Rivista internazionale di diritto comune 18 (2007) 43-70

kenneth pennington

The “Big Bang”:

Roman Law in the Early Twelfth-Century

The rebirth of Roman law in Italy during the twelfth century is an old story with a long pedigree. In the thirteenth century Odofredus de Denariis, a native of Bologna, who taught Roman law in his home town’s law school, regaled his students with what must have been Bolognese oral traditions about the school’s origins[1]. According to Odofredus Pepo “of no name” first began teaching Roman law. He was followed by Irnerius, and later by the “Four Doctors,” Bulgarus, Martinus, Hugo, and Jacobus[2]. The exact chronology of the teaching of Roman law in Bologna has given scholars a golden opportunity. Burdened by almost no evidence and bolstered by a healthy skepticism, they have created alternative universes limited only by their imaginations. Traditionally Pepo’s and Irnerius’ teaching careers were dated between 1075 and 1125. The “Four Doctors” succeeded the two founders and taught in the middle of the twelfth century (ca. 1130- ca. 1170). Because the evidence is so sparse and because much of it is from later reports – like that of Odofredus – some scholars have doubted the existence of Pepo and the teaching activity of Irnerius[3]. Examining the histories of individuals and their works is one way to answer the question about the origins of the teaching of Roman law. There is another way to approach the question. In this essay I will concentrate on when can we see the first signs that Roman law has begun to shape other legal systems and institutions. And when it began to shape the thought of non-Roman law jurists and the statutes of local communities. When, in other words, Roman law became a “practical science”.

In an essay that I published recently in this journal I examined in some detail the influence of Roman jurisprudence on the legislation of King Roger II of Sicily[4]. My main conclusion in that piece was that King Roger imported jurist(s) from Bologna to shape his legislation and his compilation of laws that he promulgated ca. 1140. I would like to extend my examination of the influence of Roman law in the first half of the twelfth century in this essay. Before I do, I should explain exactly what I mean by the term “jurisprudence”. It is a term often misunderstood, even by scholars. The ancient Roman jurist Ulpian gave a philosophical definition of the term when he defined jurisprudence as a “knowledge of justice and injustice”[5]. The modern definition of jurisprudence encompasses the Roman definition as well as jurists’ “knowledge of or skill in law”, the more general “the science of... human laws,” and, perhaps most importantly, their knowledge of “the practical science of giving wise interpretation to the laws and making a just application of them to all cases”[6]. The last definition neatly combines the more philosophical understanding of the term with the practical and is what I mean when I use the term.

The jurists of the early twelfth century had no jurisprudence to draw upon. When they confronted Justinian’s Corpus iuris civilis and the libri legales they struggled to understand basic institutions like court procedure and fundamental concepts and principles of law. The jurists also had to confront the rules that either ancient Roman jurists or imperial statutes had laid down to govern very practical questions. What should the proper length of time for a prescription of a property right be? Could women testify in court? Could they hold public office? What role should ignorance of the law play in determining guilt? These questions could be simple and practical, but even the most simple had moral and ethical ramifications. The first teachers of Roman law and their students had to decide how Justinian’s authoritative compilation could be used to provide guidance for constructing legal systems that were based on justice and reason, not on arbitrary customary practices. Manlio Bellomo has aptly described Europe as being a “An Age without Jurists” before the revival of Roman law[7]. Without jurists there cannot be any jurisprudence. Without jurisprudence there cannot be a legal system that is based on equity and justice. As any student of statutory interpretation learns very quickly, laws cannot just be read literally; they must be interpreted[8].

I would like to begin my examination of the influence of Roman law outside of Sicily with Gratian. His use of Roman law in his Decretum is another example of how the teaching of a dead legal system began to shape law and institutions. Since Anders Winroth and Carlos Larrainzar discovered earlier versions of Gratian’s Decretum our understanding of Gratian and the text that he used for teaching canon law in Bologna has been greatly deepened. Many questions, however, remain about the development of Gratian’s book[9]. In the comments that follow I will assume that a manuscript in St. Gall is the earliest version of the Decretum and was compiled in the 1120’s. A Florentine manuscript is the next earliest version and was compiled in the 1130’s[10]. I believe that the Vulgate text of Gratian’s Decretum was finished shortly after 1140. Not all scholars yet agree with this history of Gratian’s text, but I will try to illustrate how Gratian’s use of Roman law supports my general conclusions about its evolution. Some scholars have argued that the St. Gall manuscript is an abbreviation of the Florentine. I shall present evidence that the St. Gall cannot be an abbreviation of Gratian’s text. I shall also demonstrate how Gratian began to use Roman law to establish procedural rules and norms in canonical jurisprudence in the first two versions of his Decretum.

