ARTICLES

 

Primitive prohibition of direct representation in Roman law scholarship: origins, sources and flaws

 

 

Javier E Rodriguez Diez

Assistant Professor of Roman law, Pontificia Universidad Católica, Chile

 

 

---

ABSTRACT

Roman law scholars since the nineteenth century have claimed that Roman law originally banned any form of direct representation, and that only through juristic innovations was this general prohibition of the ius civile partially overcome. Such theory was built on the assumption that some texts found in classical jurisprudence were manifestations of a general principle that inspired early Roman law. However, modern scholars have discarded many of the assumptions on which this theory was built, granting a much more limited scope to the texts which restrict the possibility to act on behalf of someone else. Moreover, the sources show that early legal institutions did not exclude agency-like figures, and that Roman jurists resorted to different criteria to determine whether the principal was affected by his agent depending on the particular legal act that was performed, for example the conclusion of a contract, the transfer of ownership, payment, acquisition of possession, etcetera. Accordingly, legal historians should avoid approaching the Roman sources through the notion of "direct representation". A piece-meal approach serves to understand when and under which conditions Roman jurists enabled an agent to affect the legal position of the dominus negotii.

Keywords: Direct representation; agency; voluntas domini; nuntius

---

 

 

1 Introduction

It is a commonplace among scholars since the nineteenth century to claim that Roman law forbade direct representation.¹ While there are different ways in which this idea has been presented, the common thread behind it is that early Roman law knew a negative principle which forbade anything what one would nowadays label as "direct representation". It is argued that the ius civile was initially too unsophisticated to allow that the legal position of an individual could be affected by the acts of another one acting on his behalf. However, at some point the needs of commerce would have urged jurists to develop exceptions to this primitive set of rules, particularly through the activity of the praetor. This would have been the case regarding, for instance, the actiones adiecticiae qualitatis.

Despite being broadly accepted, this general reconstruction has some troubling issues. Most of them are related to its inherently anachronistic phrasing. It is in fact odd to claim that Roman law prohibited a legal doctrine that would come into existence some two thousand years later.² It is, moreover, rather obvious that Roman jurists ignored the main features of such doctrine. Despite these self-evident remarks, scholars have made curious claims regarding what the alleged prohibition of direct representation meant for Roman law, which are directly related to the scope they grant to the doctrine of direct representation. Particularly relevant is the key role often granted to the contemplatio domini of the agent, that would determine whether - and why - a specific case fell under the prohibition on direct representation or not.³ Such idea pictures Roman jurists distinguishing between direct and indirect representation, or between Stellvertreter and Bote, distinctions that are troublesome considering the absence of analogous clear-cut dogmatic notions in the sources.

Another issue which rises when approaching the Roman sources through the notion of direct representation is the exact field of application of this theory - and its related prohibition - within Roman law. As it will be shown below, the doctrine of direct representation found its origin in the ius commune in the context of the law of obligations, and it was not until the nineteenth century that scholars applied it to other fields of patrimonial law, such as the acquisition or transfer of ownership. Following this trend, Roman law scholars approach different situations where a legal act is concluded by an intermediary as part of a common phenomenon of "direct representation", covering both the law of obligations and the law of property. This leads once again to the rather puzzling conclusion that somehow Roman jurists approached different situations - namely a pledge, sale, delivery, payment or any other act concluded through an intermediary - as part of a common phenomenon of "direct representation". Such idea is of course even more awkward considering that Roman law never developed a comprehensive doctrine which would simultaneously deal with every form of legal intermediation.

Leaving aside these remarks, perhaps the most problematic aspect of the theories dealing with "direct representation" in Roman law is the underlying thought that early Roman jurisprudence completely banned such notion, and that accordingly the ius civile did not allow any form of legal intermediation on behalf of the dominus negotii. Such idea was built on the interpretation of a handful of texts which to a large extent has become obsolete through the studies of later scholars. However, the general theory built around such texts has remained alive and well, influencing in turn the way in which scholars approach the evolution of numerous legal institutions.

In a recently published dissertation I have attempted to revise the idea that Roman jurisprudence had to overcome a primitive ban on direct representation regarding the transfer of ownership.⁴ The aim of this contribution is to show the wider picture of the problem of direct representation in Roman law, presenting the origins and flaws of this theory when approaching the sources. As a result, a more source-oriented analysis of the sources is proposed in order to present the piece-meal approach of Roman jurisprudence to the problem of agency.

 

2 Origins of the theory in German scholarship

Christian Friedrich Mühlenbruch was the first jurist to develop a general theory regarding the prohibition of direct representation in early Roman law. His ideas on the subject were presented in three successive editions of his work Die Lehre von der Cession der Forderungsrechte, between 1817 and 1836.⁵ However, the theory was to some extent already in the air. In fact, some of Mühlenbruch's starting points can be found in an earlier discussion regarding the possibility to perform actus legitimi - namely acts belonging to the ancient ius civile - through another person.⁶ This discussion can be traced back to the sixteenth century, when Jacobus Raevardus (1535-1568) claimed that the rule "nemo alieno nomine lege agere potest" of D 50 17 123pr⁷ necessarily implied that no actus legitimus could be carried out through another person.⁸ Raevardus attempted to demonstrate this regarding the cretio, refuting the ideas of Duarenus concerning the possibility of a procurator to acquire the hereditas. However, most of Raevardus' energy is devoted to proving that the mancipatio could only be performed by the owner. According to him, this circumstance would explain the role of the ancient fiducia (cum creditore or cum amico) since the only way in which the mancipatio could be performed by someone else was by transferring him ownership in the first place. These views were subject to discussion in the following centuries. For example, Conradi (1701-1748) denied Raevardus' textual support and conclusions regarding the cretio and the mancipatio,⁹while Averanius (1662-1738) upheld the idea that Roman law excluded that an actus legitimus - particularly the cretio, acceptilatio and mancipatio - could be concluded through an agent.¹⁰

This briefoverview shows that by the nineteenth century it was not an original idea that the ius civile was reluctant to various forms of agency. However, Mühlenbruch gave a much broader scope to this claim.¹¹ He approached through the notion of Stellvertretung various institutions in which someone carried out a legal act on behalf of someone else, such as the actiones adiecticiae qualitatis or the acquisition of ownership through a non-owner. He then proposed a general historical reconstruction for the evolution of Stellvertretung in Roman law, claiming that it was a general principle in early Roman law that no one could act on behalf of another as D 50 17 123pr would show. In support of his theory he brought the ideas of Averanius.¹² But he went even further, quoting other general statements¹³ such as "alteri stipulari nemo potest"¹⁴ and "per extraneam personam nobis adquiri non posse".¹⁵ Moreover, he also argued that originally ownership could only be transferred by the owner, which would explain the importance and role of the fiducia. Accordingly, Mühlenbruch saw a common thread running through early Roman private law, namely that everybody should perform legal acts personally.¹⁶ Such general principle would have proven problematic for trade, which is why Mühlenbruch proposed that all the cases where someone legally acts on behalf of another would have been exceptions to the original prohibition on direct representation. Such exceptions, however, would never have affected institutions belonging to the ius civile where the original ban on direct representation would have retained its validity.¹⁷

