussions between the disputing parties and all the other procedural subjects: asking questions and answering them, making claims and counterclaims, giving explanations, interpretations, expressing different views on the facts, giving judgements on the legal of factual content etc. He could even allow cross examinations, which altogether gave considerable contribution to fair and just administration of justice, which was one of the main purposes of the Roman arbitration. The Herculanean inscription indicates each disputing party had to be given opportunity to present its arguments and a chance to adequately respond to arguments of the opposing party. Subsequently, it means that each disputing party could submit all relevant evidence and demonstrate decisive legal facts and, on the other hand, submit counter-arguments on the claims and evidence of the opposite party. If further importance of such views is observed within the Roman arbitration, we can define such modus operandi as a means by which comprehensive pursuit for the truth was promoted.

4. Schedule of hearings

21Camodeca’s publication “Tabulae Pompeianae Sulpiciorum” (TPSulp.) gives a detailed and accurate insight on how the hearings were scheduled and what their importance was. Its chapter entitled “Iudicium arbitrale ex compromisso” contains inscriptions which document scheduling of hearings (dies datio ad audiendum) and the perquisites for the hearing to be attended by the disputing parties.

22The arbitral proceeding was usually fragmented into hearings, unless the parties agreed that the dispute should ultimately be resolved at a single one. Hearings were scheduled by an arbiter in presence of both parties who could be legally represented. Epigraphic evidences⁵¹ suggest several data that should be included in the schedule: (1) arbiter’s name and remark he was acting ex compromisso; (2) full names of the disputing parties; (3) remark that the hearing was scheduled in presence of both parties (or by their legal representatives) – utroque praesente; (4) command (iussum) for the parties to be present (adesse) at the hearing; (5) the exact day (dies) when the hearing will take part; (6) exact hour (hora) when the hearing will begin; (7) the purpose of the hearing expressed by the wording ad audiendum, which indicates the hearing were conducted orally; (8) the exact place (actum in...), day (dies), hour (hora) and year (consulibus...)⁵². The fragments in Digest proclaim the arbiter could schedule the hearing and issue command to the disputing parties to be present or by messenger or in a letter (arbiter adesse litigatores vel per nuntium vel epistulam iubere potest)⁵³. An arbiter could postpone the hearing, which he could do in person, by messenger or in a letter (Diem proferre vel praesens vel per nuntium vel per epistulam potest)⁵⁴.

5. The disputing parties and their representatives

23Roman arbitration did not follow the determinants of civil litigation which knew of two disputing parties whose procedural positions were dependent upon the fact whether they were plaintiffs or defendants. Arbitration was primarily a means of dispute resolution which paid considerable attention not to the procedural positions (actor and reus) and claims (petitiones), but rather to parties’ interests. Furthermore, Roman arbitration did not necessarily have two parties in dispute. In the aforementioned inheritance legal matter of Ausdius Curianus, which is recorded in Pliny’s letter to Annius Severus, there was one party only. This arbitration did not resolve the dispute, but it examined whether the disinheritance of Asudius Curianus was legally sustainable and properly handled.⁵⁵ Though, all the other evidences on arbitration document two parties in arbitral dispute, but no constrain was set. Therefore, especially in boundary dispute (for example in the case of trifinium), it is imaginable that there could be more than two disputing parties, each with different claims and interests.

24Party’s position in arbitration was strictly personal, no changes or successions could take place. Compromissa that are preserved as well as records of scheduling dies ad audiendum indicate that the disputing party were always mentioned by their full name, which leads to the conclusion that their personal identity was essential. Ulpianus explicitly says that compromissum (and consequently the arbitration ex compromisso as well) ceases to exist if one of the parties dies unless the succession into his position had been foreseen by compromissum.⁵⁶ The succession of party’s procedural position was possible only if it was defined by a special clause in compromissum.⁵⁷ Practical realization of such circumstances is well documented in inscription of one Herculanean tablet. When the identity of the disputing party was documented in compromissum it was said L. Cominium Primum heredemve eius⁵⁸, which means a party named Lucius Cominius Primus can be proceduraly succeeded by his heir. The possibility of succession had nothing to do with succession mortis causa because it was provided by mutual agreement of the parties expressed in the text of compromissum. Therefore the legal ground for procedural succession was compromissum, which means this succession was not pertinent to inheritance law. Inheritance law provided legal perquisites by which a successor could be identified in person. Though, his involvement in the arbitral procedure was completely dependent upon the content of compromissum.

