Since the publication, in 1927, of Hersch Lauterpacht’s doctoral dissertation, entitled Private law sources and analogies of international law, the idea that private law sources and analogies contributed much to the development of early modern and modern public international law has been well established in historiography1. Several studies have delved into the reception and adaptation of Roman private law, showing the extent to which its concepts, principles and rules have influenced modern international law (and treaty law in particular), and even how often it provided arguments and a source of inspiration for international legal practice during the second half of the twentieth century and the early twenty-first century2. In recent years, private law institutions such as occupation and acquisitive prescription have also been studied in relation to the notion of empire in Western legal and political thought and to the ideological context of European imperialism3. The function of private law sources and analogies in late medieval scholarship on public international law has, however, received much less attention4. The purpose of this paper is to shed light on this function by analysing the impact of late medieval scholarship on private law agency on the conceptualisation of diplomatic representation, in the broader context of a discussion of the significance of late medieval legal scholarship for the elaboration of the status of ambassadors5. After providing some introductory remarks on the origins of the debate on the office of ambassador, this paper examines the definition of the basis and limits of the negotiating power granted to diplomatic envoys in the late medieval ius commune. The conclusion will illustrate with some examples the persistence of this notion of diplomatic representation in the juristic discussion of the matter right up until the late eighteenth century.
In Northern and Central Italy, the practice of diplomacy developed significantly over the twelfth century, in parallel with the growth of the communes. The Lombard League – formed in 1167 as a military and defensive alliance against Frederick I and his attempts to curtail the freedoms of the commune – played a major role in the strengthening of a close web of relations between its member cities, and greatly influenced their institutional evolution through the creation of supra-city institutions and the establishment of shared rules of mediation and political communication6. The Roman Church – which began to dispatch emissaries and legates as early as the fourth century – had also relied heavily on papal legates since the late eleventh century, in its endeavours to centralize political and administrative power7. A great difference, however, existed between papal legates and secular ambassadors : the former were agents empowered by the pope with public authority – the power to « govern » a province, in the words of Bernardus Papiensis (d. 1213) and later of Henricus de Segusio (Hostiensis, c. 1200-1271), « iurisdictio, sive potestas » in those of Guillaume Durand (1236-1296), and even « imperium » in those of Baldus de Ubaldis (1327-1400)8. In other words, papal legates were primarily sent out not to negotiate an agreement, but to exercise jurisdiction in territories which were subject to the pope’s (spiritual, at least) sovereignty9. The ambassadors exchanged between Italian cities, on the other hand, had no such authority : since they operated beyond the territorial limits of their principal’s jurisdiction, they had no power to oblige the recipient of their mission to do anything and were solely authorised to deliver a message, or to negotiate as best they could to achieve their principal’s goals. Even agents exchanged between the emperor and cities could be considered simple ambassadors by legal scholars, as is shown by two comments in which Odofredus Denariis (d. 1265) mentions Frederick I and the Lombards’ « ambasiatores » who engaged in the negotiations which led to the Peace of Constance (1183)10.
The intense diplomatic activity that characterised Northern and Central Italy caused jurists to start pondering the office of ambassador from the late twelfth century onwards, even before local statutes began to establish a regulatory framework for diplomatic activities (which, in any case, throughout the late Middle Ages was always far from being complete)11. One of the main questions upon which the jurists focused was the public role assigned to the ambassador (legatus), referred to by the terms officium and munus publicum. The latter in particular, rooted in the Roman system of the munera – the public services which Roman citizens had to perform for their municipium – indicated the compulsory nature of the office, which (with certain exceptions) had to be fulfilled by the person appointed for as long as had been stipulated12. Other aspects of the office that were discussed by late medieval jurists include the range of actors entitled to send ambassadors, the ambassador’s appointment, his right to a salary or reimbursement for the losses incurred during his mission, his capacity to accept or solicit gifts and benefices from the recipient of his mission, and his inviolability, immunities and other privileges13. The definition of the (public) status of the ambassador was, moreover, linked to that of the cities themselves. As is well known, Roman law defined cities as municipia and, reserving the status of respublica only to Rome, considered them to be mere private entities14. It is noteworthy that one of the first legal works to elaborate on the status of ambassadors, the Summa Trium Librorum written by the Tuscan judge and advocate Rolandus de Lucca in 1195-1197, and then revised until the 1230s, is also among the first works in which the public nature of cities was strongly asserted. This was indeed a pioneering claim, which as late as the second half of the thirteenth century would still not be unanimously accepted in legal scholarship15.