Mühlenbruch's historical reconstruction on the evolution of Stellvertretung in Roman law would be adopted and developed by Savigny. Already in his famous Vom Beruf unserer Zeit (1814) Savigny presented the idea of an early prohibition of direct representation,¹⁸ a theory which, as pointed out before, was not completely original to Mühlenbruch. However, Mühlenbruch does seem to be his main source of inspiration when Savigny further developed his theories in his System (1840) and later in his Obligationenrecht (1853),¹⁹ where he reproduced largely the ideas of Mühlenbruch, adding additional references from the sources and offering a more elaborate historical outlook. Perhaps Savigny's greatest innovation was to add a practical twist to the theory by claiming that it showed that there should be no obstacle for the acceptance of direct representation in legal practice in Germany. This had been an ongoing debate during the nineteenth century, which concerned specially the alteri stipulari-rule. Savigny would limit the scope of the texts which seemed to exclude direct representation by arguing that they belonged to the old ius civile, which excluded direct representation due to its formal and primitive character. However, he argued that Roman jurisprudence would have developed a series of exceptions to the general prohibition of direct representation, thus reducing its practical impact on trade. Such exceptions, in any case, would have only taken place regarding informal legal acts, since the formal legal acts of the ius civile - such as the stipulatio - remained under the old principle that excluded direct representation. Savigny's conclusion came thus as self-evident: yes, Roman law seems to exclude direct representation in a series of texts, but these refer only to an early prohibition which was overcome by Roman jurists themselves. Therefore, if already Roman law overcame this prohibition to a large extent, why should German legal practice have remained stuck with it? In fact, such ban on direct representation would be even more difficult to endure in the nineteenth century, since some Roman legal institutions that relieved this limitation had disappeared by that time, such as the possibility of the paterfamilias to act through his sons-in-power or slaves.²⁰

Savigny is, moreover, more explicit than Mühlenbruch concerning the scope of his conclusions since he claims that the doctrine of direct representation applies generally to patrimonial acts inter vivos?²¹ It is precisely this systematic approach which leads him to deal with different situations where an agent performs an act on behalf of the principal - the acquisition of possession, the conclusion of contracts, the transfer of ownership, etcetera - as part of a general problem of Stellvertretung which would develop a common evolution regarding all of them. This comfortably anachronistic approach served well the practical objective of his theory, advocating in favour of granting an unlimited application to direct representation in the German legal system.

Savigny's theories on direct representation had a dissimilar impact on later scholarship. On the one hand, many of his views concerning the dogmatic framework of direct representation for his own time were soon overshadowed by the contributions of authors such as Ihering and Laband. On the other hand, his historical reconstruction was enormously influential among Roman law scholars. This is all the more surprising considering not only that many of the starting points of his theory were abandoned by later scholarship - as it will be shown in the following section - but also because his theory had a rather evident practical aim. Nonetheless, even after legislation had served that practical aim, scholars upheld the theory that Roman law originally knew a prohibition on direct representation (Verbot unmittelbarer Stellvertretung), which was gradually overcome through juristic innovation regarding informal acts, while the formal acts of the ius civile remained bound to it. This idea is to be found among several nineteenth-century Roman law scholars, including Exner, Pernice and Sohm,²² but it found its most influential formulation in Mitteis' Römisches Privatrecht bis auf die Zeit Diokletians .²³ Many Roman law handbooks retained a special title regarding "direct representation" which largely reproduced the views of Mitteis, even if it seemed old-fashioned to make use of this pandectistic category when dealing with classical Roman law. Accordingly, many Roman law scholars continue to adhere up to this day to the main aspects of the theory concerning an early prohibition on direct representation in Roman law.²⁴

 

3 The theory's starting points fall, but the theory stands

As it was mentioned before, a curious fact behind the ongoing success of the traditional theory regarding direct representation in Roman law is that several of its original starting points have long been discarded. Many scholars may agree up to these days with the main conclusions on the subject offered by Mühlenbruch, Savigny and Mitteis, but few could share their approach to specific problems or texts. This is the case regarding, for instance, the possibility to carry out an actus legitimus through an agent. Mühlenbruch built his theory on earlier discussions concerning this point, claiming that no actus legitimus whatsoever could be carried out through another person. Savigny and Mitteis would also rely on this idea. However, Roman law scholars have long ago identified cases of actus legitimi carried out through an agent,²⁵ which immediately jeopardises the very foundations of a primitive ban on direct representation. We are aware, for instance, that a slave could acquire ownership for his master through mancipatio ²⁶Accordingly, not every actus legitimus falls under the rule "nemo alieno nomine lege agere potest". In fact, its palingenetic analysis shows that it was originally related to the centumviral courts,²⁷ where the legis actiones were still in use.²⁸ This shows that this rule was referred in particular to procedural representation under the legis actiones, and not even in this context did it have an over-arching validity.²⁹

The objections are even more serious regarding other foundations of the theory. For instance, Savigny sought confirmation for his theories in the analysis of D 41 1 53, which according to him would prove the fundamentally different approach to direct representation between formal acts of the ius civile and informal legal acts. Later scholars dealing with the acquisition of ownership and possession have long discarded the scope given to this text, which occupies a central position within Savigny's theory.³⁰ Moreover, the study of the acquisition of property and possession through an agent in general has undergone significant changes in the last century, with countless articles and monographic works dedicated to this topic. Several authors discard in fact that the per extraneam personam-rule had an overarching validity concerning the acquisition of ownership and possession through an agent.³¹On the contrary, such maxim would simply convey the idea that slaves and sons-in-power would automatically acquire for the paterfamilias, since they had no goods of their own, while the acquisition through a sui iuris would only take place if specific requirements were met. The per extraneam personam-rule would therefore have a limited scope within Roman law concerning the acquisition of property, and would not imply a general or ancient prohibition.