In the St. Gall and Florence manuscripts Gratian used or cited Roman law in a number of places:[11] he placed a reference to a chapter from Justinian’s later legislation when he discussed the time permitted to litigants to appeal unfavorable decisions and when he defined the difference between judges and arbiters[12]; he defined the restitution of office or property to plaintiffs before their cases are heard in court using Roman law terminology;[13] he cited two laws from Justinian’s Codex when he discussed a women’s right to testify in court[14]; and he employed Roman law concepts and borrowed the language of ancient Roman jurists when he formulated his definitions of marriage in Causa 29[15].

In the Second Causa Gratian discussed procedure in the courtroom. He focused on the legal questions that arose from the case of a cleric who had been accused of sexual crimes. In the sixth question of the Causa Gratian broached the issue of appeals to higher judges. How long did litigants have to appeal? In the St. Gall manuscript Gratian stated that certain Theodosian laws permitted appeals within five days after the court had rendered its decision. Justinian, he continued, corrected the Theodosian rule and permitted ten days for an appeal[16]. Gratian did not include both texts of the laws that he cited even though he could have taken them from a number of canonical collections[17]. He did add both (C.2 q.6 c.22 and c.28) to his next version of the Decretum in the Florence manuscript[18]. In St. Gall he had not included the entire text of Justinian’s legislation that he cited for the “ten-days-to-appeal rule,” but only the incipit[19]. Later Gratian added the entire text to the Florence manuscript that he took from a Latin translation of Justinian’s Novella 23 (dated 536)[20]. Gratian probably knew several textual traditions of Justinian’s law; however, we cannot know from where he took these texts. It is not too far-fetched, but by no means certain, that if Gratian had a glossed text of Bulgarus’ treatise like that in Vat. lat. 8782, he could have learned about the ten-day rule for appeals and Justinian’s Novella there[21]. What is striking about his adoption of Roman law rules for appeals in canon law is his tacit assumption that Roman law contained norms that could be accepted into canon law. His presumption is not surprising. Canonical collections had long included Roman law texts[22], and the recovery of the text of Justinian’s Digest in the late eleventh century prompted canonists to include texts from it into their collections[23]. There was a conviction among the canonists that Roman law could contribute to canonical jurisprudence and could solve practical questions – as, for example, how long litigants had to appeal a court decision.

Bulgarus’ tract on procedure illustrates my generalization. It is one of the first “treatises” written by a Roman lawyer. Both he and the recipient of his tract, the Papal Chancellor Haimeric, must have presumed that Roman rules could be applicable to procedural rules in ecclesiastical courts. Gratian would have known Bulgarus in Bologna. He knew at least one of Bulgarus’ other works as well. Bulgarus wrote a short treatise on De iuris et facti ignorantia, a subject that Gratian also took up in his Decretum[24]. The similarities of theme and language demonstrate a probable connection between the two jurists in the 1120’s and 1130’s[25].

Another connection between Bulgarus, Roman law, and Gratian can be found in a passage in which Gratian discussed the differences between arbiters and judges. In Causa two where he discussed appeals, Gratian noted that there are some judges from whom one may not appeal[26]. He laid out the differences[27]:

Here, it must be understood, (i.e. C.2 q.6 c.33) he speaks of arbitration. Some judges are arbiters and some are ordinary judges. Ordinary judges are appointed by the pope as ecclesiastical judges or by the emperor as secular judges and thus accept legitimate power and authority. Arbiters having no power <obtain it> only with the consent of the litigants <when they> are elected judges, who, so that their decisions would be firm, are chosen by both sides. Those who are ordinary judges and arbiters are considered delegated arbiters in the laws. Others are only arbiters. Arbiters and ordinaries are those who have legitimate power from those who elect them to arbitrate and who are not subject to their authority and jurisdiction. Nonetheless one may appeal from arbiters who sit as judges but not as ordinaries, by privilege of their ordinary dignity, if one has been gravely injured.

There are several points to make about this text. First, the terminology is purely Roman. The concept of the ordinary judge is found throughout Justinian’s Codex and the Digest. One cannot find any discussions of ordinaries in earlier canonical collections. Arbiters were also an important part of Roman jurisprudence. Roman law defined two types of arbiters: an “arbiter delegatus (datus vel pedaneus)”[28] and an “arbiter ex compromisso”. By the twelfth century the concept of the “arbiter delegatus” had disappeared from practice if not from the pages of the Corpus iuris civilis[29]. In his first discussion of arbiters Gratian included a reference to “arbitri delegati,” which was deleted from all subsequent versions of his text. That is a significant piece of evidence that the St. Gall manuscript is not an abbreviation. No abbreviator would have added that detail. 

