The early Middle Ages produced a series of law codes for the new “barbarian” kingdoms of Europe, which succeeded the western Roman Empire. These law codes were often inspired by the precedent and sometimes the content of Roman vulgar law as well as the customs of the respective peoples for whom they were written and the interests of their rulers.Footnote ¹ The making of law could often play a vital role in the stabilization of kingdoms, especially under new rulers. Early medieval secular lawmaking falls into three broad periods: the early royal laws of the Frankish, Burgundian, and Visigothic peoples in the fifth and sixth centuries; the interrelated composition of Lombard, south German, and perhaps also early Anglo-Saxon law in the seventh and eighth centuries;Footnote ² and the writing up of the last “ethnic” laws for peoples subject to Charlemagne's empire, such as Frisians and Saxons, in order to accommodate them into a multiethnic empire committed to the principle of personality of the law.Footnote ³ The subject of this article, the law of the Bavarians (Lex Baiuvariorum, hereafter abbreviated “Lb”), belongs to the second of these stages. However, scholars have never reached consensus as to the date of its composition nor where it was created. This has inhibited the use of the Lb for any but they most general discussion of Bavarian society.Footnote ⁴ This article will review the evidence for the Lb's date and place of composition, to suggest that we can plausibly identify them more precisely than has been done, and therefore argue that the distinctive features of this text can be tied to specific political needs.

The early medieval Bavarians occupied an area between the Alps and the Danube, with the Lech River as their western border and an open eastern frontier that reached from the Enns River toward the Slavic and Avar peoples of the middle Danube.Footnote ⁵ On their west, the Bavarians had the Alemanni, living roughly in modern Baden-Württemberg and northern Switzerland. Beyond the Alemanni were the Franks, ruling not only modern France but most of the Low Countries and parts of modern Germany, already the strongest single kingdom in Europe. The continent's other major kingdom, the Lombards, lay across the Alpine passes in Italy.Footnote ⁶

The Bavarians first coalesced as a group in the early sixth century, and were under Frankish influence from an early point, if not the very beginning.Footnote ⁷ By the end of the seventh century, however, they seem to have become de facto independent. At that time, Duke Theodo invited foreign ecclesiastics to help incorporate his people into Christendom, and visited the pope personally in 715 to advance this process.Footnote ⁸ When Theodo died soon thereafter, the duchy was divided among his sons. The last of these, Grimoald, was killed in a coup in 725, a conflict that brought renewed Frankish intervention to the duchy under Charles Martel.Footnote ⁹

When Grimoald's nephew Hucbert died in 735, there were no more heirs of Theodo available. Instead, probably with Frankish influence, a new duke arrived. Odilo was a member of a distant branch of the ruling family, the Agilolfings, from neighboring Alemannia.Footnote ¹⁰ Odilo led a coalition of Germanic and Slavic peoples to war against the Franks in 743, but was defeated. He ruled until his death in 748, to be succeeded by his 8-year-old son Tassilo III. Tassilo ruled in virtual independence until his cousin, Charlemagne, began to impose his authority on the Bavarians in the 780s, finally deposing Tassilo in 788 and incorporating his lands into the growing Frankish empire.Footnote ¹¹

The Lb was composed at some point in the midst of these political upheavals. It is a complex text that was compiled from numerous sources, but also contains original elements. It is composed of twenty-two titles, each with anywhere from one to thirty-two items, in addition to a prologue.Footnote ¹² The first eleven titles mainly constitute a wergild code, assigning the composition due for offenses against various classes of society.Footnote ¹³ Of these, titles I–III specify the rights of leading sectors of society: the church, the ducal household and officers, and a kind of nobility, the “genealogiae.” Therefore, they also sketch out a kind of political constitution. Titles IV–XI complete the wergild code with laws on offenses to freemen, freedmen, and slaves, followed by laws pertaining to marriage, women, theft, arson, and miscellaneous violence. The next section deals with economic relationships: property boundaries, pledges, trusts, and sales.Footnote ¹⁴ These are followed by two titles dealing with participants in conflict resolutions; first witnesses and then champions in trials-by-combat.Footnote ¹⁵ The final section, generally agreed to be a Carolingian supplement, contains various laws on animal property, including dogs, falcons, and bees.Footnote ¹⁶

Two main approaches have been taken to dating the Lb, each combining internal evidence with political and ecclesiastical context.Footnote ¹⁷ Curiously, both positions follow general lines laid out by Heinrich Brunner in the late nineteenth and early twentieth century. Brunner initially argued for a later dating, between 744 and 748.Footnote ¹⁸ Subsequently, however, believing that he could discern the text of a lost Merovingian royal decree in the superscription to title I of the Lb, he changed his view and argued for an earlier composition instead.Footnote ¹⁹ Most adherents of the early dating rely on the prologue of the law. This claims that the original legislation was issued by a Frankish king Theuderic (I, r. 511–533/4), updated by Childebert (I, of Austrasia 575–596) and completed by Chlothar (II, of Austrasia 613–622). Finally, Dagobert I (d. 639) is said to have consulted four noblemen to improve and reissue the whole, not only for the Bavarians but all the peoples under his rule. Originally skeptical of the prologue, Brunner was converted by a superscription to the text, which pronounces that the law was issued for peoples living in the “Kingdom of the Merovingians,” the dynasty to which Theuderic, Childebert, Chlothar and Dagobert belonged.Footnote ²⁰

This early dating, however, has failed to win a consensus of scholarly opinion. For one thing, the prologue's description of successive waves of legislation is difficult to reconcile with Bavarian history. During Theuderic I's reign, the region was not even under Frankish rule, but was governed by the Ostrogoths from Italy. Also, the Merovingians' authentic documents never refer to their realm as the “Kingdom of the Merovingians,” but simply as the “Kingdom of the Franks.”Footnote ²¹ Furthermore, there is no manuscript evidence for earlier versions of the Lb than the one known in the eighth century; most other early medieval law codes that were revised and updated in this way, such as the Alemannic Law or the Franks' own Salic Law, can still be found in their sixth- or seventh-century versions in some manuscripts.Footnote ²²

This prologue had a life of its own in the manuscripts and was not inextricably bound to the rest of the Lb. It does not seem to have been composed specifically to introduce Bavarian law; the Bavarians are simply listed with the Franks and Alemanni as people for whom King Theuderic and his successors supposedly produced law, “for each of the peoples under his dominion, according to their own custom.”Footnote ²³ It thus works equally well as a prologue for Alemannic law or for Frankish law, and in fact its manuscript transmission reflects this. For example, the prologue appears in manuscript K21 (first third of the ninth century) of Rosamond McKitterick's handlist of Carolingian legal manuscripts, a codex that includes the Lex Salica, Lex Ripuaria, and Lex Alamannorum—that is, the laws of the other peoples the prologue refers to—but not the Lb. K44 (tenth or eleventh century) includes the prologue with Lombard and Salic law, and K72 (middle of the ninth century) with Salic, Ripuarian, Burgundian, and Alemannic law as well as a few other texts, but, again, not the Lb.Footnote ²⁴ It is, therefore, possible that the prologue did not originally refer to Bavarian law in particular. Perhaps, prior to the Lb's creation, it was assumed that Frankish or Alemannic law applied to their Bavarian cousins as well.