Concerning the law of obligations, the scope of the alteri stipulari-rule seems to have been restricted as well. Hans Ankum, who argued that the rule only concerned stipulations involving a dare oportere, showed this some decades ago in an influential contribution.³² The original limitation would stem from the features of the formulary procedure, where the stipulator would be unable to claim an obligation of dare that favoured a third party, since he would have no actionable interest himself. This is why most of the exceptions to this rule had to do with cases where the stipulator did have a pecuniary interest, thus rendering the stipulation valid. Further exceptions could only be introduced under compelling reasons, as it was the favor dotis.³³

The specific scope of the alteri stipulari-rule does not imply that the Roman law of obligations accepted agency in general and only had to face this particular limitation. One may indeed agree to some extent with the basic claim that the Roman law of obligations knew no such thing as direct representation, since it was never possible for an agent to bind the principal to a contract by acting on his behalf.³⁴This was not even the case regarding the actiones adiecticiae qualitatis, where the agent would still be personally bound to the contract. However, it is essential to bear in mind that this is a particular feature of the law of obligations, where it is in principle rather awkward that someone who does not personally conclude a contract is bound to its terms. In other words, one can understand the limitations of Roman law on agency in the context of the peculiar features of the law of obligations, and not because of an underlying principle that would span through the whole of Roman patrimonial law. This is why the alteri stipulari-rule should be approached on its own scope, and not as a manifestation of a wider prohibition of direct representation in early Roman law. The same can be said concerning the applications of the per extraneam personam-rule in the law of obligations. Moreover, regarding this latter principle, the sole fact that it is alternatively used within the law of property and the law of obligations,³⁵ in occasions coupled with the expression "vulgo dicitur",³⁶suggests a loose rule of thumb rather than an iron rule stemming from the ius civile.

Another maxim which is usually brought up as proof for the original prohibition of direct representation is Ulpian's nemo plus iuris-rule,³⁷ which allegedly would have excluded the transfer of ownership by a non-owner according to the ius civile for the case of the mancipatio and the in iure cessio..³⁸ Leaving aside the inaccuracies this implies for the evolution of the transfer of ownership by a non-owner - which will be reviewed in the following section - it is worth noting that the origin of this rule is by no means as ancient as it is usually assumed. In fact, the rule first appears in different philosophical and rhetorical writings,³⁹ from where it was adopted by jurists in order to enhance legal arguments in the most various contexts, including the law of possession, pledge, succession and of course the transfer of ownership.⁴⁰

Accordingly, both the origin and the scope of this rule discard that it had anything to do with an early prohibition on direct representation.

This brief overview shows that nowadays one cannot grant the same scope originally given to the various texts which once were seen as concluding evidence for an early prohibition on direct representation. On the contrary, these texts have a particular origin and scope of application that must be understood individually, and not as part of a common notion of "direct representation". It is even more far-reached to claim that this handful of texts would describe a general principle which underlies all inter vivos acts in early Roman law. The objections on this point are not only theoretical, but lead to serious distortions when approaching the sources, as it will be shown in the following sections.

 

4 A flawed approach to the sources

The traditional theory regarding the evolution of direct representation in Roman law has had a direct impact in the way scholars have approached the early stages of development of different legal institutions. Since there is not much information available concerning early Roman law, scholars have filled in the gaps through this general theory, often neglecting what little evidence is available in the sources. It is particularly the case regarding the transfer of ownership by a non-owner.⁴¹ As noted above, several scholars since the sixteenth century have claimed that a non-owner could not perform the mancipatio. Mitteis developed this idea in detail in the context of the mancipatio by slaves,⁴² which in turn gave place to wide controversy in the twentieth century.⁴³ However, most of the texts involved in the discussion are rather inconclusive, since it is almost impossible to determine whether the alienation takes place fiduciae causa or not.⁴⁴ Due to this innate ambiguity, personal preconceptions ultimately stir the discussion. Moreover, some scholars have dismissed on various assumptions what is perhaps the only truly conclusive text dealing with mancipatio by a slave.⁴⁵

The debate on mancipatio has also determined the approach to other related legal institutions. Particularly noteworthy is the analysis of the satisdatio or repromissio secundum mancipium, a mysterious archaic surety given in the context of the mancipatio. Ankum has conducted extensive research of these institutions, but his results rely heavily on his ideas concerning the impossibility of a non-owner to perform the mancipatio,⁴⁶thus leaving the door open for a further revision of the subject.

The impact has been much more widespread regarding the transfer of ownership by traditio. The traditional theory on direct representation considers that this was one of the cases in which direct representation was only developed at a later stage, on the grounds of being an informal act that would fall outside the prohibition of the ius civile. This, in turn, has affected the way scholars understand the development of various legal institutions. For instance, regarding those cases where the faculty to dispose is granted by a statutory provision, such as the curator furiosi or the tutor, it has been traditionally assumed that the legal guardian originally should acquire ownership over the administered goods in order to alienate them. Otherwise - it is argued - he would not be able to transfer ownership, due to the old prohibition of direct representation of the ius civile.⁴⁷ Other authors have toned down this idea by claiming that legal guardians could indeed transfer ownership over someone else's property, but could only transfer praetorian ownership, since Quiritary ownership could not be transferred through an agent.⁴⁸ There is, however, hardly any evidence in the sources to support such ideas, which is why scholars seem to rely almost exclusively on the theory of a primitive prohibition when making these claims.⁴⁹

The interpretations stemming from the traditional theory have been even more distorting concerning the cases where the traditio is authorised by the owner. Again, scholars have traditionally claimed that this could only take place through praetorian innovations, being impossible for a non-owner to transfer Quiritary ownership.⁵⁰This discussion has shaped the way modern scholars approach several institutions, such as the pledge, regarding which some claim that the pledge creditor could only transfer praetorian ownership.⁵¹ However, there are countless texts where ownership is transferred by a non-owner who acts under the authorisation ofthe owner (voluntate domini).⁵²Some of these texts clearly show that the authorised non-owner may transfer Quiritary ownership. An opinion of Javolenus even presents the owner's consent for the delivery as the criterion that stems from the strict interpretation of the ius civile, which can be corrected through the intervention of the praetor.⁵³ However, scholars have traditionally neglected these references to the owner's consent at the traditio, claiming that such texts reflect post-classical doctrines.⁵⁴ Such criticism is coupled with the idea that the Roman traditio was strictly causal in nature, which is why every reference to a particular intent at the delivery should be regarded as post-classical.⁵⁵ This line of reasoning - which again makes use of strict modern legal notions to approach the sources - has refrained scholars from attempting to understand the way Roman jurists approached the voluntas domini in the transfer of ownership by a non-owner, as well as its evolution within classical Roman law.⁵⁶

The traditional theory also faces serious problems when faced with the evidence concerning other legal institutions. There are, in fact, various archaic legal institutions where the legal position of an individual is affected by the acts of someone else. Whether one wants to label them as forms of "direct representation" is as arbitrary as it has always been to approach the sources through this notion, but the point remains that some agency-like figures can be found in early legal institutions. For example, we have noticed that a sacred and ancient ritual such as the consecratio of holy objects could be carried out by someone other than the owner of the consecrated object, as long as the latter authorised it.⁵⁷ There are also numerous texts where a non-owner - particularly a son-in-power - performs a manumissio vindicta,⁵⁸a form of manumission which was modelled after the in iure cessio. Scholars advocating for an ancient prohibition on direct representation - and specially Mitteis - have resorted to the unappealing argument of post-classical corruption in all of these texts,⁵⁹ which Buckland rebuffed.⁶⁰ There is even evidence of cases where a sponsio or stipulatio is concluded by a slave.⁶¹ One can also identify in Plautus a case where someone attempts to persuade his friend to marry off his sister through a sponsio.⁶²To return to the transfer of ownership: in the context of the ancient consortium ercto non cito one of the co-owners could transfer ownership over the object as a whole, thus affecting his co-owners even without their authorisation.⁶³ Leaving aside the peculiar features of these institutions, it seems at any rate clear that early Roman jurisprudence was not abiding to a prohibition on direct representation.