Over time, a substantive body of jurisprudence grew out of this reflection, to which all three branches of late medieval ius commune – i.e. Roman, canon and feudal law – greatly contributed. As well as being a supra-national legal system, the ius commune was also at the core of legal education, providing a common language and a common form of legal reasoning. The sixth-century compilation of Roman law ordered by Emperor Justinian and the several compilations of canon law circulating in the late Middle Ages (namely Gratian’s Decretum of c. 1140, the Liber Extra of 1234, the Liber Sextus of 1298 and the Clementinae of 1314) were authoritative legal texts extensively commented upon by legal scholars ; although they consisted mostly of Lombard laws and customs, the Libri feudorum or Consuetudines feudorum included imperial laws too, and – since they formed part of the last volume of the Corpus iuris civilis (the Volumen) – also generated a stream of scholarly interpretation16. These texts, however, included only a handful of sections that dealt with « public international law » issues as such17: this lacuna may contribute to explaining why « public international law » did not develop in the late Middle Ages as a distinctive branch of legal scholarship, with its own literature and method. And so, although jurists wrote extensively on matters of diplomacy, treaties, war and peace, they were obliged to elaborate the concepts, principles and rules governing these matters based on the adaptation of those that regulated other (often private law) issues, or by referring to explicit or underlying general legal principles18. The following discussion aims to illuminate this process through one specific example, the adaptation to the (public) office of the ambassador of concepts, principles and rules originally elaborated for private law agency.
Before turning to this discussion, however, it must be pointed out that, although the abovementioned body of jurisprudence was being generated by legal scholars working in universities, it was far from being just bookish law, unrelated to actual practice. As their biographies amply demonstrate, legal scholars not only held positions at university but were also actively engaged in the social and political life of their time, as diplomats, public administrators, or sitting in either ecclesiastical or secular courts19. The legal science they elaborated was largely informed by this experience, which meant that their interpretations of the legal texts were effective adaptive responses to contemporary practical needs. The extent to which this legal science permeated late medieval society has recently been emphasised by Alain Wijffels, who defined it as a « science of public governance »20. A clue to the pervasive influence of legal scholarship is provided by the enormous number of legal opinions (consilia) produced by jurists and preserved in manuscripts and printed editions : each of these opinions related to the solution of an actual case, which could concern not only private law litigations, but also inter-polity disputes and diplomatic issues21. As a result, although the notion of ius commune as « learned law » correctly conveys the idea of a body of law that was mainly elaborated by legal scholars in their interpretation of authoritative legal texts, it would be a mistake to understand this expression to imply that ius commune had little influence on the shaping of legal systems, or to dismiss it as barely relevant to actual practice22.
What defined an ambassador in the late Middle Ages ? In Western European diplomatic practice, an official envoy always carried a letter of credence, that is a formal document – written in Latin, and sealed and signed by the sender – addressed to the recipient of a mission, which the ambassador had to deliver at his first audience with his host. This letter did not usually explain the details – or even, sometimes, the general object – of the mission ; its real function was to request the addressee to give full credence to the envoy or envoys’ words, uttered on their sender’s behalf. This document was the only prerequisite for the recognition of ambassadorial status23. An envoy might also carry other documents, but if his mission was simply to deliver a message, a letter of credence sufficed. Among the other documents that the ambassador might be carrying was an instruction, which was not essential – the use of verbal rather than written instructions was frequent – but became increasingly common from the thirteenth century on. Instructions were informal documents and intended for the envoy’s eyes only : they told him what to do and say, and sometimes how to do or say it. They usually granted no power to bind the principal24. Until well into the twelfth century, there were two possible ways to conclude a negotiation : either the rulers themselves could meet, or the recipient of a mission had to trust the ambassador and have confidence that the latter’s action would be endorsed by his principal on his return home. The letters of credence gave no such assurance, but envoys tasked with conducting negotiations were generally men whose dignity, political stature, and close association with their principal guaranteed the endorsement of their actions ; these men were usually referred to as credibiles personae, credibiles legati or probabiles viri25.