There have been two recent efforts to defend the older dating for the Lb. When the Lb makes use of Visigothic law, it draws on the Lex Eurici from before 507 rather than the Liber Iudiciorum of 654, which is far better represented in surviving manuscripts. Hermann Nehlsen has argued that the use of the Lex Eurici must mean that the law was compiled prior to the Liber Iudiciorum's promulgation.Footnote ²⁵ He suggests that copies of the Lex Eurici must have passed to the Lombard kingdom and then to the Lb's compilers, who may even have worked in Italy. Taking a different tack, Wilhelm Störmer argued that the Bavarian legislation of the 770s shows a distinct noble class that was unknown to the Lb; in Störmer's view, the single generation from the 740s to the 770s is too brief a time for such a deep-reaching social change.Footnote ²⁶ By taking these different approaches, both Störmer and Nehlsen argue that the Lb's prologue must be essentially accurate, and that the law or a core of it was composed in the seventh century, probably not later than the reign of Dagobert I (d. 639).

Neither of these arguments is compelling, however. Nehlsen envisions busy scriptoria of Lombard jurists collecting and copying laws from all over Europe. The Lombard laws themselves do show considerable sophistication and we might imagine the Lombard court having a relatively cosmopolitan group of scribes and legal officials, but there is no manuscript evidence to suggest that the Lex Eurici was actually transmitted as he suggests. Nehlsen can only point to the well-known royal marriage connections between Lombards and Bavarians to depict the possibility that legal manuscripts might have passed between the two areas already in the early seventh century.Footnote ²⁷ The Lex Eurici is known only from a palimpsest in a late Merovingian (seventh or early eighth century) manuscript, possibly from Corbie; the original was written in uncial letters in the sixth century, apparently in southern Gaul.Footnote ²⁸ Therefore, the only manuscript evidence we have—which is not much—would point to the Frankish kingdom as the route of transmission, not the Lombard.Footnote ²⁹ Lombard law never cites the Lex Eurici. A further problem with Nehlsen's argument is the assumption that the compiler of the Lb would have availed himself of the most up-to-date Visigothic law available, so that the Lex Eurici could not have been used after the appearance of the Liber Iudiciorum in 654. However, there is no reason to suppose that the compiler thought in terms of laws' relevance or obsolescence in their homelands when making his collection, nor that he was even aware that different versions of Visigothic law existed. His goal was to create a law for the Bavarians, and he borrowed whatever seemed useful in the law books he had access to. In other words, the compiler presumably did not go looking for up-to-date Visigothic law; he looked for law on particular topics (mostly commercial matters such as pledges and sales, in this case) and found what he needed in a handy copy of the Lex Eurici. How or where he found it, we do not know. We do not know how widely available the Lex Eurici would have been outside of the Visigothic kingdom in the first century or thereabouts after its publication; the earliest surviving manuscripts of the Liber Iudiciorum date to the eighth century, a century or after their original promulgation, or thereabouts, and the bulk of them are products of Carolingian efforts to compile “national” law codes for the peoples of their empire.Footnote ³⁰ Therefore, Nehlsen's argument that the promulgation of the Liber Iudiciorum in 654 means that the Lb's compiler must have worked before that date does not stand; it seems unlikely that scribes elsewhere in Europe would even have had access to it until much later.Footnote ³¹

Störmer's argument, likewise, rests on the premise that significant social change (the emergence of a defined legal nobility) cannot happen in a single generation. However, one can hardly say that Bavaria in the 740s did not have a class of powerful landholders who dominated the duchy's economic and political life by virtue of their inherited wealth and social connections, even if the Lb did not grant them a distinct legal status or call them “nobles”; we meet them all the time in eighth-century charters, as Störmer's own prosopographical studies have amply shown.Footnote ³² The sudden appearance of “nobles” in the legislation of Tassilo's assemblies of the 770s does not necessarily reflect the appearance of new social realities, but rather the usage of new vocabulary and conceptions for existing social groups, perhaps because of the influence of Frankish terminology.Footnote ³³

It should be noted that there have also been some efforts to “split the difference” and treat the Lb as the product of gradual evolution or revision, first promulgated in the sixth or seventh centuries but only taking its “final form” as it survives in the manuscripts in the eighth century. Such an approach was taken by Joachim Jahn, for example, who first presented the Lb as a document of the duchy's sixth-century origins but then also used it to explore the duchy's social and political structure under Duke Odilo, nearly 200 years later.Footnote ³⁴ Such an approach has the virtue of compromise, and, since the nineteenth century, the scholarly community has been accustomed to texts that evolve and develop over generations. However, the evidence for such evolution is lacking in the case of the Lb. No scholar has ever been able to excavate an “earlier layer” of the Lb's text, and the clear exemplar of the law as it exists is the Lex Alamannorum (hereafter La) as redacted under Duke Lantfrid, who died in 730. It is far more straightforward to see the text as a single work that took the La as a template and proceeded to revise and supplement it, than to argue that there was an earlier independent version of the Lb that was then somehow overhauled in imitation of Lantfrid's version of the La. Whether the La itself, particularly the seventh-century Pactus Alamannorum, once counted as law for the Bavarians, before they had a law code of their own, is an interesting question, but it is a separate one that need not detain us here.Footnote ³⁵

Most scholars have therefore followed some variation on Brunner's earlier approach, which would date the text's composition between 744 and 748.Footnote ³⁶ Tassilo III succeeded to the ducal office at eight years of age in the latter year, and only came of age in the mid-750s, by which time Bavarian documents already indicate their conformity to Bavarian law.Footnote ³⁷ This makes composition under his auspices unlikely. Brunner chose 744 as a terminus post quem, based on Duke Odilo's unsuccessful war with the Frankish palace mayors, Pippin III and Carloman, in 743. As the Lb often emphasizes the dependence of the Bavarian duke on the Frankish king's authority, it is argued that it could not have been issued prior to Odilo's defeat in 743.

Unfortunately, this argument from ducal politics no longer makes sense in the light of contemporary scholarship on the Agilolfings. Peter Landau, in one of the more recent studies of the Lb, has already suggested that the choice of 743 as a terminus post quem exaggerates Odilo's aspirations to independence from the Franks.Footnote ³⁸ It is not within the scope of this study to trace the political narrative afresh from the primary sources, but it can be summarized as follows: from the fall of Duke Grimoald in 725, the Franks played a leading role in buttressing ducal authority.Footnote ³⁹ Odilo, even before 743, relied heavily on Frankish support; later sources (albeit Frankish ones) claimed that he only obtained the ducal office in the first place through Charles Martel's assistance, and that he took refuge at the Frankish court when a rebellion by his own nobles drove him into exile, probably shortly before Charles' death in 741.Footnote ⁴⁰ He married one of Charles' daughters, Chiltrud, although apparently without the approval of her half-brothers Pippin and Carloman. The war of 743, in light of these events, was probably not a struggle for Bavarian independence from Frankish control. Rather, Odilo was intervening in the struggle for the succession between Pippin and Carlomann, on one side, and their half-brother (Odilo's brother-in-law) Grifo, on the other.Footnote ⁴¹ There is therefore no reason to assume that the Lb was compiled after 743.

On the basis of these contextual arguments, Peter Landau has already argued for a composition date between 737 and 743. Internal evidence from the text also makes a dating after 743 unlikely. The first title of the Lb addresses church law. A very unusual and controversial clause in this title, I.10, specifies the procedure for trying a bishop in the duchy. The text is wholly original and not based on any prior legal collection of the early Middle Ages; therefore, it was presumably custom written for the Bavarians.