25There are numerous records of juristic persons being the disputing parties in arbitrations. Considerable fund of boundary inscriptions from Roman province of Dalmatia records tribal communities as parties in boundary disputes, disputes over water rights and rights to access the water sources.⁵⁹ They are mentioned just as tribes (Ortoplini, Parentini, Corinienses, Salviates, Stridonenses, Nediti) or occasionally as res publicae and identified by their tribal names (res publica Asseriatium, res publica Alveritarum). The expressions res publicae on boundary stones should be understood as self-governing tribal communities which were awarded with a certain degree of municipality⁶⁰. In dispute resolution proceedings tribal communities were recognized as parties entitled to stand before an arbitral tribunal and to take all the procedural actions necessary to protect their rights. Although some of them (especially those communities which did not reach the status of res publicae recognized by the Romans) were not considered to be juristic persons, in judicial or arbitral proceedings they were regularly awarded with the status of the party. In arbitral proceedings they had the full capacity (ius standi in iudicio) recognized on an ad hoc basis only in a single dispute⁶¹.

26The second important source on juristic persons as disputing parties in Roman arbitration is the Histonian inscription:

27C. HLEVIDIVS PRISCVS ARBITER EX CONPROMISSO INTER Q. TILLIVM ERYLLVM PROCURATO REM TILLI SASSI ET M. PAQVIVM AVLANIVM ACTOREM MVNICIPI HISTONIENSIVM VTRVSQVE PRAESENTIBVS IVRATVS SENTENTIAM DIXIT IN EA VERBA QUAE INFRA SCRIPTA SVNT⁶².

28The inscription records a dispute between a private individual who was legally represented on the contractual ground. On the other side it records municipium Histoniensium which was a party in dispute and represented by the head of municipal community. Actor Marcus Paquius Aulanium was entitled by his position to take legal actions in arbitration ex compromisso handled by an arbiter. This means that actormunicipii had previously arranged compromissum with a private individual. The inscription indicates difference between municipium Histoniensium who was the disputing party and an actor who procedurally acted on its behalf.

29A party was free to conduct his case personally and at the same time to be assisted in arbitration. The assistant was legal advisor or an expert or a person who had better understanding of the dispute. In their absence, the disputing parties could be represented in arbitration, which is well documented in Roman legal sources.⁶³ Legal representation i arbitration should be observed from technical point of view – where a party availed himself of help – and as a means to soften strict rules on immediacy which required parties’ presence to all procedural acts. Legal representation was a substitute to parties’ presence in personam. The Romans realized that rigid rules on being personally present to all hearings eventually would stop arbitration, especially if the legal matter was complicated to resolve or if many hearings needed to be held on different place with purpose to discuss the dispute comprehensively.

6. The issue of the double procedure

30A question could be raised whether the disputing parties in arbitration could initiate some other proceedings (arbitral or judicial) for the purpose of resolving an identical dispute between identical parties. The classical Roman jurists had no clear understanding of this issue: Paulus explained the case in which the parties agreed upon ex compromisso arbitration and later, during the arbitral proceedings, one of the parties abandoned the arbitration and initiated proceedings before the state court.⁶⁴ Paulus understood this as legally possible, but a contractual penalty (poena compromissa) was imposed nonetheless. Herculanean tablets record a prohibition of initiating successive arbitral proceedings for an identical dispute between identical parties. If both parties abandoned the previous arbitration to initiate another, a penalty could be imposed on them.⁶⁵ A classical Roman jurist proclaimed that good faith does not allow the same thing to be exacted twice.⁶⁶ This point of view was applicable to arbitration because it was a means of dispute resolution firmly grounded on good faith and equity. Quintilian’s principle was bis de eadem re agere non licet.⁶⁷ The identical legal matter should not be legally examined or resolved twice.

7. The procedural context of award’s rendering

31The final award (sententia) had to be rendered by an arbiter himself or by a person whom he had delegated. It had to be proclaimed orally in presence of both parties or, if they were absent, in presence of their legal representatives. Otherwise the award had no legal effect. Rendering of the final award ended the arbitration and the ad hoc arbitral tribunal ceased to exist because they both accomplished their purpose. By rendering of the award competences of the arbiter ceased to exist (arbiter esse desierat⁶⁸) because they were consumed, which meant that the arbiter could no longer impose penalties or exercise coercion.