It was not until the twelfth century that the use of a new – formal – written document originated in Italian diplomacy, and quickly spread across Western Europe : this document was a procuration, or mandate (the two terms were synonymous in medieval legal practice and scholarship)26. Actually, the earliest known example of a clause de rato – i.e. a clause by which the sender promised to uphold whatever his agent did on his behalf – is found as early as 1063, in a letter by Alexander II which accompanied Peter Damiani in his mission to France : the pope declared that Damiani was sent « in his own place » to wield an authority second only to the pope’s, and expressly stated that whatever decree Damiani adopted would be confirmed as if the pope himself had issued it27. In secular diplomacy procurations appeared some time later. Initially, somewhat hybrid documents were produced – letters of credence including a clause de rato. A famous example, studied by Donald Queller, dates from 1201, when Geoffroi de Villehardouin and his colleagues carried letters from the Counts of Champagne, Flanders and Blois, on whose behalf they negotiated the transport of the Fourth Crusade army to the Levant with the Venetian government28. But soon the procuration became a separate document, addressed to the ambassador himself (not to the recipient of the mission, as was the letter of credence) and granting him the power to conclude negotiations on behalf of the principal. An early example is that of the procuration containing full powers given by Frederick Barbarossa to the bishop of Asti and two other persons to negotiate and conclude a peace treaty with the rebellious Lombards : « whatever you accept – wrote the Emperor –, we will accept it as well, and whatever you confirm by your promise or oath, we will hold it to be valid and put it into effect ». It was on the grounds of this procuration that the peace, later ratified by the Emperor at Constance, was concluded by Barbarossa’s envoys at Piacenza, in April 118329.
The procuration, therefore, differed from the letter of credence. Its most distinctive element, in the final instance, was the clause de rato. Canon lawyers soon started to consider this clause necessary in any procuration, as evidenced in the 1190 s by Bernardus Papiensis’ Summa decretalium30. The Ordinary apparatus to the Liber extra, composed and repeatedly revised by Bernardus Parmensis between 1234 and 1266, lists analytically the elements that a procuration should include. Two texts collected in the section de procuratoribus are of particular relevance. The first is a letter by which Gregory the Great, in November 602, refused to comply with the request of the bishop of Salona to invalidate the action carried out by two people who had acted in the bishop’s name without his authorisation : the pope observed that the envoys dispatched to him with the bishop’s request lacked a written mandate, and asked the bishop to send an « instructed person » carrying a procuration legally drawn up and authenticated with the signatures of the bishop and his witnesses31. The second text, a decretal by Gregory IX, stated that if a party to a litigation revoked his procurator’s mandate after the litis contestatio without making the other party and the judge aware of the revocation, the judgement subsequently delivered, although it mentioned this procurator, was nonetheless valid32. A gloss to the first letter states that any mandate should include three elements, i.e. the name of the principal, the causa for which the procurator is appointed and the de rato clause33. Further elements are added in a gloss to the second decretal, which mentions the names of the principal and the procurator, the causa for which the latter is appointed, the de rato clause and the judge before whom the case is brought, especially if he is a delegated judge34. Elsewhere Bernardus Parmensis expressed the need for a de rato clause in any mandate by saying that there was no difference between a mandatum and literas de rato35.
In the fourteenth century, Bartolus de Sassoferrato (1313/14-1357) distinguished a procuration from a letter of credence when he commented on the lex Lucius in the section de fideiussoribus et mandatoribus of the Digesta. This lex presented a private law case concerning a certain Lucius Titius, who had bound himself by letter as guarantor for a loan a third person had made to his brother. In his comment, Bartolus stated that it is one thing to ask someone to trust what an envoy says, or to recommend an envoy to him ; but quite another thing to grant the envoy the power to enter into a contract in the sender’s name. And so, general formulas such as « trust the bearer of these letters in everything he tells you about me », « trust [him], who is fully informed of my intention » or « trust him fully in this matter » only ask the other party to believe what the bearer of the letters says, not to negotiate with him. In other words, they are mere letters of recommendation, which do not bind the sender to uphold the action carried out by their bearer, unless the particular situation requires otherwise (for example, if the bearer is the sender’s factor or institutor in a certain business, because then the sender seems to be asking to trust his agent in relation to this specific business) or the sender explicitly assumes the risk of the action36. Bartolus’ opinion was accepted by later jurists like Baldus de Ubaldis and Bartolomeus de Saliceto (d. 1411), and was explicitly applied to the ambassador in the two most important late-fifteenth-century legal writings on this office, namely the entry « Ambasiator » in Johannes Bertachinus’ Repertorium iuris (1481), and Gonzalo de Villadiego’s Tractatus de legato (1485)37. Bertachinus referred to Bartolus’ comment on lex Lucius after briefly stating that an « ambassador (ambasiator) » could not enter into any contract in his principal’s name by virtue of mere credentials, and Villadiego reproduced a large portion of Bartolus’ comment (with only very slight changes) in the third section of his treatise, which deals with nuntii ad negotia38. These sources are clear evidence of the transfer of notions and rules from private to public law, and of the former’s contribution to the discussion of ambassadorial powers.