In some of the earlier manuscripts of the Lb, I.10 carries the superscription “De solis episcopis et illorum interfectione,” meaning literally, “On bishops solis and the killing of them.” This raises the question of what the writer meant by “solis,” and whether it was part of the original text or a later addition. “Solus” normally means “single, sole, individual.” That it would refer to one bishop out of many being killed (“On individual bishops…”) seems unlikely, as there would be no reason to distinguish the killing of one bishop from the killing of several. It might mean that the law envisions times when only a single bishop was present in the duchy (“On unique/sole bishops…”), perhaps carrying his authority by association with the ducal court rather than a canonical see. A third meaning would take solis adverbially, so that the clause reads, “Concerning bishops only, and the killing of them.”Footnote ⁴²

The meaning of this heading cannot be settled by grammatical arguments alone. Taking this heading on its own, the adverbial translation makes the most sense. The content of the law, however, is worth consideration. It describes the bishop as chosen by the king or people, without reference to other bishops of the province consecrating him, and calls him “highest pontiff” (summa [sic] pontifex). This language is found elsewhere in Bavaria, in the hagiographies of Bavarian saints from before Odilo's time. The Life of St. Corbinian (d. ca. 728), written in 770 or thereabouts, describes a miracle in which an eagle delivers a fish to the feet of the “highest pontiff” (summi pontifici); In the Life or Passion of St. Emmeram, the same author, Bishop Arbeo of Freising, stresses the term by having Duke Theodo ask Emmeram to be his people's “pontiff” twice in a single sentence.Footnote ⁴³ In both of these texts, particularly Emmeram's, the bishop in question is treated as unique in the area; unlike Frankish episcopal saints' lives from the same period, we never see them interacting with colleagues in the episcopal office. Any medieval Christian writer might refer to a saintly bishop as pontifex or even summus pontifex; therefore, the term is not unique to these Bavarian hagiographies. We can, however, at least be sure that the usage was consistent with the period of singular, foreign bishops presiding over the duchy from court.Footnote ⁴⁴

Furthermore, if the bishop was accused of a crime, he was to be judged “according to the canons” by the duke (or king), not by his fellow bishops, as canon law would normally have required.Footnote ⁴⁵ The overall tone of the text, then, carries the impression that the bishop is a unique individual in the dukes' territory, not one in a class of several individuals. Lb I.10 cannot be taken to prove absolutely that there was only one individual in Bavaria at a time with an episcopal consecration (the episcopates of Emmeram and Corbinian may have overlapped with others', such as Rupert's or Erhard's), but it is easy to associate with a kind of “court bishop” operating in the duchy at ducal invitation, of the kind seen in the early Bavarian saints' lives.Footnote ⁴⁶

Was the problematic “solis” part of the original text? The heading does not appear throughout the Lb's manuscript tradition, only in a few manuscripts, and some of those omit “solis.” It appears in the index capitulorum of the law's oldest manuscript, from 800 or thereabouts; some subsequent scribes copied it over as a chapter title to the main text.Footnote ⁴⁷ Given the ambiguity of the language, it would seem that the solis was more likely to have been part of the original text, as the lectio difficilior. That is, it seems more likely that the term made sense to an earlier scribe, but came to be dropped in the manuscript tradition by copyists who found it ambiguous, meaningless, or obsolete, than that a later copyist would have thought that solis should be added to clarify the meaning. In short, it seems possible that the title with “solis” is original to the index of the Lb and that it envisions a duchy with a bishop who was unique, if not exactly “alone,” in the province.

I would suggest, therefore, that the Lb does not envision a situation in which multiple bishops are available in the duchy to adjudicate the misdeeds of a fellow bishop; not so much because the heading of Lb I.10 might be translated “on sole bishops,” but because the content of the law itself seems to point in this direction. This suggests that the Lb would have been issued at the very beginning of Odilo's reign as duke, between 736 and 738. The year 738 was a terminus ante quem because in 738/9, St. Boniface, the papal legate, church reformer, and sometime missionary, visited Bavaria at Odilo's invitation and erected four episcopal dioceses.Footnote ⁴⁸ These new episcopal sees were based on existing religious communities that had, at least occasionally, housed “wandering bishops” since Theodo's time. Boniface's work created four stable, canonical bishoprics instead of the improvised provisions of the prior situation. It is, therefore, unlikely that the Lb was composed after 738, because at no point after that were there ever fewer than four bishops present, and it is hard to imagine Boniface (who had the Dionysiana canon law collection) consenting to its provision for ducal jurisdiction over bishops. Boniface complained often and loudly about the uncanonical behavior of his Frankish colleagues in the episcopate: despite his close relationship with Charles Martel and then later with his sons, he never suggested that these secular rulers should have enforced canon law on them. In fact, as Odilo was apparently in communication with the pope and Boniface about the organization plan in the year 738 itself, he probably would not have made himself the final authority over churchmen in that year either. This pushes the likely date of composition back to 737 at the latest.

If references to the Frankish king indicate that a Merovingian was reigning at the time the Lb was promulgated, this dating would be reinforced. Charles Martel had maintained a puppet monarch, Theuderic IV, in the earlier years of his tenure as mayor of the palace, but from Theuderic's death in 737 he did without the figurehead. His sons, Pippin and Carloman, only had a new king enthroned in 743. At one point Bruno Krusch thought that the attribution of the Lb to a “Theuderic” in the prologue is an allusion to Theuderic IV, in which case it was written no later than 737.Footnote ⁴⁹ This is not a strong argument, however; no one would have presumed after 737 that the interregnum would be permanent, or even as long as it turned out to be, in which case new laws would still have referred to the king's authority although no actual king was reigning at the time. In any case, Charles Martel regularly issued decrees in his own name without putting the king's name to them. It is unlikely that he would have used the king's name here, still less that he would have had a prologue falsified that used the long-dead Theuderic I to allude to his own puppet, Theuderic IV.Footnote ⁵⁰ Nevertheless, we can at least say that the chronology is suggestive.

On the basis of this reassessment of the eighth-century political context, then, we can say that the old argument for dating the Lb between 744 and 748 is highly unlikely. Assuming that the law was promulgated by Odilo, the Lb was probably compiled between 735/6 (the year of his accession to the duchy) and 737, or at least no later than 738.