In his tract on procedure Bulgarus outlined the general differences between judges who were appointed by public authorities and arbiters[30]. In fact, Gratian’s discussion of arbiters and judges is more sophisticated than Bulgarus’. There has been some debate about the date of Bulgarus’ treatise, but its contents speak for an earlier rather than a later date. The window for its composition is the period that Haimeric was papal chancellor, 1123-1141. The evidence from Gratian would also argue for an earlier rather than a later date (there is another crucial piece of evidence that I will present shortly). Until now no one has uncovered a link between Bulgarus’ tract and the work of the papal chancellery. Such evidence does, however, exist. During the pontificate of Pope Innocent II (1130-1143) the pope sent two almost identical letters to the bishops of Northern France and all the bishops of Germany warning them that they should not hinder appeals to the Apostolic See[31]. One of the letters is dated to the year 1133-1136 and the other might be dated 1137. The second letter may bear the endorsement of Haimeric, the papal chancellor. The letters are laden with references to papal authority and with quotations from Roman law.

After an arenga that emphasized that the primacy of Saint Peter, Innocent (or Haimeric or his clerics) stated that since Peter was the prince of the Apostles and had been given the headship of the Church by Christ, the pope had the duty to correct mistakes and to give to each person his rights[32]. The pope’s reference to giving each person his rights is a clear reference to a famous passage in Roman law[33]. He then stated that “causae maiores” were reserved to the jurisdiction of the Roman Church. The doctrine that the papacy reserved more important cases to the papal curia first entered canonical jurisprudence in the early Middle Ages, was confirmed by the Pseudo-Isidorian Decretals, and was well-established common doctrine by the twelfth century[34].

The pope declared that everyone knew that the custom of appealing legal decisions was necessary because then the “iniquity and ignorance of judges” could be corrected. Bulgarus’ treatise on procedure that he sent to Haimeric ended with the same issues; the text also cites the Digest – the earliest example that we have of a citation to the Digest in a twelfth-century papal letter:

In the next sentence Innocent made an even more explicit reference to Bulgarus’ tract:

Finally Innocent used a norm of Roman law to impose penalties on those bishops who did not permit appeals to Rome. “Imperial statutes have established that if ordinary judges reject appeals they will be fined thirty pounds of gold. Their officials will also be assessed an equal amount unless they can prove that they publicly opposed the ordinary’s refusal to permit an appeal”[35]. Innocent cited a decree of Constantius II (337-361) to establish his point[36]. It is also significant that Innocent did not alter the Roman terminology “iudices ordinarii” to describe the judicial role of bishops. As we have seen Gratian had introduced this term into canonical jurisprudence in the first version of his Decretum a few years earlier. Innocent’s mandate to fine episcopal courts for hindering appeals to Rome never became a canonical norm. The appeal process was just being introduced in canonical courts in the first half of the twelfth century[37]. Within a few years no pope had to threaten bishops for not respecting litigants’ right to appeal. Only in the second half of the twelfth century did appeals to the papacy become an almost unregulated floodtide[38].

Innocent’s letter is another signpost for the importance of Roman jurisprudence in shaping institutions outside the classrooms in Bologna. The letter also provides a terminus ad quem for dating Bulgarus’ tract. Bulgarus’ discussion of judicial procedure is elementary (another example of this in a moment), omits much of the complexity embedded in his sources, and was probably written before 1130 – in the formative years of the law school[39]. In any case it cannot have been sent to Haimeric after ca. 1135 and most likely some time before. Innocent’s letter provides some evidence that the Roman Curia already had a more sophisticated understanding of judicial procedure than they had found in Bulgarus’ tract.

Anders Winroth has discussed Gratian’s use of the Roman law of restitution in Causa three at some length.  He used this text to argue that Gratian became more sophisticated and knowledgeable in his use of Roman law as he compiled later versions of his Decretum.  There is no question that Winroth is right about Gratian’s growing sophistication.  By the time he compiled the last recension of his Decretum his understanding and use of Roman law norms had developed considerably (see the example of C.3 q.7 c.2  that I discuss below) [40].   Winroth points to different levels of competence in treating Roman law texts in the earlier recension of the Decretum ― as well as other evidence ―  and concludes that the person who compiled the last recension of the Decretum was a different person from the compiler of the earlier recensions[41].  One reason why he comes to this conclusion is that he dates the earliest recension of Gratian’s Decretum to ca. 1140 and the last recension to ca. 1150 or even later[42].  If Gratian had compiled the Decretum aaround mid-century, we would be presented with a serious problem: why did Gratian not draw upon more Roman jurisprudence when he compiled his first two recensions in the 1140’s even though the teaching of Roman law had been under full sail for decades?  Did he reject Roman jurisprudence?  Was he ignorant of it?  Did the teaching of Roman law begin much later than we have traditionally thought (Winroth’s conclusion)?  However, if we date Gratian’s recensions between 1120 and 1140, the problem disappears.  Gratian absorbed his knowledge of Roman law as the teaching of Roman law in Bologna evolved.  As he learned more about Roman jurisprudence, especially its principles of procedure, he incorporated what he had learned into his Decretum.