Although today we are used to the notion of perfect, or direct, representation, which is grounded solely in an agreement between an agent and his principal, this general notion did not exist in the Middle Ages and was not elaborated until early modernity, within the framework of a general theory of contracts based on notions like the subject of law and the autonomy of the will39. For a long time, even the de rato clause was not explicitly interpreted as granting an agent the power to represent directly his principal40. In fact, medieval scholarship did not envisage a general theory of representation, instead endeavouring to offer solutions in a variety of situations in which Roman law allowed people to act for others despite the general interdiction expressed in the Digesta (or at least what medieval jurists considered to be a general interdiction – « alteri stipulari nemo potest », Dig. 188.8.131.52)41. Civil and canon lawyers soon began to ponder issues like representation in lawsuits, the acquisition of possession or property through an agent, marriage by proxy, or the legality of taking an oath on the soul of one’s principal42. Within this debate, different types of agent were outlined, the most important of which were the nuntius and the procurator. Medieval jurists stressed the distinction between these two kinds of agent, considering the nuntius to be no more than the voice of his sender, whose words he repeated, and comparing him to a « magpie (pica) » and an « instrument (organum) ». In so doing, they implied that a nuntius had no autonomy or discretionary powers, but could only deliver a message or sign a contract which had already been agreed upon : everything a nuntius did was, in fact, automatically considered to have been done by his sender. In contrast, the procurator could negotiate and conclude an agreement : he was not simply held to be an instrument in the hands of his principal, but an agent provided with certain discretionary powers. As a result, the general view was that the act performed by a nuntius immediately produced legal effects in the principal’s legal sphere, while the act performed by a procurator only produced legal effects in the principal’s sphere after having been transferred to him (through a cessio actionis or a traditio)43. Other kinds of representative were the syndicus and the actor. According to Accursius (d. ante September 1262), although an actor could also « sometimes » represent individuals, both were usually appointed by a community : the actor on an ad hoc basis, the syndicus for all matters. While the Digesta equated the syndicus and the defensor civitatis in some passages44, Accursius used defensor as a general term for any agent of a universitas, and considered actor and syndicus to be kinds of defensores45. Given that a section de syndico (just as a section de procuratoribus) had been included in the canon law collections since the Breviarium extravagantium, canon lawyers extensively wrote on the office of syndic46. They also provided a decisive contribution to the conceptualisation of collegia and other bodies composed of a plurality of human beings as legal persons, using to this end the concept of persona ficta or repraesentata47. A syndicus who carried a procuration, therefore, had procuratorial powers and was held to represent the whole community – understood as a corporate body (universitas) – which had elected him48.
That said, it is worth pointing out that the abovementioned distinction of different kinds of agents was nuanced in legal scholarship. Bartolus de Sassoferrato, for example, argued that, in some situations, the procurator could act as a representative who produced immediate legal effects in his principal’s legal sphere. What were these situations ? Bartolus said that this happened when the procurator accomplished some « natural acts », or acts related to rights or obligations that were « so personal, that they cannot be transferred from the procurator to the dominus ». For instance, when a procurator was appointed to take an oath on his principal’s soul, or to take possession of a good, or to marry a woman on the principal’s behalf, he did not need to transfer the rights and obligations deriving from such acts to the principal, since they were immediately produced in the latter’s legal sphere. In such situations, Bartolus wrote, « the procurator directly represents the person of the dominus, just as a nuntius does »49. This is a key passage, in which the phrase « personam (alicuius) repraesentare directo » is used in a technical sense to define direct representation50. The originality of this affirmation deserves special emphasis. Admittedly, the concept of « representation » had already been used by Pierre de Belleperche (c. 1247?-1308) in relation to the action of a servus, which – according to Roman law – directly bound the dominus, but Bartolus’ remark was far more general, and also applied to the action carried out by a free man51. Moreover, as convincingly argued by Hasso Hofmann, the aforementioned concept of persona repraesentata – the first occurrences of which Robert Feenstra traced back to Jacques de Revigny (c. 1230/40-1296), and possibly his master Jean de Monchy (d. c. 1266) – did not refer to the exercise of vicarial representation (representation as Stellvertretung), but was intended to define, through a constructive use of fiction, corporations as abstract legal entities (representation as Nachbildung)52. However, we can see that, even in Bartolus, direct representation was not a general institution depending solely on the will of the parties : it was considered to be an exceptional case, and depended on the specific quality of the acts involved, which produced rights and obligations that could not be transferred to another person. This, at least, was the situation under the ius commune : from the late thirteenth century on, Italian statutes and customs began recognising a much broader applicability of direct representation, which undoubtedly facilitated commerce and economic exchanges53.