The possibility remains that the law was promulgated by Odilo's predecessor, Hucbert.Footnote ⁵¹ However, attributing the text to Hucbert then raises questions about the place of composition. Bavaria in Hucbert's time had, as mentioned previously, no standing episcopal sees and only a handful of monasteries, which do not seem to have been large or well equipped with libraries and scriptoria. Literacy was not unknown—Duke Theodo had a chancellor near 700—but not on the scale one would have needed to compile a law code like the Lb.Footnote ⁵² The evidence for Bavarian scriptoria only begins from the mid-eighth century, suggesting that the episcopal foundations of 739 helped encourage regular scribal activity. None of the known manuscripts produced in Agilolfing scriptoria contain secular law. Even in the ninth century, the only secular law to appear in Bavarian manuscripts is the Lb itself, not its sources.Footnote ⁵³

The size of library that would have been needed can be seen from a review of the sources for the Lb's various titles. To judge by the attributions in von Schwind's edition, the compiler of the Lb drew on several earlier “Germanic” law codes. References to and influences from Visigothic, Lombard, Frankish, Burgundian, and Alemannic law have all been identified or posited, many of them near-verbatim quotations. However, derivations from these older law codes are not distributed evenly through the Lb, and their influence is far from equal. Some of the similarities are only of subject rather than specific language; these are as likely to come from a common heritage of Roman vulgar law as from the use of specific post-Roman codes. Different sources feature differently in the various sections of the Lb. Therefore, the bulk of the wergild code in titles I to XI comes almost verbatim from the Law of the Alemanni, Odilo's own people. Within this section, however, the treatment of the ducal court and its officers (title II) makes use of the seventh-century Edict of Rothair, from Lombard Italy, in addition to Alemannic law.Footnote ⁵⁴ Visigothic law, particularly the Codex Euricianus, is another important source for the Lb. The titles on pledges, sales, and property boundaries are heavily Visigothic in character.Footnote ⁵⁵ Other sections, meanwhile, are wholly original, including the titles on witnesses and champions, as well as various individual laws (I.10, mentioned previously, title III, on the noble kin groups called “genealogiae,” and title IX, on arson).

Manuscripts that contain substantial compilations of law texts from different areas are quite common in the early Middle Ages. Rosamond McKitterick compiled an impressive list of them in 1989, which might create the impression that the Lb's compilers could easily have had all this material ready to hand, wherever they worked. However, a closer look at McKitterick's list shows that it would not have been so simple. First of all, the earliest of the legal manuscripts McKitterick mentions dates from 770, and the vast bulk of them from the ninth century. This suggests that the practice of legal compilation was a product of the Carolingian Renaissance and the practicalities of governing a multiethnic empire where the personality of law was a recognized principle; factors that were not relevant in Odilo's time. McKitterick argues that such collections flourished precisely because they were needed for practical use under Charlemagne and his heirs. Also, when McKitterick's list is examined for manuscripts that contain Alemannic, Frankish, Lombard, and Visigothic law—the main sources for the Lb—it turns out that not one Carolingian manuscript contains all of them together.Footnote ⁵⁶ Those that come close, only missing one source (usually either the Visigothic or Lombard elements), are from the ninth century or later and usually include the Lb itself. This makes it clear that the compiler of the Lb would not have worked from a single manuscript of collected legal texts, even if anyone before the Carolingians were interested in compiling such; he must have had a library. How common were libraries with all the requisite material in the eighth century?

The Lb would have required a substantial library and a learned scribe or group of scribes to bring together all of these various precedents. The influence from other legal texts was the product of actual books, not simply the familiarity of individuals with those cultures: both the Lombard and Visigothic texts are quoted in versions that were obsolete by the eighth century.Footnote ⁵⁷ The predominance of the Alemannic Law at the heart of the Lb, however, gives us our best clue to its origins. The composer of the Lb worked from an Alemannic core, supplementing it with material from other law codes and adding original material that presumably reflected the particularity of the Bavarian situation or perhaps even Bavarian custom (the practice of tugging the ears of witnesses, for example, seems unique to the area, mentioned in both the Lb and Bavarian charters). As Odilo himself came from Alemannia—the version of Alemannic law used in the Lb is that produced by his older brother, Duke Lantfrid, between 724 and 730Footnote ⁵⁸—the obvious conclusion is that he, and not his thoroughly Bavarian predecessor Hucbert, was the moving force behind the Lb's compilation.

The exact location of the Lb's composition cannot be determined with certainty because our evidence for pre-Carolingian book collections and scriptoria is so sketchy. But Carolingian evidence can give us some hints. From the later eighth century onward, Charlemagne's court patronized a greater quantity and quality of text copying and correction; the “Carolingian Renaissance.”Footnote ⁵⁹ As part of this movement, the Alemannic monasteries of St. Gall and Reichenau wrote up catalogs of their libraries early in the ninth century.Footnote ⁶⁰ The list from Reichenau was written in 820/1, and includes seven codices of secular law and two of canon law, in a total collection of 415 items. Three of the secular manuscripts were monographs, containing a single legal text each: one of Frankish (Salic) law, one of Alemannic law, and one of Lombard law. The other four manuscripts collected multiple legal texts; of these, the first contained Roman law (from the Theodosiana) in addition to Ripuarian, Salic, and Alemannic laws, and Carolingian capitularies. The three others each combined Alemannic law with other texts: the Ripuaria and capitularies, the Salica, and capitularies, respectively. Of the sources for the Lb, only Visigothic law is absent from this list. St. Gall's booklist was compiled near 850/60, with items added as late as 880, and lists 294 codices containing 426 texts among them. In this collection, in addition to three volumes of canon law, there were five codices of capitularies and only one book of pre-Carolingian secular law. This contained three Roman texts (the Theodosiana, Ermogeniana, and Lex Papiani), Frankish law (the list does not specify if this was the Salica or Ripuaria) and Alemannic law.

These lists were made generations after Odilo's time, and the books mentioned in them (or their exemplars) might only have come to the Alemannic monasteries in the late eighth or early ninth centuries. However, they do shed some light on the problem. Even in the full swing of the Carolingian renaissance, a center of literary culture as massive as St. Gall did not have a library that could have supported the compilation of a text like the Lb. Only one manuscript there contained any of the needed texts, and was missing any representative of either Visigothic or Lombard law. Reichenau seems more promising, especially if we may imagine its apparently strong interest in secular law going back to the eighth century. The monks of Carolingian Reichenau had several copies of Alemannic law, several copies of both Salic and Ripuarian Frankish law as well, and, perhaps most importantly for the purposes of this suggestion, a copy of Lombard law.

Reichenau had been founded by the peregrinus St. Pirmin only in 724.Footnote ⁶¹ It was a representative of Frankish influence in Alemannia—Charles Martel participated in the foundation—which embroiled it in various regional struggles. Reichenau, however, also may have had ties to Odilo. In 742, he founded a monastery at Niederaltaich with monks from Alemannia.Footnote ⁶² The new house was consecrated not by the bishop of Regensburg, in whose diocese it lay, but by Heddo, Bishop of Strasbourg, a former abbot of Reichenau. It is, therefore, a fair guess that the monks came from Reichenau as well. Odilo also supported other Pirminian monasteries and was ultimately buried in one of them, Gengenbach.Footnote ⁶³ As a church institution connected both to the Frankish court and to Odilo, Reichenau makes sense as being a place where Odilo might have looked for help compiling the laws to support his reign.

None of these secular law collections from Reichenau was used in the composition of the Lex Alamannorum issued by Duke Lantfrid, Odilo's brother, between 724 and 730. This may mean simply that Lantfrid did not turn to Reichenau as Odilo would do; alternatively, it could mean that the composition of the La stimulated Reichenau's interest in secular law, leading to the acquisition of manuscripts, perhaps in the early 730s, which ended up helping Odilo's project. Any attempt to place the Lb's composition is necessarily speculative, and it is unlikely that we will ever determine the location with any certainty. However, given the absence of evidence for active Bavarian scriptoria before the second half of the eighth century, it seems reasonable that we should look abroad; Odilo's origins make it likely that we should look to Alemannia; and the later Carolingian evidence for Alemannian monasteries implies that of the two largest houses to exist in Odilo's generation, Reichenau is more likely to have had the requisite source texts than St. Gall.