     The evolution of a dictum in Causa three illustrates my point.  Gratian posed the hypothetical that a bishop had been ejected from his see.  He asked whether the bishop should be restored to his see before his complaint was heard in court.  He cited four texts from Pseudo-Isidore that all quoted the Roman law remedy, “restitutio in integrum” ― the principle dictated that plaintiffs should be restored to their rights before a trial began.  Gratian’s texts, however, did not provide all the answers to how a plaintiff’s rights should be restored.  In his dictum Gratian asked whether the bishop should be granted possession of his bishopric through the judge’s will (animo) or possession in fact (possessio corporalis).  Gratian’s distinction was based on Roman property law.  In the first case the bishop had a legal claim to his see but not actual possession; in the second he had full, legal possession of his see.  Gratian concluded that the bishop must receive full rights to his see in court (per presentiae executorem)[43].

In his next recension Gratian tweaked his text.  One of the Pseudo-Isidorian texts used the term “presentialiter” to describe the how the bishop should be restored to his position.[44]  This was not a Roman law term but entered the canonistic tradition through Pseudo-Isidore[45].  In his next revision Gratian fiddled with the wording of his dictum and made it very clear that a judge’s declaration in court did not bestow full property rights.  He emphasized that the bishop should be present (praesentialiter) in court and that the judge had to mandate an order that the bishop must receive his see “corpore”[46].  Winroth argues that Gratian’s dictum reflected vulgar Roman law, not Justinian’s terminology and thought.  Nonetheless, Book 43 title 2 of the Digest supports the early Gratian’s formulation of the legal problem in his dictum.  It is clear from many texts in the Digest that the will (animo) to possess property was a necessary step towards the acquisition of property. The will could be present without physical possession.  In order to obtain full rights of ownership the will must be followed by physical possession.   A text of Paulus stated the Roman legal distinctions most clearly:  “We acquire possession factually and through our will, not through our will alone nor factually alone”[47].  Gratian’s point in his dictum would have been supported by this text and many others[48].  The bishop must not be satisfied with only receiving a judge’s declaration restoring him to his see.  Gratian’s quotation of the judge’s words “Censeo te in integrum restituendum” was a warning to his fledgling students[49]. A declaration of the judge should not be accepted as a substitute for full restoration of the bishop’s rights.   The bishop and his procurators must be certain that his right to be restored to his see would be vindicated outside the courtroom.  In his final revision of dictum, Gratian introduced two more Roman law terms, “naturalis possessio’ and “corpus alienum”.  The first was a technical term in Roman law that described mere possession, without an intention (animus) to own.  The bishop must, argued Gratian, have actual possession of his office.  “Naturalis possessio” made that point absolutely clear. Gratian used “corpus alienum” to acknowledge that the bishop could gain possession through a procurator or some other person.  Gratian knew that this was a very real possible situation if the plaintiff were in a court that was distant from his see[50].  Roman jurisprudence used the term to describe a person who had the will (animus) to acquire ownership but did so through another (corpus alienum).[51]  As Gratian wrote and revised this dictum he betrayed no ignorance of Roman principles of possession, rights, and ownership.  Quite the contrary, he based his thought on Book 41, title two of the Digest.  His revisions reveal a growing understanding of the Roman law of property and principles of ownership.

Gratian’s next brush with Roman law in the earliest versions of his text occurs in Causa 15. He posed a case similar to the hypothetical of Causa two: a cleric fornicated. Could the woman’s testimony convict him of the crime?[52] Bulgarus treated the same question in his treatise. It is revealing to compare the two:

Bulgarus and Gratian broached two different questions. Bulgarus discussed all persons who were generally prohibited from testifying, while Gratian focused on whether women could accuse the clergy. Gratian went on to note that in Roman law women were generally not permitted to bring suits but could do so if their own families or interests were involved[53]. He then cited examples from the Old Testament in which women exercised judicial offices. However, “today because of the sin that women committed, the Apostle declared that they should feel shame and be subjected to men”[54]. After emphasizing the misogynist tradition in the Western tradition, Gratian turned to Roman law. “On the other hand, nevertheless, while widely and universally women are not permitted to bring suit in court, there are certain crimes for which women are not forbidden from bringing suit”[55]. To support his conclusion Gratian cited a constitution of Diocletian and Maximianus in which the emperors permitted women to bring suit for “certain reasons”[56]. This statute was not part of the canonistic tradition. Gratian must have taken Diocletian’s and Maximianus’ statute directly from the Codex. Proof of its source can be found in the inscription that he provided in the St. Gall manuscript: “Imperatores Diocletianus et Maximianus”. The Florence manuscript and all the later versions of Gratian’s Decretum attribute the text only to Diocletian[57]. An abbreviator of Gratian’s text would have never bothered to restore the original inscription in the Codex. Once again Bulgarus’ glossed text may have provided Gratian with the information that he needed to deal with the question of women’s legal rights. In Bulgarus’ text in Vat. lat. 8782, the glossator cited Diocletian’s and Maximianus’ constitution next to the place where Bulgarus made his general statement about a woman’s lack of rights to bring suit[58].

After Diocletian’s and Maximianus’ constitution Gratian inserted another text from the Codex that equated the crime of simony with treason and other criminal cases[59]. In his comment to that constitution Gratian concluded that since women were permitted to bring suit for treason and for crimes, they could also accuse clerics of simony[60]. In his next draft of the Decretum in the Florentine manuscript Gratian added four texts from the Digest and one from the Codex that introduced more nuance into his general statement. He argued that women could bring suit for crimes like adultery or treason and also move cases involving liberty[61].

One other point should be made. When Gratian placed these texts from the Digest in his Decretum he provided them with inscriptions with the most primitive form of citation found in the juristic literature: “in libro digestorum” or “digestis”[62]. As Gero Dolezalek has pointed out, this form of citation was no longer used after the middle of the twelfth century[63]. Dolezalek also has noted that the capital with an angled line through the letter D became common practice by the middle of the twelfth century. In the second half of the twelfth century the ff. became standard. Gratian commonly used “In digestis” even in the last version of his Decretum, which is good evidence that he finished it earlier rather than later in the first half of the twelfth century.

The last place I will discuss where Gratian used Roman law to shape his thought was in Causa 29. He posited a case in which a man married a woman under false pretenses. She thought he was a nobleman; he was actually a slave. Gratian asked whether this was a valid marriage. He took the opportunity to define a Christian marriage as he solved the problem that he had posed[64]. The textual tradition of the long dictum that constitutes Question one of Causa 29 is very complicated and significantly different in the three versions of Gratian’s text. It is also, perhaps, one of his most creative causae. When he analyzed the case he created a doctrine of error in marital contract law that still has jurisprudential force in canon law today[65].

Borrowing from Roman law, Gratian defined marriage and then described the key element of a valid marriage:

Marriage is the union of a man and woman keeping an undivided way of life. Mutual consent makes marriage... Consent occurs when two or more perceive the same thing. [Coniugium uel matrimonium est uiri et mulieris coniunctio indiuiduam uitae consuetudinem retinens (Instit. 1 9.1.). Item consensus utriusque matrimonium facit... Consensus est duorum uel plurium sensus in idem (Dig. 2.14.1.2)].

Gratian’s definition of marriage is taken from Justinian’s Institutes 1.9.1:

Nuptiae autem sive matrimonium est viri et mulieris coniunctio, individuam consuetudinem vitae continens.

Anders Winroth has argued that Gratian did not draw his text from Roman law but from an obscure French treatise on the sacraments (or a related work)[66]. The definition in the Institutes was wide-spread and found in many canonical collections[67]. For that text there was no need for Gratian to turn to an anonymous northern author. Winroth also thinks that since the French author and Gratian have “retinens”at the end of the definition instead of “continens” as found in Roman law, this variant proves that Gratian did not take his text from Roman law. However, that particular reading is also found in a canonical collection from Central Italy that was close cousin to collections that Gratian used[68]. But the St. Gall manuscript proves conclusively that Gratian and his readers thought that the text was taken from Roman law. St. Gall manuscript abbreviates the sentence: “Coniugium uel matrimonium est ui. et m. c. in. u. c. c.” Medieval authors and scribes only abbreviated texts with initials that were very well-known, like passages from the Bible. It should also be noted that in Gratian’s Urtext in St. Gall the reading was “continens” not “retinens”. This too is a change that no abbreviator would have made.