Not only was the distinction between different kind of agents nuanced in legal scholarship, but it was also scarcely observed in actual practice. In fact, in late medieval diplomatic practice, the terms nuntius, procurator and syndicus were often used interchangeably, or even together, or in conjunction with other titles like actor, factor, oeconomus, orator, commissarius, negotiorum gestor or ambasiator. The quality and powers of each envoy, therefore, always had to be inferred not from his title, but from the documents he carried with him and the precise powers that these granted him. In some cases, legal scholars, especially in the fields of canon and procedural law, recognised and partly accepted this redundancy, stating that, in the interpretation of the documents borne by the envoy, the principal’s intention had to prevail over the literal words54. In a famous passage of his Summa, Henricus de Segusio declared that « we do not give great force to names, for whether [an envoy] is called procurator, or syndic, or economus, or ass, or even if no name is expressed, it doesn’t matter as long as the mind of the person constituting him is clear »55. In other cases, however, legal scholars continued to use each term with due precision, as demonstrated by the aforementioned late-fifteenth-century works on the ambassador written by Johannes Bertachinus and, above all, Gonzalo de Villadiego56.
Roman law distinguished between two kinds of mandate, mandatum generale and mandatum speciale. A special procurator was appointed for a specific task, while a general procurator was appointed for the entirety of his principal’s business, either for a limited period of time or in perpetuity57. The degree of precision required for a mandate to be deemed « special » could vary in jurists’ opinions. In the fourteenth century, Guillaume de Cun (d. 1336) argued that a special mandate had to indicate both the exact situation in which the procurator was to accomplish his act, and the person with whom he had to deal ; Baldus de Ubaldis criticised this view, and stated that indicating the act to be performed in general terms sufficed58. On the other hand, a general mandate conferred rather limited powers and was of very little use in diplomatic practice59; however, it was considered to produce the same effects as a special mandate if a particular clause was added to it, granting the procurator plena potestas or libera administratio60. Like Cynus de Pistorio (c. 1270-1336), Baldus refers to the opinion of Jacobus de Arena (d. ante 1302), according to which syndics appointed by universitates like cities, castra and villae held a publicum officium, and therefore did not need a special mandate to perform acts which normally required such a mandate ; syndics appointed by collegia without public status, by contrast, did not held a publicum officium and so needed a special mandate to perform those acts61. Commenting on Cod. 2.4.12, which referred to a compromise (transactio) made by the administrators of a city, both Bartolus and Baldus observed that, since compromises required a special mandate, the mandates granted to city administrators had to be understood to include implicitly libera administratio62. However, Baldus does not seem to have fully adopted Arena’s view, as revealed by his comment on a clause of the Peace of Constance which required that the Lombard cities – except for those in which a bishop held comital office – send a nuntius to the emperor every five years, to request the investiture of new consuls. After reproducing a previous gloss by Odofredus clarifying that, in Italy, such nuntii were called syndici, Baldus here mentioned Jacobus de Arena’s opinion, but then, drawing on Johannes Andreae, concluded that it was « safer » to say that, when they were required to accomplish acts requiring a special mandate, syndics had to bear either such a mandate, or a general one cum libera63.