The above analysis of the sources shows that Roman law acknowledged different forms of agency. It is worth bearing in mind that the doctrine of direct representation originally developed in modern law specifically within the law of obligations, particularly regarding the conclusion of contracts through an agent. This tells us in itself a lot about the sources. Until the nineteenth century, jurists had no problem to explain how an agent could carry out countless legal acts on behalf of the principal, such as the transfer of ownership, pledge, payment, etcetera. The sources provided abundant criteria to determine the outcome of the agent's acts in all of these cases. This is why, for instance, a jurist like Pothier bestowed considerable attention to overcome the alteri stipulari rule within his Traité des Obligations,⁶⁴but did not feel the need to apply such notions to explain, for example, how ownership was transferred through an agent.⁶⁵ Only in the course of the nineteenth century did German jurists feel the need to approach every form of legal intermediation under a common doctrine of Stellvertretung, even if the sources did not present any obstacles for an agent to act on behalf of the principal in countless situations.⁶⁶

At this point it becomes clear that Roman law knew no such thing as a general prohibition on direct representation. One may, moreover, add that Roman jurists developed general distinctions of their own to determine under what conditions an individual could affect the legal position of someone else, considering whether an act is beneficial or detrimental to the dominus negotii..⁶⁷ Particularly interesting in this regard is an opinion of Gaius in D 3 5 38(39),⁶⁸ which generally distinguishes between acts which imply a patrimonial loss for the principal and those which imply a patrimonial gain. In the first case, the principal will be bound as long as he authorises the agent to perform such acts, while in the second case it is only necessary that the agent acts on his behalf.

The general formula offered by Gaius agrees with the legal reasoning behind numerous texts,⁶⁹ since the sources show that an agent could in fact carry out voluntate domini several acts that involved a patrimonial loss. This is, for instance, the case regarding the transfer of ownership, the constitution of various iura in re aliena,⁷⁰pledge,⁷¹ consecration of holy objects, manumission,⁷² etcetera. On the other hand, several acts involving a patrimonial gain to the dominus negotii required that the agent acted nomine domini, such as the acquisition of ownership - although the sources do not offer a uniform view on this regard - or payment.⁷³ Moreover, Ulpian's explanation to the alteri stipulari-rule abides by the general distinction between acts that imply a patrimonial gain and those that cause a patrimonial loss,⁷⁴as does Gaius when claiming elsewhere that "melior condicio nostra per servos fieri potest, deterior fieri non potest".⁷⁵

Gaius' formula in D 3 5 38(39) and the texts related to it also show that Roman jurists were mainly concerned with the problem of whether the position of an individual was affected by the acts of someone else, and not whether an agent was validly fulfilling the requirements to bind the principal. For example, Roman jurists did not ask themselves whether an individual who buried a dead man in someone else's property was acting as his agent, but whether the owner was affected by granting his consent. Modern scholars would rather approach these problems through different dogmatic categories, distinguishing for instance between direct representation and the abdicative acts of the owner. Such distinctions are however completely absent from the Roman sources.

Despite these interesting constructions offered by Roman sources, one should avoid relying on them too much. Texts such as D 3 5 38(39) are little more than a guideline or rule of thumb which were obtained through the observation of specific solutions, and they should not be adopted as over-arching principles which can explain every form of legal intermediation in Roman law.⁷⁶ Accordingly, scholars should refrain from replacing the general theory of "direct representation" in Roman law by a new theory of "patrimonial gain or loss" in order to approach the sources. Instead, the particular set of texts and solutions governing specific legal institutions - the conclusion of contracts, transfer of ownership, payment, etcetera - should be approached individually, as it will be shown in the following section.

 

5 A piece-meal approach, anachronistic preconceptions and the nuntius

These different features of the various agency-like situations in Roman jurisprudence show that one cannot take as a starting point that every form of agency in early Roman law evolved from a uniform prohibition on direct representation. The common thread that scholars identified since the nineteenth century concerning direct representation in early Roman law is nothing but an illusion. This is why one should favour a piece-meal approach of each of the different cases where an individual carries out a legal act through someone else. This approach is already to be seen to a large extent in different contributions by Buckland, where he pointed out the often careless constructions made around the notion of direct representation in general, as well as the evidence in the sources regarding specific cases.⁷⁷ Such a piece-meal approach when approaching the problem of agency in the sources was moreover explicitly described as an agenda for the study of agency in Roman law by Kreller, Cappellini and - more recently - Coppola Bisazza and Miceli, who have urged to discard a joint analysis when dealing with institutions that were governed by various sets of rules.⁷⁸

This piece-meal approach to agency in Roman law must also be complemented by an effort to uproot some seriously anachronistic conceptions which the traditional theory has brought into the analysis of Roman law. It could be argued that Roman law scholarship tolerates to a considerable degree the use of modern legal terminology to describe the sources, particularly when it provides the legal historian with a mere working hypothesis.⁷⁹ However, such use should be restrained when it brings along relevant distortions, explaining legal outcomes through dogmatic distinctions that are alien to the sources.⁸⁰ This is precisely what happens regarding the significance of the contemplatio domini in the Roman law of agency. Since scholars have traditionally claimed that Roman law forbade "direct representation" - which implies acting nomine alieno - many have assumed in turn that what was acceptable to Roman law was "indirect representation", namely to carry out a legal act for someone else while acting on behalf of oneself (nomineproprio). This, in turn, has led countless scholars to claim that any agent who acts on behalf of the dominus negotii should be labelled as a nuntius⁸¹

The significance granted to the contemplatio domini in Roman law on account of the traditional theories on direct representation is utterly alien to the sources. The scope given to the nuntius in particular is part of a typically pandectistic construction,⁸² and in fact the legal significance of the nuntius in Roman law is a problem which remains largely unaddressed by modern Roman law scholarship.⁸³ Accordingly, whether an agent acts on behalf of his principal or not may of course be relevant in some situations, but the sources do not offer a general distinction in this regard, and certainly do not set the nuntius as a uniform figure for all the cases where the agent acts nomine alieno. For instance, it could be relevant for the acquisition of ownership through an agent whether the latter acts on behalf of the owner or not.⁸⁴ Similarly, as mentioned above, the payment of someone else's debt needs to be done on behalf of the debtor, or otherwise the debtor will remain bound. This outcome is not only related to the fact that the debtor is favoured by such act -as pointed out in D 3 5 38(39) - but also to the practical fact that the creditor cannot know which debt is extinguished unless pointed out by the person who pays. Such considerations are not relevant for the transfer of ownership by a non-owner, where the contemplatio domini is by no means decisive to determine whether ownership is transferred or not. In fact, some of the sources dealing with the delivery voluntate domini reveal in passing that the traditio was carried out on behalf of the owner, but that this circumstance is only relevant to determine the good faith of the acquirer for the usucapio.⁸⁵The decisive element for the transfer of ownership remains thus the voluntas domini. Burdese was critical of the attempts to analyse the transfer of ownership through the notion of direct representation, particularly regarding the significance of the contemplatio domini⁸⁶ Accordingly, the evidence shows that one cannot claim that the contemplatio domini is equally relevant in all of these cases, or that every legal act performed nomine alieno should be approached under the common idea of a nuntius. Neither should one attempt, for instance, to determine by way of analogy the significance of the contemplatio domini in the transfer of ownership by resorting to the rules for the acquisition of ownership.⁸⁷ These are all problems that have a logic of their own in Roman law, where over-arching dogmatic constructions on direct representation are completely absent.