Gratian’s definition of marriage may be ultimately based on Roman law, but it had been thoroughly “canonized”. However, his analysis of the central act – consent – that created a valid marriage was taken from an excerpt of Ulpian in Roman contract law:

Dig. 2.14.1.2: Pactum autem a pactione dicitur – inde etiam pacis nomen appellatum est – et est pactio duorum pluriumve in idem placitum et consensus [Moreover pact is derived from agreement (the word peace comes from the same origin) Agreement is the consent of two or more persons to the same thing].

One may have doubts that Gratian went into the Justinian’s Digest to find a definition for legitimate consent[69]. Again the St. Gall manuscript provides conclusive evidence that Gratian had Ulpian’s definition in mind when he wrote his dictum. Roman law always used standard names when referring to persons. “Titus” and “Seius” were most common. In the title on contracts Stichus and Pamphilius were used, just a few chapters from Ulpian’s definition of consent. Gratian wrote[70]:

Vnde sciendum est quod error alius est personae, alius fortunae, alius conditionis, alius qualitatis. Error personae est ut cum putatur Stichus ipse est Pampbilius.

Dig. 14.1.7.2 is another excerpt from Ulpian and reads:

Ulpianus... et ideo puto recte Iulianum a mauriciano reprehensum in hoc: dedi tibi Stichum, ut Pamphilum manumittas: manumisisti: evictus est Stichus [I think that Julian was rightly reproved by Mauricianus in the following case: I gave Stichus to you so that you would manumit Pamphilius. You manumitted Stichus].

The first point is that the Roman jurists always used Stichus and Pamphilius as generic names for slaves. In Gratian’s Causa 29 the error of person was about a slave. It is not clear in the text of the Digest whether the slave owner purposely manumitted the wrong slave, but Gratian might have thought so. In any case, the St. Gall manuscript has generic slave names taken from the Roman law of contracts. However in the Florence manuscript and in all subsequent versions of the Decretum, Gratian changed the names to Virgil and Plato[71]. He may have made the change simply because his canon law students were flummoxed by the Roman names – or for some other reason. Whatever his reasons this text is also good evidence that Gratian dipped into the Digest while formulating his thought in Causa 29 – and that the St. Gall manuscript is not the work of an abbreviator. No abbreviator would have changed Virgil and Plato to generic slave names taken from the Digest. He might have been clever but not that clever.

There are two other probable influences of Roman law in Causa 29. Gratian broke new ground with his classification of error into four elements: person, fortune, condition, and quality. He could have found some inspiration for his analysis in Justinian’s legislation that treated people who married slaves whom they thought were free[72]. Although Justinian did not use the word “error of person” in his statute, it was the topic of his legislation[73]. If Gratian was not aware of this legislation, other canonists were. One of them added an authentica based on Novella 22.11 that was commonly written into the margins of Codex manuscripts during the early twelfth century. It summarized Justinian’s legislation on persons who mistakenly married slaves (for a further discussion of this text see below)[74]. Gratian did use authenticae added to the Codex in the last version of his Decretum. His archaic citation to one of them confused later scribes of Decretum manuscripts[75].

In the second version of his Decretum Gratian inserted a passage in which he attempted to further clarify the distinction that he wished to make about errors that negated consent[76]. He described what he called “error materiae” Suppose, he wrote, that “someone promised to sell me gold and instead of gold gave me orichalcum and thus deceived me, might I be said to have consented to orichalcum?” Ulpian used the a similar hypothetical in the Digest[77]. Anders Winroth, again, thinks that Gratian did not consult the Digest, but his case is stronger in this instance because he can point to a number of authors who used the same language[78]. I think, however, that Gratian’s use of Roman law in Causa 29 could lead one to a different conclusion.

Gratian added the vast majority of Roman law to his Decretum in his last version of the text. He inserted forty-three more passages, primarily in the causae in which he treated procedure. In its final form the Decretum presented a much more sophisticated analysis of procedure than Bulgarus had given in his treatise – but Gratian benefited from Bulgarus’ path-breaking work composed at least a decade earlier[79]. In many cases Gratian simply incorporated texts taken from Roman law without much alteration, but in a few cases he creatively shaped the texts of Roman law to suit his purpose. A particularly good example of his methodology is in Causa three where he treated the question who could plead a court case. He relied on texts in the Digest for his answer.

As a comparison of the texts demonstrates Gratian rearranged and clarified the material that he found in the Digest. Adam Vetulani conjectured many years ago that Gratian did not put much Roman law into early versions of his Decretum. When he did turn to Roman law he inserted texts with a sure hand and with a sophisticated understanding of Roman jurisprudence[80]. He added these extensive Roman law texts to the last recension of the Decretum ca. 1140.