The impact of the principles and rules of private law agency on diplomatic representation and public international law is particularly evident in the juristic discussion about ratification. In medieval legal scholarship, ratihabitio or ratificatio was the act whereby a dominus could accept, with retroactive effect, an act performed by a false procurator – that is a procurator acting without procuration : according to the principle « ratihabitio mandato comparatur », after the ratihabitio a legal act performed by an unauthorised agent had the same effect as if it had been performed by an authorised agent64. But what about an act performed by an ambassador within the boundaries of his mandate ? Did it immediately bind his principal ? Or did the principal have to confirm it in some way ? In his study on the office of ambassador in the Middle Ages, Donald Queller demonstrated that representation in late medieval diplomatic practice could be either direct or indirect. According to Queller, ratification was not always necessary in strictly legal terms to make a treaty binding upon the principal, although it was nonetheless sometimes performed to give more publicity or solemnity to the outcome of negotiations65. In particular, it was not necessary when an ambassador had been granted full powers (plena potestas) to conclude an agreement. This thesis seems to have been confirmed by Victor Crescenzi’s research on late-thirteenth-century Siena, where two kinds of agent could be dispatched : on the one hand, « ambaxatores et sindici » were empowered to negotiate and enter into agreements that were directly binding on the city ; on the other hand, simple « ambaxatores » could either present their commune’s proposal and conclude an agreement if this were fully accepted (but had no power of negotiation), or negotiate a deal (although they were powerless to conclude it)66. When it was performed, the ratification of treaties normally entailed the taking of an oath, either by the political leaders themselves or by duly entrusted envoys – which brought the enforcement of these treaties under the jurisdiction of ecclesiastical courts67. In their discussion about the extent to which public officials were entitled to represent parties in concluding agreements, late medieval jurists largely focused on the function performed by public notaries in the conclusion of contracts between private individuals68; other kinds of representative, however, were mentioned too – although much less frequently69. Angelus de Ubaldis (1327/28-1407), for example, wrote that « those who fulfil a public office […] can form the words of the payment directly in the person of the principal », with the result that « any universitas is entitled to the actio [arising] from a contract [concluded by] his administrator without cessio »70. In one of his consilia, dealing with a provision in the Florentine statutes about the contract concluded by a public notary (notarius) on behalf of a third person, Angelus noted that in some cases the ius commune admitted the possibility of producing legal effects in the principal’s legal sphere immediately, that is without having to transfer these effects to him (« sine cessione »). Such cases, according to him, included the acts accomplished by a syndicus and an actor universitatis : these titles were, in fact, often attributed to ambassadors in medieval diplomatic practice and theory71.
III. The influence of the late medieval notion of diplomatic representation on early modern scholarship on international law
Later sources give more explicit instances of the immediate validity of an act performed by ambassadors provided with full powers being accepted in legal doctrine, even though these sources also warn us that, in practice, ratification was often requested72. Although it is not possible, within the limits of the present article, to carry out a close analysis of the issue of diplomatic representation in early modern scholarship on international law, some examples will illustrate the persistence of the late medieval notion and approach in the discussion of the issue right up until the late eighteenth century, thereby allowing us to appreciate the importance of the contribution made by the ius commune to the early modern discussion of the status of the ambassador. A first example is provided by Pope Pius II’s Commentaries, written in the years 1462/63 : in one passage, Pius II talks about the Florentine citizens who, when reminded, in early 1460, of the promise, made in Mantua by their ambassadors the year before, that Florence would contribute to the Crusade, answered that the ambassadors
had made the pledge in the name of the city, acting on [a mandate] (mandatum) from the priors and certain richiesti ; this was normal procedure, but the pledge would not be valid until it was ratified (ratum habeatur) by the Great Council of the City. Only rarely did this body refuse to ratify what ambassadors with such a [mandate] (mandatum) had promised, but in this matter, which was of the gravest importance, there was not the slightest chance that the people would give their consent. […] The pope was astonished that the Florentines could be so shameless. He inveighed at length against the injustice (iniustitiam) of those who lied and deceived in matters concerning the defence of the faith. To give ambassadors a mandate that had no authority (invalida… mandata) was a disgrace ; it was even worse to exploit it to their own advantage […]. But the pope was wasting his time. He decided to take what he could get. The state [i.e. Florence] was acting like a haggler (not to say a whore) and he dispensed with further negotiations73.
This case was cited by Alberico Gentili in his De iure belli (1598), in the chapter of book III concerning the law of agreements, to argue that a prince was always bound to observe a treaty concluded by his ambassador within the limits of his procuration, and had no right to invoke his discretionary power of ratification. Commenting on the case of the Florentine ambassadors, he asked : « How will you be able to rely on a mandate if you accept this way of thinking ? Or shall a thing so essential in human intercourse as the employment of procurators be abolished ? Against those Florentines, Pius II declared that it was shameful to send invalid mandates, and still more so to make use of them »74.