Similar considerations can be made concerning contracts concluded by commercial agents. The sources often refer explicitly to contracts concluded on behalf of the dominus negotii,⁸⁸ something that must have been as common in the ancient world as it is today. Just as in the case of the transfer of ownership, whether the agent acts on behalf of the principal or not is largely irrelevant for the ensuing legal consequences.⁸⁹ The agent will bind himself to the other contracting party, whether he acts on behalf of the principal or not.⁹⁰ However, the sources often show that the third party is fully aware of contracting with an alieni iuris, particularly if he is dealing with an institor or an excercitor. In fact, it is precisely this knowledge that would normally enable the third party to resort to an actio adiecticiae qualitatis. This knowledge is, moreover, important in order to know the scope of the appointment of the agent (praepositio) and the limitations he faced (proscriptio), information which sometimes was even displayed through a lex praepositionis .⁹¹ Scholars should therefore avoid picturing Roman tradesmen as artfully hiding who their principal was in order to avoid violating a prohibition of direct representation. Accordingly, the intermediary who acts on behalf of the principal cannot be labelled as a nuntius only on account of acting nomine alieno.

 

6 Concluding remarks

Two main conclusions can be obtained from the above analysis. Firstly, the notion of "direct representation" should be left aside when approaching the Roman sources. The doctrine of direct representation is a relatively modern construction, the features of which are alien to Roman law. For example, while a modern jurist may approach different situations - such as contracts concluded by an agent, delivery by an agent, payment by an agent - as part of a common phenomenon of direct representation which has a general set of rules, such uniform approach was completely unknown to Roman jurists. Moreover, the contemplatio domini has a decisive role in the modern doctrine of direct representation, while it was only occasionally meaningful to determine whether the acts of an agent could affect the dominus negotii. Accordingly, "direct representation" is an anachronistic notion that is inadequate for the legal historian approaching Roman sources. Instead, a more source-oriented understanding of agency in Roman law favours a piece-meal approach, which identifies the peculiar reasoning behind the different situations where the dominus negotii is directly affected through the acts of his agent.

A second conclusion is that there is no evidence to claim the existence of a primitive ban on direct representation in Roman law. Since there was no common legal doctrine to approach the different forms of agency in Roman law, one cannot claim that all of these cases were equally subject to a general prohibition. Nor is there evidence that early Roman law consistently prevented an agent from validly affecting the legal position of the principal. Moreover, in those cases where agency faced some sort of limitation, there is no evidence that such limitations sprung from a primitive general principle. Accordingly, legal historians should refrain from referring to a primitive prohibition on direct representation when describing the evolution of different forms of agency and intermediation in Roman law.

 

BIBLIOGRAPHY

Ankum, JA (1970) "Une nouvelle hypothese sur l'origine de la règle alteri dari stipulari nemo potest" in Études offertes ä Jean Macqueron (Aix-en-Provence): 21-29        [ Links ]

Ankum, JA (1978) "Mancipatio by slaves in classical Roman law?" in W de Vos, W Dean, I Leeman (eds) Essays in Honour of Ben Beinart vol 1 (Cape Town, Wetton, Johannesburg): 1-18        [ Links ]

Ankum, JA (1981) "Alla ricerca della repromissio e della satisdatio secundum mancipium" Atti dell'Accademia Romanistica Costantiniana 4: 741-792.         [ Links ]

Ankum, JA (1984) "Mancipatio by slaves in classical Roman law" in J van der Westhuizen (ed) Huldigingsbundel Paul van Warmelo (Pretoria): 5-17        [ Links ]

Ankum, JA (2013) "Noch einmal: die repromissio und die satisdatio secundum mancipium im klassischen römischen Recht" in M Armgardt, F Klinck & I Reichard (eds) Liber Amicorum Christoph Krampe zum 70. Geburtstag (Berlin): 13-28        [ Links ]

Betti, E (1935) Diritto romano Parte generale vol 1 (Padua)        [ Links ]

Briguglio, F (2007) Studi sul procurator, L 'acquisto del possesso e della proprietä (Milan)        [ Links ]

Buckland, WW (1903) "Manumissio vindicta par un fils de famille" Revue Historique de Droit Franqais et Étranger 27: 737-744        [ Links ]

Buckland, WW (1908) The Roman Law of Slavery: The Condition of the Slave in Private Law from Augustus to Justinian (Cambridge)        [ Links ]

Buckland, WW (1918) "Mancipatio by a slave" Law Quarterly Review 34: 372-379        [ Links ]

Buckland, WW (1931) The Main Institutions of Roman Private Law (Cambridge)        [ Links ]

Burdese, A (2009) "Potestas alienandi di non domini e principio di rappresentanza" in A Trisciuoglio (ed) Valori e Principii del diritto romano. Giornata di Studi per i cento anni del prof Silvio Romano (Turin): 15-29        [ Links ]

Cappellini, P (1987) "Rappresentanza (diritto intermedio)" in Enciclopedia del diritto vol 38 (Milan): 435-463        [ Links ]

Claus, A (1973) Gewillkürte Stellvertretung im römischen Privatrecht (Berlin)        [ Links ]

Coppola Bisazza, G (2008) Dallo iussum domini alla contemplatio domini, Contributo allo studio della storia della Rappresentanza (Milan)        [ Links ]

Corbino, A (1976) "La legittimazione a 'mancipare' per incarico del proprietario" IVRA, Rivista internazionale di diritto romano e antico 27: 50-71        [ Links ]

Corbino, A (1982) "La 'pro Caecina' di Cicerone e l'acquisto della proprietà mediante intermediario estraneo" in Studi in onore di Arnaldo Biscardi vol 3 (Milan): 277-287        [ Links ]

Corbino, A (1984) "Forma librale ed intermediazione negoziale" in Sodalitas. Scritti in onore di Antonio Guarino vol 5 (Naples): 2257-2272        [ Links ]

D'Ors, A (2004) Derechoprivado romano (Pamplona)        [ Links ]