The association of the Lb with Reichenau has also been made recently by Clausdieter Schott, as a byproduct of Schott's work on the La.Footnote ⁶⁴ Schott argues that the final, “Lantfridana” version of the La was in fact an ecclesiastical forgery, principally on the argument that its demands regarding church property were more extreme than anything Lantfrid is likely to have accepted. Judging by similarities between the phrasing of some of the La's clauses and charters from Reichenau, and given that Lantfrid was memorialized favorably there, Schott believes that monastery to be the most likely culprit behind the forgery. In Schott's view, the La would have been forged soon after the duke's death in 730 (not between 724 and 730 as generally thought). The consequence Schott draws for the Lb is that Konrad Beyerle's idea was correct, that it was composed at Niederaltaich by monks from Reichenau who brought a copy of the forgery with them, but that it was a private project of the monks there rather than an initiative of the duke's. Schott suggests that the Lb was also “published” after Odilo's death, perhaps at the Synod of Ascheim when the Bavarian bishops tried to dominate the young Duke Tassilo.

However, this argument provokes several doubts. Schott is certainly correct in asserting that the La, and the Lb for that matter, are generous in their protection of church interests and that they bear the marks of composition in an ecclesiastical setting. However, this does not demonstrate that no duke would have approved of them; as many early medieval princes knew, secure church property under a ruler's protection could be an excellent resource to help tame recalcitrant aristocrats. There is also the question of the nobles themselves; one wonders how they were persuaded to yield, with no protest that we can discern, to the sudden presentation of a law code, under the pretext of being “good old law,” which they had not heard of 5 years earlier, and which was supposedly so inimical to their interests. Third, if these are ecclesiastical forgeries, they are unique. No church institution managed to forge an entire law code before this, and none did so afterwards either. It seems a very labor-intensive process to generate an entire code for the sake of inserting a handful of clauses protecting church property. Forgers' interests could be served much more simply and credibly by a false charter of donation or immunity. It seems likely, then, that the suspicion of “forgery” in this case is not justified. It is much more probable that the dukes in question—Lantfrid and Odilo—knew and supported what their scribes were working on. Without granting Schott's claims of forgery, however, it appears that his observations are consistent with what has been proposed here: that the La and Lb were composed in close association with each other, by monks who were probably working from Reichenau in the 720s and 730s.

To sum up the argument so far, it appears that the Lb was probably produced 10 years earlier than the conventional view, or thereabouts, between 736 and 738 rather than between 744 and 748, and in Alemannia (perhaps by scribes at or associated with Reichenau) rather than in Bavaria. Is this adjustment merely scholarly pedantry, or does it carry larger implications for our understanding of the role and use of law in early medieval society?

An earlier dating and an origin in Reichenau suggest that the Lb is less a product of Bavarian society itself than a political tool, designed to support the change in ducal dynasty. If the Lb was compiled in 736 or 737, it must have been one of the first projects that Odilo undertook upon becoming duke. I would argue that the text shows Odilo using legislation to gain control of his new duchy as an outsider, in the absence of military force to assure his authority.

Frankish chroniclers would later claim that Odilo had obtained the duchy through the good graces of Charles Martel, and although these are hardly disinterested sources—they needed to justify Charlemagne's deposition of Odilo's son in 788—there is no reason to discard the claim.Footnote ⁶⁵ In any case, we have no better information to explain how the younger brother of the duke of Alemannia was chosen to become duke of Bavaria. Charles Martel had led an army to Bavaria in 725, and apparently a second time in 728, interventions which helped assure Hucbert's position as duke.Footnote ⁶⁶ However, no Frankish campaign in Bavaria is recorded for 735 or 736; the next time Frankish armies ventured into the region was in 743, when Odilo gathered an army in support of Grifo. It would appear that Odilo did not arrive in his new duchy at the head of a column of troops, at least not that any chronicler cared to record. The threat of a renewed Frankish incursion doubtless underlay Charles Martel's role in the process, but that role was presumably one of diplomacy and influence. Charles was in a good position to exercise influence through personal connections. When he returned from his 725 expedition, he brought Swanahild (or Sunnichild) with him, a daughter of the Bavarian Agilolfings.Footnote ⁶⁷ Thus, Odilo came into the ducal office through the influence of the Frankish court, but any compulsion from his backers was only potential. He had to have a means to gain acceptance as the Bavarians' leader without overawing them with a military demonstration. The Bavarian aristocracy, like most magnates in the early Middle Ages, was not a group whose support a ruler could take for granted; Grimoald was killed in a coup of 725, Odilo was forced into exile in 740/1, and his son Tassilo was handed over to Charlemagne by the testimony of some of his own disgruntled nobles in 788.

The Law of the Bavarians was the key to attaining this legitimacy. Through it, Odilo was able to generate support in key sectors of Bavarian society, which could secure his position. There were three groups who stand out as the beneficiaries of Odilo's legislation. The first group to benefit was the Christian Church, to whom the Lb gave formal protection, but whose protection was also made dependent upon the duke himself. The second was servants of the ducal household, whom the Lb made a privileged group within Bavarian society; thus, proximity to the duke himself could become the basis for individuals' political and social advancement. The third group was aristocratic kinship networks, which the Lb calls genealogiae. Five such kin-groups were named in the text, and the law's provisions for members of these groups lifted them above the run of other free landholding Bavarians, beginning the process of creating a true nobility in the region.Footnote ⁶⁸

The first title of the Lb consists of laws protecting the church's personnel and property. There are ten original sections to this title, plus three more that are generally believed to be later additions.Footnote ⁶⁹ Sections 1–6 concern the protection of the church's property, including unfree persons and protection against arson as well as its right to receive and hold real estate and movables. Section 7 guarantees the right of sanctuary in churches, and sections 8–10 set the compensation rates for members of the clergy against personal assaults. The later additions, sections 11–13, cover miscellaneous matters: nuns removed from their convents for marriage, a canonical prohibition of clerical cohabitation with women, and a detailed guide to the disposition of church estates and their revenues.

Section 1 details the procedure to be used in giving property to the church. Its main provisions are as follows. Any free Bavarian can give property to the church so long as he has already divided the property to give his sons their share; no one can prevent his doing so. Such gifts are to be made by charter, with at least six witnesses, on the altar of the church itself. Finally, such gifts are irrevocable; the legal guardian of the church (its defensor) can grant the property out in benefice but not give it away, and any disputes must be handled in the presence of the bishop.Footnote ⁷⁰ The stipulation that a would-be donor must first provide shares for his heirs is an important amendment to the Alemannic model, and would be the occasion for numerous lawsuits over church property in the coming generations.Footnote ⁷¹ Sections 2 and 3 indicated the composition owed for seizure of church property; double for “unjust seizure” (presumably the kind of property disputes forbidden in section 1), ninefold for actual theft and twenty-seven-fold for theft of liturgical items. The chalice, paten, and altar cloth are specified as examples of the latter, items that were probably the most valuable of a church's movable possessions and also sanctified by their role in the Eucharist. Unfree persons are likewise treated as church property to be protected in the subsequent two sections. Persuading an unfree person to flee results in a fine of 15 solidi, plus the restoration of the refugee, whereas killing one requires double the compensation normally owed for the unfree.Footnote ⁷² The penalties against arson set in section 1.6 are consistent with title X, a generally original title which shows a strong concern with this crime.Footnote ⁷³