One last important piece of evidence about the early use of Roman law in canonical texts during the 1130’s is provided by marginal additions of Roman law in the St. Gall manuscript. Several of them were not incorporated into Gratian’s Decretum[81]. Consequently, we may assume that Gratian was not responsible for these additions. Rather, the form in which the texts were added indicate that they were placed in the margins by the scribes of a scriptorium where this version of Gratian’s Decretum was being used, presumably, I shall argue, at a center where Roman and canon law were taught. The scribal hands that added them were professional; in three cases the scribe even left a space for a rubricator to provide the first letter of the incipit[82]. These additional marginal texts were, in other words,  not added by a jurist who owned the manuscript and made notations to the text. Rather they are the product of the scriptorium that produced it. These additional texts, the glosses and cross-references in the margins are solid evidence that St. Gall was used for teaching canon law. The additional texts of Roman law, as we shall see, raised significant points of law that Gratian had ignored.

The jurist(s) who originally placed these texts in the margins of this early version of Gratian’s Decretum could have done so only with good knowledge of contemporary developments in Roman law. An inevitable conclusion must be that the St. Gall manuscript was being used in a center where Roman law was being taught. As we shall see the manuscript could not have been the product of a small provincial school[83]. These texts of Roman law provide yet another example of the main theme of this essay: how Justinian’s compilation began to shape the thinking of jurists who were working with non-Roman legal texts.

To turn now to the evidence for these generalizations. At the end of Causa two the text of Cod. 9.46.10 was added to the margin[84]. This would not be remarkable, except for that fact that the text of the Codex was provided with an interlinear gloss that can also be found only in early twelfth-century Codex manuscripts[85]. The combination of the two, the text of the Codex and the gloss, are evidence that the jurist who was responsible for adding this text to the manuscript must have been in a center where he had access to Roman law texts and to the commentaries on them.

Two more texts from the Codex were added to the margin of C.32 q.1 d.p.c.10. Gratian later incorporated the first into his dictum[86]. He did not include the second text, Cod. 9.9.5, in later recensions. Causa 32 treated the issue of adultery. In question one, Gratian focused on the issue whether and under which conditions a man could keep a wife who had committed adultery. Cod. 9.9.5 raised another point of law: was there a statute of limitations that should be placed on an accusation of adultery? A constitution of Severus Alexander (223 A.D.) established a time limit of fifteen years. A rubric in the margin of St. Gall declared: “Adultery cannot be prosecuted after 15 years”[87]. Although Severus Alexander’s constitution may have been persuasive to the jurist(s) who worked with the St. Gall manuscript, as far as I know, this rule of Roman law was never accepted by the canonists.

The authenticae that were added to the St. Gall manuscript provide more examples of the jurist’s keen knowledge of Roman law.[88] Excerpts from Justinian’s Novellae were added to Codex manuscripts during the first half of the twelfth century. These texts have been attributed to Irnerius, but his authorship is far from certain[89]. The importance of these authenticae has never been fully appreciated. Whoever excerpted and translated these texts from the Greek (there is evidence that more than one person was responsible) and added them to the margins Codex manuscripts accomplished a demanding and important project[90]. The Novellae of Justinian in their Greek and Latin versions were turgid pieces of prose[91]. An abbreviated Latin text of the Novellae, the Epitome Iuliani, had been published in the sixth century. It circulated widely in the West[92]. Strangely, when the jurists of the late eleventh or early twelfth decided to add excerpts from the Novellae to the margins of Codex manuscripts, they did not use the translations in the Epitome Iuliani; rather they crafted their own, shorter, translations. The need for shortness and clarity may be the main reason why Irnerius and/or his colleagues were compelled to retranslate and abbreviate Justinian’s Novellae even more than had been done for the Epitome Iuliani. Their primary purpose for creating these short texts was to update the Codex[93].

The authenticae that were added to the margins of the St. Gall manuscript copied the same format that we find in early Codex manuscripts: They are introduced by “C.N. (Constitutio nova)” and are provided with large paragraph markings at the beginning of each text. The authenticae in St. Gall raise points of law that Gratian did not consider in his hypothetical cases and, consequently, provide evidence that teachers who used Gratian’s Decretum turned to Roman law in the 1130’s for a deeper understanding of canonical problems.

In Causa 29 Gratian had posed a hypothetical in which a man of servile status married a free, noble woman. Although he treated various questions raised by the marriage, he did not address a very important point: could a slave marry without his master’s permission? In all three stages of the Decretum that we have that issue was not joined. The jurist who added the authentica to the top of the folio 171 implicitly raised the issue. The text stated[94]:

If a slave owner gave his female slave to a Christian in marriage and presented her as a free woman or if knowing that she were to be married remained silent, he lost his ownership of her. She became free. The same is true of a male slave.