Düll, R (1950) "Über Ansätze direkter Stellvertretung im frührepulikanischen römischen Recht" Zeitschrift der Savigny-Stiftung für Rechtsgeschichte: romanistische Abteilung 67: 162-184        [ Links ]

Erxleben, F (2017) Translatio iudicii, Der Parteiwechsel im römischen Formularprozess, (Muenich)        [ Links ]

Exner, A(1867) Die Lehre vom Rechtserwerb durch Tradition nach österreichischen und gemeinen Recht (Vienna)        [ Links ]

Finkenauer, T (2008) Direkte Stellvertretung bei Stipulationen?, Zeitschrift der Savigny-Stiftung für Rechtsgeschichte: romanistische Abteilung 125: 440- 497        [ Links ]

Finkenauer, T (2010) Vererblichkeit und Drittwirkungen der Stipulation im klassischen römischen Recht (Tübingen)        [ Links ]

Guzmán Brito, A (2013) Derecho Privado Romano vols 1 & 2 (Santiago)        [ Links ]

Hallebeek, J & Dondorp, H (eds) (2008) Contracts for a Third-Party Beneficiary, A Historical and Comparative Account (Leiden, Boston)        [ Links ]

Hamza, G (1980) "Aspetti della rappresentanza negoziale in diritto romano" Index, quaderni camerti di studi romanistici 9: 193-229        [ Links ]

Hamza, G (2010) "Aspetti teorici della rappresentanza negoziale" Revista General de Derecho Romano 14: 1-21        [ Links ]

Hoetink, HR (1955) "Les notions anachroniques dans l'historiographie du droit" Tijdschrift voor Rechtsgeschiedenis 23: 1-20        [ Links ]

Hölzl, FJ (2002) Friedrich Carl von Savignys Lehre von der Stellvertretung, Ein Blick in seine juristische Werkstatt (Göttingen)        [ Links ]

Jörs, P, Kunkel, W & Wenger, L (1949) Römisches Recht 3^rd ed (Berlin, Göttingen, Heidelberg)        [ Links ]

Jörs, P, Kunkel, W & Wenger, L (1987) Römisches Recht 4^th ed by H Honsell, Th Mayer-Maly & W Selb (Berlin, Heidelberg, New York)        [ Links ]

Karakasis, E (2003) "Legal language in Plautus with special reference to Trinummus" Mnemosyne 52(2): 194-209        [ Links ]

Kaser, M (1950) "Ueber Verfügungsakte Gewaltunterworfener mit Studien zur Natur der 'manumissio vindicta'" Studia et Documenta Historiae et Iuris 16: 59-85        [ Links ]

Kaser, M (1971-1975) Das Römische Privatrecht vols 1 & 2 (Munich)        [ Links ]

Kaser, M & Hackl, K (1996) Das Römische Zivilprozessrecht (Münich)        [ Links ]

Klinck, F (2004) Erwerb durch Uebergabe an Dritte nach klassischem römischen Recht (Berlin)        [ Links ]

Kreller, H (1948) "Das Rechtsinstitut der Stellvertretung" Juristische Blätter 70: 221-226        [ Links ]

Lenel, O (1889) Palingenesia iuris civilis vols 1 & 2 (Leipzig)        [ Links ]

Ligios, MA (2013) Nomen negotiationis (Torino)        [ Links ]

Longo, G (1982) "Nuncius (diritto romano)" in Novissimo Digesto Italiano 11 (Turin): 514-515        [ Links ]

Lovato A (2001) "Traditio e conventio nel settimo libro delle disputationes ulpianee" Studia et Documenta Historiae et Iuris 67: 79-133        [ Links ]

Miceli, M (2008) Studi sulla 'rappresentanza 'nel diritto romano (Milan)        [ Links ]

Mitteis, L (1885) Die Lehre von der Stellvertretung nach römischen Recht mit Berücksichtigung des österreichischen Rechtes (Vienna)        [ Links ]

Mitteis, L (1901) "Ueber die Manumissio vindicta durch den Haussohn" Zeitschrift der Savigny- Stiftung für Rechtsgeschichte: romanistische Abteilung 21 (1900): 199-211        [ Links ]

Mitteis, L (1904) "Die manumissio vindicta durch den Haussohn" Zeitschrift der Savigny-Stiftung für Rechtsgeschichte: romanistische Abteilung 25: 379-381        [ Links ]

Mitteis, L (1908) Römisches Privatrecht bis auf die Zeit Diokletians (Leipzig)        [ Links ]

Mühlenbruch, CF (1836) Die Lehre von der Cession der Forderungsrechte: nach den Grundsätzen des römischen Rechts dargestellt (Greifswald)        [ Links ]

Pernice, A (1873) Labeo, römisches Privatrecht im ersten Jahrhundert der Kaiserzeit vol A (Halle)        [ Links ]

Pringsheim, F (1933) "Animus in Roman law" Law Quarterly Review 49: 43- 60, 379-412        [ Links ]

Rabel, E (1955) Grundzüge des römischen Privatrechts (Basel)        [ Links ]

Riccobono, S (1930) "Lineamenti della dottrina della rappresentanza diretta in Diritto Romano" Annali del Seminario Giuridico dell'Università di Palermo: 389-447        [ Links ]

Rodriguez Diez, JE (2016) Potestas alienandi, Transfer of Ownership by a Non-Owner from Roman Law to the DCFR (Oisterwijk)        [ Links ]

Savigny, FC (1814) Vom Beruf unsrer Zeit für Gesetzgebung und Rechtswissenschaft (Heidelberg)        [ Links ]

Savigny, FC (1840) System des heutigen römischen Rechts vol 3 (Berlin)        [ Links ]

Savigny, FC (1853) Das Obligationenrecht vol 2 (Berlin)        [ Links ]

Schulz, F (1917) "Die Lehre vom concursus causarum im klassischen und justinianischen Recht" Zeitschrift der Savigny-Stiftung für Rechtsgeschichte: romanistische Abteilung 38: 114-208        [ Links ]

Schulz, F (1951) Classical Roman Law (Oxford)        [ Links ]

Siber, H (1928) Römisches Recht in Grundzügen für die Vorlesung, vol 2 Römisches Privatrecht (Berlin)        [ Links ]

Sohm, R (1908) Institutionen, ein Lehrbuch der Geschichte und des Systems des Römischen Privatrechts (Leipzig)        [ Links ]

Stagl, JF (2009) Favor dotis: die Privilegierung der Mitgift im System des römischen Rechts (Vienna, Cologne, Weimar)        [ Links ]

Talamanca, M (1990) Istituzioni di diritto romano (Milan)        [ Links ]

Urbanik, J (1998) "Sponsio servi" The Journal of Juristic Papyrology 28: 185-201        [ Links ]

Voci, P (1952) Modi di acquisto della proprietà (Milan)        [ Links ]

Weimar, P (1993) "Zum Eigentumsübergang beim Pfandverkauf im klassischen römischen Recht" in JA Ankum et al (eds) Mélanges Felix Wubbe (Fribourg): 551-569        [ Links ]