These property laws, taken together, indicate a strong commitment on the part of the legislator to secure the position of the church as a legally privileged estate and a major landholder in Bavarian society. Most of them have precedents in Alemannic law, but are not common to other early medieval legal texts. The Salic Law, for example, hardly notices the clergy or the church as a distinct group at all prior to its Carolingian rescensions, which postdate the Lb in any case.Footnote ⁷⁴

The laws in question also seem to have been anticipatory and prescriptive, rather than descriptive of any customary legal regime for dealing with the church. There were certainly churches and clergy in Bavaria before Odilo's arrival, and even some bishops. However, whether these earlier bishops exercised rights over church personnel and property as envisioned here is unclear; Theodo is supposed to have granted Rupert a broad swath of territory around Salzburg, but the other pre-Bonifatian bishops do not appear in the sources ordaining or disciplining clergy or consecrating churches and defending their property. The requirement that donations to churches be made by written charter is another case in point. The oldest surviving charter from the duchy dates from 744, and that not in the original.Footnote ⁷⁵ Property listings from the bishopric of Salzburg include items going back to Theodo's reign at the beginning of the century, and even mention Theodo's chancellor, showing that written documents were not unheard of.Footnote ⁷⁶ But these early donations to Salzburg were all made by the dukes themselves. This suggests that only ducal grants were recorded in writing prior to Odilo's time, not “private” grants by free Bavarians as the Lb describes. If so, then Odilo, by issuing the Lb, both extended to private Bavarian freemen a mode of recording grants previously customary only for the ducal household, and also equipped the duchy's churches with a new form of legal protection of their property. More or less monastic communities had already existed at Salzburg, Freising, Regensburg, and possibly Passau; but the years from the 740s on would see an explosion in the founding of new monasteries and churches following the elevation of these older houses to episcopal sees.Footnote ⁷⁷

The impression that Odilo was trying to reinforce the church's position in society is strengthened by the laws protecting church personnel. Members of minor clerical orders receive double compensation for any injuries that they would have received on the basis of their family status, as do monks, whereas priests and deacons get triple compensation. It is worth noting that these clauses do not make the clergy into an entirely separate social order with its own scale of compensations; rather, their rights are still pegged to their birth status. However, Odilo has distinguished the clergy from the laity more than he would have found in his models: the Alemannic law allowed triple compensation only to priests; deacons received double, and lower clergy the same compensation as other family members. The Alemannic law had set the compositions for outright killings of clergy higher, however: 600 solidi for priests and 300 for deacons, to the Lb's 300 and 200, respectively (both laws had set a free man's blood price at 160 solidi).Footnote ⁷⁸ The difference in value may reflect economic differences between the two areas, but this is hard to be certain of. The Lb presents the clergy as highly privileged members of their families, standing over and above their relations in the court of law. Seen from another perspective, this law could have given families incentive to place members in the clerical orders in order to gain stronger protections for them.

Law I.10, concerning the killing of a bishop, has been referred to already, particularly that it seems to envision that only one bishop would be present in the duchy rather than the four who were consecrated by Boniface in 739. Two features of this unique law stand out. One is the dramatic way in which compensation for the killing is determined; the offender must pay the weight in gold of a leaden garment fitted to the bishop's body, or the equivalent value in property, or sell himself into slavery if he does not have enough to make up this princely sum. This seems a highly dramatic, rhetorical way to make a point about the inviolability of bishops. It is unlikely that even the wealthiest aristocrats possessed so much gold.Footnote ⁷⁹ The other striking feature of this law, again already mentioned briefly, is that it calls for the king or duke to try the case if the bishop is accused of a crime. This is clearly to head off private revenge killings in such cases, again stressing that bishops may not be entangled in feuds with secular persons. It also, however, makes the bishop legally dependent on the secular ruler as his judge and protector.

The unusual nature of Lb I.10 has led to much discussion of its relationship to an altogether different text from Agilolfing Bavaria, the Life or Passion of St. Emmeram, written between 770 and 772.Footnote ⁸⁰ According to this hagiography, set in Theodo's reign sometime between 680 and 715, Bishop Emmeram was falsely accused of having impregnated the duke's daughter. He set off for Rome to plead his innocence, but the duke's son Lantfrid overtook him and killed him, after a series of mutilations that the text dwells on at distasteful length. It is possible that Emmeram's story was well known before being written up and that Lb I.10 is a deliberate effort to prevent such things happening again. However, the hagiographer's sources for the story are unclear; it is even possible that he embroidered a vague tradition of Emmeram's death with details inspired by Lb I.10.Footnote ⁸¹ Whatever the case, it is likely that bishops were not exempt from violence prior to Odilo's time; another bishop's hagiography, the Life of St. Corbinian, also by Arbeo, depicts its hero narrowly escaping an assassination plot hatched at Duke Grimoald's court in the 720s.Footnote ⁸²

In issuing the Lb, then, Odilo made a sustained effort, not only to win over ecclesiastical support for his reign, but even to build the church up into a powerful social institution whose support would be politically meaningful. He made significant new concessions to encourage the accumulation of church property, and to protect that property once gained, and he offered strong legal protections to encourage the recruitment of clergy and to enable them to stand above the conflicts of Bavarian society; provided that they also accepted the authority of the duke himself to judge their highest representative, the bishop.

The next sphere to which the Lb turns is the ducal administration itself, in title II. Here also Odilo's legislation builds on Alemannic precedents, and many of title II's clauses also have connections to Lombard law.Footnote ⁸³ This section deals with three main concerns; the security of ducal personnel and property, the maintenance of order in the army, and the conduct of justice. The latter two categories here largely follow precedents from the La and need not attract close attention here. But the Lb does show innovation on the question of ducal security.Footnote ⁸⁴

The opening clauses of title II cover attempts on the life of the duke. This is the only offense in the entire Lb for which the death penalty is envisioned; all others are assessed in terms of monetary compensation, although these compositions are intended to avert or settle feuds. Whereas Alemannic law had also specified death for attempts on the duke's life, the Lb changes its Alemannic model significantly. For example, the La had allowed for the accused to clear himself by an oath of innocence with twelve oath-helpers. The Lb, on the other hand, does not allow this. Instead it requires that the accusation be proven by the testimony of three witnesses, or by trial by combat if the witnesses disagree.Footnote ⁸⁵ This provision prevents high-status offenders, who might have been able to recruit oath-helpers easily, from clearing themselves without personal risk. In addition to attempted assassinations, Lb II.1 also classifies the coups with foreign intervention or the invitation of enemies into the duchy as attacks on the duke. Alemannic law had classified these offenses separately. A purely domestic uprising, however—the carmulum—does not warrant the death penalty, but the payment to the fisc of 600 solidi for the ringleader, 200 for other aristocratic conspirators, and 40 each for lesser followers.Footnote ⁸⁶

Another type of insurrection is that of the duke's own son. This can only have been an abstract possibility in Bavaria at the time that Odilo took power, as his son Tassilo was not born until 741. However, internal dissension had played a role in the prior generation of Bavarian dukes and was also an issue in the Alemannic ducal family from which Odilo had come.Footnote ⁸⁷ A son who rebels against a duke who is still competent to rule is to lose his inheritance and suffer exile. The clause is drawn from Alemannic law, but makes some modifications to the definition of the father's competence. The La had stipulated that the father's competence consisted in serving the king, leading the army, and mounting a horse. The Lb includes leading the army, mounting a horse, and executing the king's orders; but it adds being able to “contest in a judgment,”Footnote ⁸⁸ judge the people, use his weapons, and not be deaf or blind. The physical specifics may simply be clarifications of “leading the army,” but the decision of the Lb to include judicial activity is telling of Odilo's approach to his new office; the duke's authority depends not only on his military leadership, but on his ability to give justice.Footnote ⁸⁹

One other aspect of the security of the ducal administration and household is of note here; the Lb adds two clauses that appear to be original, Lb II.7 and 8. II.7 states that if anyone dies honorably in the service of the duke, especially in the army, then his heirs receive ducal protection in the maintenance of their inheritance. This apparently has minors in mind, as the clause specifies that the duke will protect their property “until they can themselves.” Lb II.8 specifies that anyone who kills another at the duke's command is exempt from prosecution or vengeance, and entitled to the duke's protection. These two clauses imply that a person might be reluctant to obey dangerous orders, either out of concern about the fate of their heirs should they be killed, or the risk of feud even if they are not. In effect, they obligate the duke to a reciprocal protection of his agents for their service over and above his responsibility to provide justice for the duchy generally.