In the same chapter, Gentili mentioned another case, which became quite famous in early modern legal scholarship. According to Francesco Guicciardini’s History of Italy, in 1503, during the war between France and Spain in the Kingdom of Naples, the Habsburg archduke Philip obtained from his parents in law – King Ferdinand II of Aragon (Ferdinand the Catholic) and Queen Isabella I of Castile – full powers (« ampia facoltà e libero mandato ») to conclude a peace with Louis XII. The peace was actually reached, and made public in Blois on 11 April, when both the French king and the archduke (as procurator of Ferdinand and Isabella) took an oath75. Louis XII immediately started to execute the peace, despatching someone to ask Ferdinand and Isabella to ratify it76 ; the latter, however, ignored the request for several months, until they expressly declared that they were not going to comply with it because, although Philip had received full powers (« mandato […] libero e amplissimo »), his instructions were limited (« istruzioni […] limitate ») and he should therefore have obeyed them77. Philip firmly denied having exceeded either his mandate or his instructions, and Louis XII – extremelyannoyed about what had happened – expelled the two ambassadors who had been sent to him by the Spanish monarchs78. Discussed by Jean Bodin in Les six livres de la République79, this case was referred to by both Gentili and Hugo Grotius. The two jurists expressed similar opinions : the former commented that Ferdinand and Isabella’s behaviour was « unbecoming for a prince, who should say : “I have spoken once and for all. What I have written, I have written. One pen, one tongue […]”. But these Catholics have two pens, and two tongues »80; as for the latter, he explicitly gave priority to the mandate over secret instructions, saying that when a treaty is concluded by an agent « whoever has appointed [him] is bound, even if the agent, while yet within the limits of his public function, has acted contrary to secret instructions »81.
In the second half of the seventeenth century, this doctrine was reaffirmed both by Samuel von Pufendorf, the holder of the first chair in the law of nature and of nations (established in Heidelberg in 1672)82, and by the Dutch ambassador Abraham de Wicquefort, who in 1680-1681 published the most important treatise on the ambassador of his time (a re-elaboration of his previous Mémoires)83. Wicquefort, in particular, clarified that « the Powers, with respect to an Embassador, are nothing else, than what a Letter of Attorney is in reference to a private Person », and considered ratification not to be necessary in strict legal terms, although he conceded that it could be usefully requested as a proof of the counterpart’s good faith84 :
Ratification is not an essential part of a Treaty ; which is by so much the more evident, not only because a Treaty is a common and publick Instrument, and the Ratification a private and particular one ; but also because a Treaty would be good, and subsist without the Ratification, if it were certain that the Treaty, and he that has made it, would not be disavow’d. […] As the Civil Laws oblige a private Person to ratify what his Mandatary has done by virtue of his Procurations ; so the Law of Nations obliges the Prince to ratify what his Minister has done by virtue of his Powers : especially if the Powers are full and absolute, without any Clause or Condition, to limit or restrain the same85.
Admittedly, in the eighteenth century, Cornelius van Bijnkershoek and Emer de Vattel cast some doubt on the automatic validity of treaties after their conclusion, observing that the rulers of their time usually reserved the right to ratify whatever they agents had concluded : while the former accepted this practice based on the idea that custom determined the principal part of ius gentium86, the latter noted that, since « sovereigns can not be constrained […] to fulfill their engagements, it is usual not to consider their treaties as final until approved and ratified by the sovereigns themselves », although he went on to affirm that « cogent and substantial reasons are needed to justify a sovereign in refusing to ratify the act of his plenipotentiary »87. However, Georg Friedrich von Martens would still write, in his Précis du droit des gens moderne de l’Europe fondé sur les traités et l’usage (first published in 1789, and translated into English in 1802), that « every thing that has been stipulated by an agent in conformity to his full powers, ought to become obligatory for the state, from the moment of signing, without even waiting for the ratification ». In Martens’ view, it had only recently (« now ») become common, « not to expose a state to the errors of a single person, […] that public conventions […] do not become obligatory, till ratified »88. As his reflection on the issue reveals, the traditional notion of diplomatic representation, grounded in the notion of private law agency developed by late medieval ius commune jurists, was – at least in theory – still valid at the turn of the nineteenth century. It was only during that century that a new doctrine would arise, and ratification would become a fully free and discretionary act, performed by national Parliaments as the final stage of the treaty-making process89.