Zimmermann, R (1996) The Law of Obligations, Roman Foundations of the Civilian Tradition (Oxford)        [ Links ]

 

 

1 For an outlook of the scholarship behind this claim, see Cappellini 1987: 440-441; Coppola Bisazza 2008: 3-10.
2 Coppola Bisazza 2008: 10; Hamza 2010: 2-4.
3 See, eg, Jörs, Kunkel & Wenger 1987: 113; Zimmermann 1996: 50-51; D'Ors 2004: 561-562.
4 Rodriguez Diez 2016.
5 Mühlenbruch 1836: 41-47.
6 Rodriguez Diez 2016: 223-225.
7 D 50 17 123pr (Ulpian 14 ad edictum), translation by Watson & Crawford: "[n]o one can legally act on behalf of another."
8 Raevardus De auctoritate prudentum liber singularis (1566): 39-51.
9 Conradi Depacto fiduciae exercitatio (1732): 12-14; 23-25.
10 Averanius Interpretationum juris libri quinque (1753) 1 12 54 at vol 1: 166.
11 For an overview of nineteenth-century scholarship on this point see Rodriguez Diez 2016: 24-27.
12 Mühlenbruch 1836: 45 n 75.
13 Idem 41-42.
14 D 50 17 73 4 (Scaevola libro singulari δρων): "Nec paciscendo nec legem dicendo nec stipulando quisquam alteri cavere potest" ("Nor can anyone stand surety for another by making a pact or laying a condition or stipulating" [translation by Watson & Crawford]). Other references to this rule can be found in Just Inst 3 19 4 and 3 19 19, as well as D 45 1 38 17.
15 Gai Inst 2 95: "Ex his apparet per liberos homines, quos neque iuri nostro subiectos habemus neque bona fide possidemus, item per alienos servos, in quibus neque usumfructum habemus neque iustam possessionem, nulla ex causa nobis adquiri posse. Et hoc est, quod vulgo dicitur per extraneam personam nobis adquiri non posse ..." ("It is apparent from this that we never acquire through a free person except where we have power over him or possess him in good faith, nor through a third party's slave unless we have a usufruct in him or possess him in good faith. That justifies the common maxim 'no acquisition to us through an outsider'..." [translation by Gordon & Robinson]). Other references to this rule can be found in Just Inst 2 9 5; D 45 1 126 2; Codex Iustinianus 4 27 1 pr.
16 Mühlenbruch 1836: 43-44: "Den in diesen und andern Stellen enthaltenen Bestimmungen lag ursprünglich ein Hauptgedanke zum Grunde, der nämlich: dass Jeder für sich in Person handeln müsse".
17 Mühlenbruch 1836: 44-47.
18 Savigny 1814: 102-103.
19 Savigny 1840: vol 3 90-98; Savigny 1853: vol 2 21-88.
20 Savigny 1840: vol 3 98; Savigny 1853: vol 2 68-73.
21 Savigny 1840: vol 3 91-92; Savigny 1853: vol 2 41.
22 Exner 1867: 125-127; Pernice 1873: vol A 488; Sohm 1908: 256-262.
23 Mitteis 1908: 203-213. For his earlier views on the subject, see Mitteis 1885: 9-77.
24 See, eg, Siber 1928: vol 2 411-416; Riccobono 1930: 389-337; Betti 1935: 400-405; Voci 1952: 81-91; Rabel 1955: 182-192; Bonfante 1968: vol II/2 210-214; Kaser 1971: 260-267; Claus 1973: 9-13; Hamza 1980: 193; Jörs, Kunkel & Wenger 1987: 111-114; Zimmermann 1996: 34-35; 4754; D'Ors 2004: 561-571; Finkenauer 2008: 442-448.
25 See, eg, Ankum 1978: 13.
26 Gai Inst 3 167.
27 Lenel 1889: vol 2 col 494.
28 Gai Inst 4 31.
29 Kaser & Hackl 1996: 62-63; Finkenauer 2010: 221; Erxleben 2017: 3 n 12.
30 Hölzl 2002: 229-278.
31 See, eg, Corbino 1982: 278 283; Corbino 1984: 2267-2271; Klinck 2004: 356-358; Briguglio 2007: 538-551; Coppola Bisazza 2008: 209-222; and Miceli 2008: 265-277.
32 Ankum 1970: 21-29. Among the supporters of this view, see Kaser 1975: vol 2 339 n 30; Zimmermann 1996: 35-40; Hallebeek & Dondorp 2008: 8-14.
33 See on this case Stagl 2009: 140-157.
34 See, eg, D 45 1 83pr; D 44 7 11; Just Inst 3 19 3 and 3 19 21.
35 Codex Iustinianus 4 27 1pr-1; D 45 1 126 2.
36 Gai Inst 2 95: "Et hoc est, quod vulgo dicitur per extraneam personam nobis adquiri non posse ..."
37 D 50 17 54 (Ulpian 46 ad edictum): "Nemo plus iuris ad alium transferre potest, quam ipse haberet" (No one can transfer to another a better right than he has himself). Similarly D 41 1 20pr.
38 Jörs, Kunkel & Wenger 1949: 125 n 4; Schulz 1951: 351-352; Voci 1952: 4-5.
39 Eg, Plato Symposium 196d-e; Aristotle De Sophisticis Elenchis 23 179" 22-24; Cicero Pro Flacco 56 4; Seneca De beneficiis 5 12 7; Quintilian Institutio Oratoria 5 10 74.
40 D 41 2 21pr; D 9 4 27 1; D 20 1 3 1; D 50 17 120; D 7 1 63; D 41 1 46. On this development, see Rodriguez Diez 2016: 265-316.
41 Mitteis 1908: 207-213; Siber 1928: vol 2 412-413; Riccobono 1930: 437-443; Rabel 1955: 186; Claus 1973: 307; Zimmermann 1996: 51; Guzmán Brito 2013: vol 1 504.
42 Mitteis 1908: 208-211. Among the supporters of a general inability of a non-owner to perform a mancipatio or in iure cessio, see, eg, Siber 1928: vol 2 412-413; Kaser 1950: 66; Kaser 1971: vol 1 267; Claus 1973: 307; Talamanca 1990: 271; Weimar 1993: 552; D'Ors 2004: 231.
43 In favour of the views of Mitteis is Ankum 1978: 1-18; Ankum 1984: 5-17. Contrary to Mitteis are Buckland 1918: 372-379; Corbino 1976: 50-71; Corbino 1984: 2257-2272; Coppola Bisazza 2008: 287-289. The controversy is analysed in Rodriguez Diez 2016: 239-252.
44 Mitteis 1908: 208 n 16 in fact acknowledges the lack of evidence on the subject: "Der Quellenbeweis für die Unmöglichkeit stellvertretender Manzipationsveräusserung ist freilich nicht ganz leicht zu führen, weil die klassischen Erörterungen über diese Frage von den Kompilatoren gestrichen sind."
45 D 21 2 39 1, analysed in Rodriguez Diez 2016: 243-252. Mitteis 1908: 209, in particular, claims that the text is highly corrupted.