Other matters relating to the security of the duke and his household are relatively minor modifications or adaptations of clauses from Alemannic law; laws on those who start disputes at the duke's court, steal ducal property, or disregard orders under the ducal seal.Footnote ⁹⁰

Title II of the Lb, then, gives several indications of Odilo's anxieties at his assumption of the Bavarian duchy and also his image of the ducal role in society, based on the Alemannic background but also anticipating and adapting to Bavarian realities. Requiring witnesses rather than oath-helpers to establish guilt or innocence in cases of treason enabled him to weaken the defenses of powerful magnates who might have attempted a coup. The extension of special legal protection to those who fought or killed on the duke's behalf would tie them more tightly to his service and reward loyalty. Underlying this also, however, is an ideology of the duke that presents him not only as the leader of the people in war but also, king-like, as the provider of law and justice, as seen in the modifications to the Alemannic law against rebellious princes.

The final section of the Lb with clear political implications for the new duke is title III, a section with no precedent in any other early medieval law code. Title III discusses the “genealogiae,” a term translated as “families” by Rivers but better understood more loosely as “kin groups,” as the structure and extent of these genealogiae are unclear. Lb III.1 names five such groups, the Hosi (or Huosi), Draozza, Fagana, Hahilinga, and Anniona. The ducal kin, the Agilolfings, are named as a sixth genealogia but distinguished from these five. These aristocratic kin groups are privileged by the grant of double composition for all injuries suffered, relative to the wergilds that will be specified for ordinary free Bavarians starting in title IV. The Agilolfings received quadruple composition and the duke a further third above that.Footnote ⁹¹ Therefore, where the base wergild for the life of a free Bavarian is 160 solidi, the genealogiae are entitled to 320, and Agilolfings to 640. The duke's own life is assessed at 900 solidi.Footnote ⁹² This last would seem to contradict Lb II.1's decree of the death penalty for killing the duke, but presumably the number is meant as a benchmark for calculating lesser penalties.Footnote ⁹³

The Lb tells us little about these other families apart from their higher compensation and the fact that they are “first after the Agilolfings.” No other early medieval law ties legal privilege to membership in specific kin groups; therefore, there is little to compare this clause with. Under the circumstances, we have suggested for the Lb's compilation, however, the most probable explanation is that these were powerful extended clans with deep roots in Bavaria.Footnote ⁹⁴ Odilo's grant of legal privilege to them would have been an effort to win their political support. Therefore, through title III, Odilo demonstrated to the most powerful Bavarian magnates that the instruments of written law could be beneficial, not only to the duke himself and to the church institutions he brought under his patronage, but to them as well. The higher compensations these families were now entitled to could have made them virtually immune to vendetta by lesser families, as the compensations owed when the feuds were settled would have favored them heavily and pushed their rivals toward economic disaster.

Unfortunately, we know very little about these genealogiae in practical terms. Only two of them are attested in any other sources. The Fagana appear in a Freising charter of 750, the same charter that is the first to mention the “ius baiovariorum.”Footnote ⁹⁵ In that document, Tassilo—who was still a minor at the time—together with members of the Fagana, give property at a place called Erichinga (Oberding, very near Freising) to the cathedral church of St. Mary at Freising. Four specific members of the Fagana are named (Ragino, Anulo, Wetti, and Wurmhart) and it appears that Tassilo himself is regarded as a member of the group. This is the only appearance of the Fagana name in any source outside the Lb.Footnote ⁹⁶

The Huosi are better attested. They are named specifically in two of Freising's Carolingian charters, one from 791 and another from 849.Footnote ⁹⁷ The charter of 791, shortly after the Agilolfings' fall, deals with a dispute within the kin over an inheritance that had to be arbitrated by Bishop Atto in the presence of the gathered relatives. On the assumption that all the names on the extensive witness list counted as Huosi, the charter has enabled the tracing of relationships through earlier charters in which these individuals appear; prosopographical research soon gives the impression that practically every important person in southwest Bavaria (around Freising and the region of modern Munich) was a Huosi or at least linked to them. Unfortunately, no source indicates what criteria, if any, the Bavarians themselves used to define members of the kin-group. Therefore, we simply cannot tell if all the earlier individuals linked by modern scholars to the witnesses of 791 really thought of themselves as Huosi or could have claimed the protections of Lb III.1.Footnote ⁹⁸ Nevertheless, we can sense the extent and influence of this kin-group from the appellation of a large district in western Bavaria as the “Huosigau” in the ninth century.Footnote ⁹⁹

The other kin-groups of the Lb, the Draozza, Anniona, and Hahilinga, never appear in any other source. This is probably because of the accidents of documentary survival. Compared with the relatively full texts of over 120 charters from Freising from the Agilolfing era, of which only one names genealogiae from Lb III (plus the two appearances of the Huosi in Carolingian-period charters), only seven charters of the era survive from the ducal capital at Regensburg, two dozen more from Passau, and a few dozen others from a handful of monasteries, plus summaries of property donations received by the bishops of Salzburg.Footnote ¹⁰⁰ It is likely that these other kin groups predominated in the north or east of Bavaria and are hidden among the kinship networks around prominent aristocrats of Tassilo's reign. Properties held by relatives of a Count Helmuni, for example, crop up at or near places with names suggestive of the “Hahilinga,” such as Hailing, near Straubing on the Danube.Footnote ¹⁰¹

The conception of a genealogia as an important social unit is not limited to the Lb or the groups named in title III; elsewhere in the laws, the term is used for the kin of lesser freemen, and Bavarian charters occasionally mention genealogiae that do not appear in the Lb.Footnote ¹⁰² The same charter that introduced members of the Fagana to us also mentions another genealogia, the Feringa, who contributed to the benefaction.Footnote ¹⁰³ Another document of 806–808 shows a group called the Mohingara (whose name survives at modern Feldmoching in Munich) trying to claim some of Freising's property.Footnote ¹⁰⁴ The Salzburg property summaries feature a group of Romani, descendants of Roman provincials, known as the “genealogia de Albina,” identifying them not with a kin-name but with a place-name (“Albina” being the modern Oberalm in Austria).Footnote ¹⁰⁵ Presumably, then, the propertied class of Bavarian society was made up of many such genealogiae, at least some with sufficiently coherent identities to have group names. The five singled out in Lb III for legal privilege must have been meant as pillars of Odilo's reign, in order to secure him in his office.