32Appeals were excluded from arbitration, which is analogue to civil litigation in Roman law (process per formulas). The case once decided before the arbitral tribunal could not be reheard or revised on the grounds of the arbitral appeal or recourse to appellate instance. D. Roebuck and B. De Loynes de Fumichon argued whether an appeal was possible if the disputing parties provided in their agreement that disappointed party could appeal.⁶⁹ Such view, though, can not be verified in Roman legal sources or in documents that record arbitral practice. Recourse to another arbitral tribunal on the grounds of the same arbitral agreement was allowed neither. All the Roman arbitral tribunals were established ad hoc, they were not permanent tribunals and would cease to exist immediately as they fulfilled their purpose by resolving the dispute. Therefore, second instance did not exist and the arbitral tribunal that had rendered the award existed no more. Furthermore, the idea of settling dispute finally by means of arbitration denied that the award could be contested anyhow. In the late classical period of Roman law the emperors enacted rules that no arbitral decision could be contested at state court (previously it was possible, but a penalty – poena compromissa – was imposed to the party who initiated the court proceedings).⁷⁰ Though, legal practice indicates this was not always or consistently followed.⁷¹

33Strictly legal point of view reveals the arbitral award was not a procedural obstacle to initiate judicial proceeding in the identical legal matter between the same disputing parties. Arbitrations handled by arbiter bonus vir or arbiter ex compromisso were private proceedings pertinent to extra-judicial sphere. This means the arbitral and judicial authorities were not interdependent or somehow connected, which raises a question on their concurrence. Contractual clauses recommended by Cato⁷² and Pliny’s epistolary record of the inheritance dispute of Asudius Curianus⁷³ explicitly say the judicial proceeding could be initiated by pleading an action in the identical legal matter in which the arbitral award had been rendered.

34Arrangement of poena compromissa should primarily be perceived as a mechanism to divert the disputing parties in arbitration to initiate court proceeding during or after arbitration. It could be imposed to the disputing party who did not execute the award or to the party who regardless of the arbitral award initiated court proceeding in the identical legal matter. Poena compromissa was not de iure mechanism which prevented parties to take actions at court, but it had considerable practical (property) and psychological effect on their behaving and legal acts. If the parties wanted to disregard the arbitral award and to plead an action at court, they would be liable for paying poena compromissa which was regularly set in much higher value than the value of the disputed matter actually was.⁷⁴ Papinianus identified in terrorem function of poena compromissa, which was noticed and commented by R. Zimmermann. In terrorem function should be perceived as the psychological purpose of poena compromissa that became operative through parties’ fear of property loss as a consequence of rejecting performance of the arbitral award. It was in the hands of the each disputing parties either to perform what was proclaimed by the award or to pay the penalty (which was intentionally stipulated as in terrorem of the offending party) and resort to civil litigation.⁷⁵

8. Conclusion

35Deeper insight into the sources of Roman law indicates there were different types of arbitration with different functions, mechanisms of dispute resolution and methods of pursuit for the truth. Roman law in concreto knew not of universal concept of arbitration.⁷⁶ The Roman legal practice indicates cumulative existence of different types of arbitrations among which the disputing parties could select the one they considered convenient for resolution of their dispute. Though diversities can be discerned among different types of arbitrations, the procedural guidelines and values of Roman arbitration were common, general and universal, independently on its exact type. The arbitral procedural values and the perpetual tensions that obviously existed among them were specific and in many aspects different than the procedural determinants of civil litigation. Such procedural determinants affirmed arbitration as a proceeding that in certain circumstances could be more attractive and even the primary or the only available and rational means of dispute resolution.

36When M. Godfrey wrote an article on arbitration in ius commune and Scots law he stated that “Arbitration seems to imitate the formality of court procedure, taking account of legal rules, and yet operate primarily within the private sphere, forming a bridge between what could be called public and private justice“⁷⁷, which is applicable to the concept and the legal nature of Roman arbitration as well. Roman arbitral procedure has unfortunately never been systematically approached or analytically studied in whole. Digest 4.8. contain many fragments with opinions on its practical operation. This title of Digest, together with its other fragments and many different types of sources of Roman law, suggest de facto existence of the law of arbitration, i.e. special rules of procedure in Roman law pertinent exclusively to arbitration.