46 Ankum 1981: 790: "Secondo l'opinione dominante, che io considero esatta, un procurator non aveva la capacitá di mancipare la res mancipi, che vende per il suo principale. Non puo dunque far naschere il dovere di auctoritas per mezzo di una mancipatio della cosa venduta ad un prezzo reale." See, moreover, Ankum 2013: 14-28, where the author revises some ideas on the subject, which do not include the possibility of a mancipatio by a non-owner.
47 Mitteis 1908: 210; Siber 1928: vol 2 413; Rabel 1955: 184-185; Weimar 1993: 551; D'Ors 2004: 379 n 2.
48 Guzmán 2013: vol 1 455.
49 For a revision of these views, see Rodriguez Diez 2016: 177-186.
50 For example, Kaser 1971: vol 1 267 n 59: "Die Zustimmung wirkt offenbar nur honorarrechtlich."
51 Weimar 1993: 551-569.
52 For example, D 41 1 9 4; D 6 1 41 1; D 50 17 165; D 24 1 38 1; D 12 4 3 8; D 17 1 5 3; D 24 1 5pr; Gai 2 64. Concerning the voluntas domini at the traditio, see Rodriguez Diez 2016: 63-164.
53 D 39 5 25. See, moreover, D 6 2 14; D 41 4 7 6; Codex Iustinianus 7 26 4. For an analysis of these texts, see Rodriguez Diez 2016: 66-69.
54 Pringsheim 1933 & 43-60 379-412.
55 Riccobono 1930: 437-443; Pringsheim 1933: 55-60; and Jörs, Kunkel & Wenger 1949: 129.
56 Among the few scholars that have proposed to study the voluntas or consensus at the delivery as a classical phenomenon, see Schulz 1917: 141-145 and Lovato 2001: 133.
57 Gai Inst 2 7; D 1 8 6 4.
58 For example, D 37 14 13; D 38 2 22; D 40 1 7; D 40 1 16; D 40 1 22; D 40 2 4pr; D 40 2 10;
D 40 2 18 2; D 40 2 22; D 40 9 15 1; D 49 17 6; Pauli Sententiae 1 13a; Codex Iustinianus
7 15 1 3.
59 Mitteis 1900: 199-211; Mitteis 1904: 379-381; Mitteis 1908: 211 n 23.
60 Buckland 1903: 737-744; Buckland 1908: 718-723. On this controversy, see Rodriguez Diez 2016: 257-259.
61 Urbanik 1998: 185-201.
62 Plautus Trinnumus act 2 scene 4 v 502. Karakasis 2003: 205 considers, however, that this had purely comical implications. I thank Professor Carlos Amunátegui for this reference.
63 Gai Inst 3 154b.
64 Pother Traité des obligations (1861): 42-45.
65 Pother Traité du droit du domaine (1772): 212-219 resorts to an independent set of rules to approach this problem.
66 Cappellini 1987: 456; Gai Inst 3 154b.
67 Kreller 1948: 222; Rodriguez Diez 2016: 37-43.
68 D 3 5 38(39) (Gaius 3 de verborum obligationibus): "Solvendo quisque pro alio licet invito et ignorante liberat eum: quod autem alicui debetur, alius sine voluntate eius non potest iure exigere. Naturalis enim simul et civilis ratio suasit alienam condicionem meliorem quidem etiam ignorantis et inviti nos facere posse, deteriorem non posse" ("Anyone paying on behalf of someone else, even without his knowledge and agreement, frees him from liability, but another person cannot lawfully demand payment of what is owing to anyone without his consent. For the principles of both natural justice and the civil law are in favour of our being able to improve another's position, even without his knowledge and agreement, but not of our being able to make it worse" [translation by Watson & Kinsey]).
69 For example, D 39 3 8.
70 Regarding the right to carry water across land see, for example, D 39 3 8; D 39 3 10; Codex Iustinianus 3 34 4.
71 For example, D 13 7 20; D 20 1 21 pr; Codex Theodosianus 2 30 2.
72 D 23 2 51 1.
73 D 46 3 53; Just Inst 3 29pr; D 46 3 17. On the latter text see Rodriguez Diez 2016: 133-134.
74 D 45 1 38 17 (Ulpian 49 ad Sabinum): "Alteri stipulari nemo potest, praeterquam si servus domino, filius patri stipuletur: inventae sunt enim huiusmodi obligationes ad hoc, ut unusquisque sibi adquirat quod sua interest: ceterum ut alii detur, nihil interest mea ..." ("No one can stipulate on behalf of another, except where a slave stipulates for his master, a son for his father; for obligations of this kind were devised in order that each man should acquire for himself what is of benefit to him; but it is of no benefit to me that something should be given to another ..." [translation by Watson, Hart, Lewis & Beinart]). Similarly D 45 1 126 2 (Paul 3 quaestionum).
75 D 50 17 133 (Gaius 8 ad edictum provinciale) (translation by Watson & Crawford: "Our condition can be improved but not worsened by our slaves").
76 I thank Prof Mr Egbert Koops for his useful input in this point.
77 Buckland 1918: 372-373; Buckland 1931: 161.
78 See Kreller 1948: 221; Cappellini 1987: 440-442; Coppola Bisazza 2008: 3-14; Miceli 2008: 12-18.
79 Hoetink 1955: 6-7 & 16; Rodriguez Diez 2016: 21-22.
80 Hoetink 1955: 15-16.
81 See, eg, Zimmermann 1996: 50-51.
82 Concerning Savigny's ideas regarding the nuntius see Hölzl 2002: 205-216; 279-280; 288-289.
83 Among the few contributions on this subject see Düll 1950: 162-170; Longo 1982: 514-515.
84 Consider, for instance, the relevance of the nominatio when a slave has more than one master, as described in Gai Inst 3 167-167a.
85 See D 41 4 14, analysed in Rodriguez Diez 2016: 131-132.
86 Burdese 2009: 15-29. See, moreover, Rodriguez Diez 2016: 103-111 128-136.
87 Rodriguez Diez 2016: 43-47.
88 For example, D 45 1 126 2 (Paul 3 quaestionum): "Plane si liber homo nostro nomine pecuniam daret vel suam vel nostram, ut nobis solveretur, obligatio nobis pecuniae creditae adquireretur ."; D 3 3 67 (Papinian 2 responsorum): "nam procurator, qui pro domino vinculum obligationis suscepit, onus eius frustra recusat"; D 12 1 9 8 (Ulpian 26 ad edictum): "Si nummos meos tuo nomine dedero ."
89 Talamanca 1990: 266 268.
90 D 3 3 67.
91 D 14 3 5 11-16; D 14 3 16; D 14 1 1 12; D 14 3 11 2-6. Concerning these institutions, see Ligios 2013: 23-78.