Title III also makes another unique statement: that the ducal office is to remain in the Agilolfing family perpetually, as this family is so pleasing to the Frankish crown. Again, no other early medieval law identifies any office with a specific family; therefore, this clause cries for explanation. It is often invoked in a direct fashion by historians to identify all dukes of Bavaria up to Tassilo III's fall in 788 as “Agilolfings,” even the shadowy figures of the sixth century found in Gregory of Tours' and Paul the Deacon's histories: Garibald I, Tassilo I and Garibald II.Footnote ¹⁰⁶ Carl Hammer, however, has recently demonstrated that the evidence does not offer certainty on this question.Footnote ¹⁰⁷ The law takes on a different meaning if we understand it as an innovation made by Odilo (or the anonymous compiler working on his behalf) near 736–738. The intent of the stipulation was to block claims to the ducal office by members of other genealogiae, not to give modern historians evidence about the history of the Agilolfings. It is likely that Theodo and his descendants had been Agilolfings, as the clause would carry little conviction if they had not been. But it cannot be invoked to envision a single ducal family stretching back for two and a half centuries. It may be relevant, in this context, that the law does not make the ducal office hereditary, so as to create a dynasty, but only indicates that candidacy for duke must include membership in the broad kin-group: “Whoever of that kin should be wise and faithful to the king, let them make him duke to rule that people.”Footnote ¹⁰⁸ This clause simultaneously legitimizes the succession of Odilo as a duke not descended from (or even closely related to) his predecessor, while reserving the possibility that he could in turn be succeeded by another Agilolfing in addition to his own son.

Instead of seeing the Agilolfing element of III.1 as a declaration of venerable Bavarian tradition, then, we should see it as an effort by Odilo to assert his own significance in the jumble of Bavarian genealogiae, if not actually to insert his own kin into Bavarian society. If it was important to exclude other genealogiae from claims to the ducal office, this can only mean that such claims were being contemplated. As is seen from the appearance of Tassilo as a member of the Feringa, membership in genealogiae was not mutually exclusive and the boundaries between kin-groups must have been very loose. No doubt there were members of the Huosi or the other groups who could trace relationships to Theodo and his heirs as well as Odilo could. Lb III, then, represents a compromise, an assertion of Odilo's exclusive legitimacy while buying off the claims of others with exalted legal status. In other words, the unstable political situation near 736 and the need for Odilo to prop up his fragile authority gave birth directly to a Bavarian nobility in the full sense of that word; a class of persons entitled to legal privilege on the basis of descent and kinship. By the 770s, the fluidity of kinship in a cognatic society meant that legislation ceased to speak of genealogiae and spoke directly of nobility.Footnote ¹⁰⁹

In summary, then, we have seen that internal evidence from the Lb itself, set against the political chronology of the duchy of Bavaria, makes it probable that the text was compiled in 736–737, not the sixth century as its prologue claims, nor between 744 and 748 as most historians since Brunner have thought. It was composed in order to support the new duke Odilo, using the laws of his native Alemannia as a framework. The work of composition itself may have been done at a church or monastery with connections to the Alemannic dukes, most likely Reichenau. With this context in mind, the unusual laws from the first three titles of the Lb jump into relief as indicators of a strategy for the outsider to generate authority for himself in his new duchy without the blunt instrument of military force. Through title I, Odilo undergirded church institutions with legal protection of their property and personnel, yet also made the highest church officials, the bishops, answerable to his own judgment for their actions. In title III, Odilo selected several of the most powerful aristocratic kin-groups to elevate above the rest of Bavarian society as the core of a new nobility, granting them special legal status signified by higher wergilds. In exchange for the privilege, he reserved the office of duke to his own kin, setting the Agilolfings above the rest of the duchy's political and social leadership. Finally, title II set out to protect the ducal household and its officers, protecting conspirators from being able to clear themselves with oath-helpers. It also assured those who risked danger in the duke's service of the direct protection of the duke against attempts on their estates or their lives. In the process, both explicitly and implicitly, Odilo added new elements to the ideology of the ducal office, making the duke not only head of a powerful family and leader of the people in war, but also the source of justice for his subjects, much in the manner of a king.

In the short run, it was not clear that this effort to effect regime change without violence would succeed. In 741, shortly after beginning a reorganization of the Bavarian churches in concert with St. Boniface of Mainz, Odilo was driven into exile by dissent in his duchy.Footnote ¹¹⁰ This reversal shows just how fragile the new duke's authority could be, and how real the threats envisioned in Lb II actually were. His opponents, unfortunately, remain anonymous in the brief mention of this setback in our sources; therefore, we don't know if the five genealogiae stood by the duke who had given them privileges or turned against him. The uprising did not last; Odilo was able to return in a matter of months, again without recorded military intervention. He even recovered enough stature to lead a coalition of Bavarians, Alemanni, Saxons, and Slavs against the Franks in 743.Footnote ¹¹¹ Charters from Freising indicate that his laws continued in effect during the reign of his son, Tassilo III, and several manuscripts of the Lb include the decisions of Tassilo's assemblies as a kind of supplement to it.Footnote ¹¹² The success of Odilo's legislation at setting the standard for political legitimacy in Bavaria might best be measured by the use his son's enemies made of it. After Charlemagne moved against Tassilo with questionable legality in 788 and deposed him, a new clause was inserted in title II in manuscripts of the law, the so-called Lex Tassilonis. Footnote ¹¹³ This new clause declared that a duke who was disloyal to the king could be removed by the king and sent to a monastery in penance for his sins, precisely what Charlemagne did to Tassilo. The reservation of the ducal office to the Agilolfings in title III maintained enough force that the Carolingians never replaced Tassilo: the region was governed by prefects and counts in the late eighth and early ninth centuries and then became a “sub-kingdom” under Charlemagne's grandson, Louis the German, avoiding the title “duke” throughout.Footnote ¹¹⁴ Not until the tenth century would the ducal title be granted to members of another family, and then not as a conscious revival of the Agilolfing duchy.

Some thirty years ago, Patrick Wormald drew attention to the highly ideological character of early medieval law texts in a seminal article.Footnote ¹¹⁵ He argued that Germanic kings needed to present themselves as lawgivers in order to legitimize their rule to a largely Roman public in the lands they had taken over. To this end, written law – any law – was needed, explaining why so few of the surviving texts seem suited to practical use in court. Wormald thought that the Lb was an exception, belonging to a southern European tradition (along with Visigothic and Lombard law) that still saw law books as authorities for actual cases. This study has sought to explore this connection between lawmaking, practical politics, and ideology in closer detail. Whereas it does not undermine Wormald's perception that the Lb's content was of practical use and not arbitrary, it also shows that this act of legislation was more actually at once practical and ideological, and, therefore perhaps more “northern,” and more royal, than Wormald thought. In the political instability of the mid-eighth century, the anonymous monks who wrote the Lb at the behest of Duke Odilo achieved a unique feat. They forged an instrument that enabled the transfer of power, not by fire and sword, but by the legitimation and incentives generated by law. Odilo's Carolingian cousins took the example to heart; Charlemagne built his own empire across Europe on the basis of law as well as his conquering armies. Even when he overthrew Odilo's own son, he did so not on the battlefield, but in the law court. In this way, the Lex baiuvariorum contributed to the making of Europe.