Quite surprisingly none of the early canonists who glossed the Decretum cited this authentica.

Another authentica was added to the end of Causa 12 q.5 in St. Gall. Gratian asked the question which episcopal property bishops could pass on to their heirs. In an authentica that was added to the Codex, the jurists adopted Justinian’s rule that bishops could put only the property into their wills that they had acquired before they entered the episcopal office[95]. All of these examples establish that the St. Gall manuscript was being used by jurist(s) who knew Roman law and incorporated its norms into canon law when it was appropriate. It also is evidence that the St. Gall manuscript was not an abbreviation. No abbreviation would have been given such professional treatment and sophisticated up-dates.  The manuscript could have only been written and used in a major center for the teaching of law.  Finally, these Roman law additions must have been made to the St. Gall manuscript in the 1130’s – exactly the time in which Roman law norms began to exercise a strong gravitation pull on European legal systems.

One last example of an early influence of Bolognese jurisprudence on twelfth-century statutes. In the thirteenth century Accursius noted that Jacobus had inserted a “false authentica” into the Codex[96]. The authentica stated that if a defendant had confessed his crime outside the courtroom to friends or neighbors, his confession was considered valid and admissible in court[97]. Not only could the text or one similar to it not be found in Justinian’s Novella, but by the thirteenth century jurists understood that it violated basic rules of courtroom procedure. I have traced the text to the margins of three early twelfth-century manuscripts[98]. When Pisans began promulgating statutes and compiling customary legal collections for the city[99], they placed the gist of the authentica in their Constitutum usus[100]. There is, as far as I know, no other possible source for the Pisan text other than the authentica created by twelfth-century Bolognese jurist(s). It is somewhat surprising that this Roman text is placed in the customary law section of the compilation[101]. However, just as Roger’s legislation was shaped by Roman law, the Constitutum usus of Pisa also had Roman jurisprudential fingerprints on it.

Roger II’s Constitutions, Bulgarus, Gratian, and Pisan legislation are separate but joined pieces of evidence that the teaching of Roman law in Bologna began to exercise influence on secular and ecclesiastical law and on medieval institutions in the first half of the twelfth century. In the context of the recent debate about when Roman law began to be taught in Bologna, Roger’s Constitutions, Bulgarus, and Gratian provide substantial evidence that Bologna was producing jurists who had a highly developed and deep understanding of the entire corpus of Roman law by the 1130’s. This evidence argues strongly for a school that was not in its infancy[102]. There has been room for debate about when the teaching of Roman law became important in Bologna because we have very few Roman law manuscripts from the period ca. 1075-1130 that might provide certain evidence for teaching activity[103]. The glosses in those manuscripts that do exist are rarely signed; and when they are signed, it is difficult to attribute them to a certain jurist with confidence. The work of the Roman jurists, Irnerius or others, to incorporate the excerpts from the Novellae into the Codex is, however, another piece of firm evidence that teaching Roman law was in full swing before the 1130’s[104].

By ca. 1140 that had changed dramatically. The texts discussed in this essay demonstrate a significant penetration of Roman law into other legal systems in the first half of the twelfth century. Bulgarus’ letter to Haimeric, the addition of extensive Roman law texts from the Digest and the Code to the dicta of the last recension of Gratian’s Decretum ca. 1135-1140, the Pisa’s Constitutum usus and, finally, the Roman law additions to the recension of Gratian in St. Gall 673, converge to give us a vivid picture of the Ius commune’s “Big Bang”. It was the historical moment when a dead legal system’s jurisprudence, which had been confined primarily to the classroom – as far as we can tell from the extant evidence – exploded and began to exercise its gravitational pull on customary law, institutions, canon law, and the intellectual life of medieval society. It was the moment in which the jurisprudence of the Ius commune was born.

A crucial element of this story is the ineffable but undeniable fact that during the 1130’s the authority and jurisprudence of Roman law as found in the libri legales became generally recognized as a source, a model, and quarry from which the iurisperiti could influence and shape the laws and institutions of the ius proprium. One last point should be made. Many of the norms of Roman law that I have cited in this essay were not accepted by later jurists. Fines were not levied against bishops who hindered appeals. A statute of limitation was not placed on sexual crimes. Confessions were not held to be valid outside the courtroom. These facts should remind us that Roman law was a quarry for norms, but the jurists did not accept every block of marble pulled out of the pit. Principles of Roman law had a profound effect on canon law and other legal systems. However, as the jurists examined each Roman norm, they slowly reached consensus about which blocks from the quarry were well-shaped and flawless for building the norms that governed medieval society. They rejected those that were fell short. The standards that they used – reason, equity, justice – are eternal.