[3] In the year 718 that part of France south of the Loire and Burgundy lay between two powers which were to affect its destiny for more than a century. To the north we find regions which were dominated by the Carolingian family -- a family which Pepin of Heristal had established in power over Austrasia and Neustria and whose authority was being further increased by his able son, Charles Martel. To the south lay Spain, conquered by the Arab and Berber forces of Musa and Tarik -- forces which were about to cross the Pyrenees in an advance toward the heartland of Western Europe.

The Midi which faced these two adversaries in the early eighth century was a region without any over-all political cohesion of its own. It consisted of four relatively distinct areas. The largest of these was an Aquitaine which lay south of the Loire and west of the Rhone Valley, with its southern borders reaching the Narbonnaise in the east and a Gascon frontier along the Garonne in the west. Aquitaine had been part of the Merovingian Frankish state since the time of Clovis, but by the eighth century was controlled by a duke or princeps called Eudes. He and his heirs, Hunald and Waiffre, were to control its destiny for the next fifty years.

We find a second rather amorphous principality known as Provence, which probably consisted of the Rhone Valley south of Lyon and east of Aquitaine and the Narbonnaise, up to the crests of the French Alps -- roughly the area occupied by the old Roman Provincia. Provence was controlled by a series of local magnates who bore the title of patrician. We know little about these patricians beyond their names with one exception, and we do not even know whether or not they belonged to the same family like the principes of Aquitaine. Provence, like Aquitaine, had long been nominally a part of the Merovingian Frankish state, though soon [4] after the death of Dagobert it had managed to achieve a large measure of autonomy under native princes.⁽¹⁾

West of Provence along the shores of the Mediterranean we find the third distinct region of Southern France, the Narbonnaise or Septimania, as it was called. Unlike Aquitaine and Provence, the Narbonnaise had never been conquered by the Merovingians. Instead, for three hundred years it had been under Visigothic rule. Its boundaries seem to have been the Rhone on the east, a series of fortified cities like Uzès, Lodève, and Carcassonne on the north, the Pyrenean high country inhabited by the Basques on the west, and the Pyrenees and Mediterranean to the south.⁽²⁾ Though it had not shared the political history of Aquitaine and Provence, like them it seems to have had an instinct for autonomy, which in the late seventh century led to a serious revolt by a local Gothic magnate called Paul, a rising which had to be suppressed by Wamba, king of Visigothic Spain.⁽³⁾

The fourth region of the Midi was Gascony which occupied the remaining region of Southern France north of the Pyrenees. It is difficult to give Gascony's boundaries in any exact way. A later Carolingian writer summed it up by saying that the Gascons lived across the Garonne and among the Pyrenees. Following his lead we might hazard the opinion that Carcassonne and Roussillon formed the eastern boundary, the Garonne its northern border, the Atlantic its western one, and the Pyrenees its southern one. On the other hand it is uncertain whether Bordeaux and the Bordelais were part of Gascony or part of Aquitaine during this period.⁽⁴⁾ It is equally difficult to be sure that the Pyrenees formed Gascony's southern border, since other Basques lived south of them in Northern Spain. These Spanish Basques, who were similar in race and culture to their French Gascon compatriots, occupied an expanse of Spanish soil northwest of Saragossa in Pallars, Ribagorça, Aragón, Navarre, and Asturias and were generally independent of and at times hostile to the Visigothic [6] rulers of Spain. They appear to have maintained close ties with those Gascons who lived north of the Pyrenees. That this represented any formal political unity, however, appears doubtful, for it seems clear that during this period the Basques of France were ruled by a native family of dukes or principes who bore the name of Loup or Lupo. They, like the Spanish Basques, were independent of their Narbonnaise and Aquitanian neighbors.⁽⁵⁾

One more region might also be mentioned -- Catalonia, which was to be closely associated with the Midi after 778, though not before. In 713-714, however, Catalonia had been overrun by the Moors who occupied all of it with the possible exception of the Urgell region, where there is some evidence of the survival of an independent native Church and monastic tradition. Yet even here one must be careful, for there is some evidence that all of this part of Northern Spain was less conquered outright by the Moslems than handed over to them by dissident Visigothic nobles, one of whom, a certain Cassius, apostatized from the Christian Church and set up an important Islamic dynasty in the region, the Banu Kasi.⁽⁶⁾ During the period covered by this chapter Catalonia followed the destiny of the rest of Moslem Spain rather than that of the Midi.

So much for the political geography of Southern France and the Spanish borderlands. But what of its ecclesiastical divisions? Only in Septimania do we find political and ecclesiastical boundaries coinciding. The Narbonnaise consisted of seven bishoprics: Elne, Carcassonne, Béziers, Agde, Maguelonne, Lodève, Nîmes, and Uzès under the archbishop of Narbonne.⁽⁷⁾ Provence on the other hand seems to have had a positive plethora of metropolitans: Aix, Embrun, and Arles near the lower Rhone, and Lyon and Vienne further north.⁽⁸⁾

Even more complex seems to have been the situation in Aquitaine and Gascony. By the late seventh century there were three archbishoprics in this part of France. The archbishop of Bourges was over the bishops of Northern and Central Aquitaine, the metropolitan of Bordeaux controlled [7] Southern Aquitaine and perhaps part of Western Gascony, the archbishop of Couserans the rest of Gascony. Gascony may also have had a special bishop of its own.⁽⁹⁾ Like Provence, then, Western France south of the Loire lacked the ecclesiastical precision of Septimania.

When one tries to examine the pre-Carolingian governmental system of the Midi in greater detail than the above generalities, one finds evidence rather scant. We do have evidence that the Narbonnaise was governed by counts,⁽¹⁰⁾ and that the Visigoths kept their mint at Narbonne in operation during the early years of the eighth century.⁽¹¹⁾ We hear also of counts in Aquitaine, where about 650 St. Didier's brother was count of Albi⁽¹²⁾ and where, at the time of Duke Waiffre, similar officials are found in Poitou, Berry, and Auvergne.⁽¹³⁾ On the other hand there is less evidence that counts existed during this entire period in Provence, though the patricians certainly exercised comital powers.⁽¹⁴⁾ As for Gascony -- it seems clear that it was without counts in the pre-Carolingian period, its only known rulers being its duces and principes who appear to have presided over tribal chiefs and families.⁽¹⁵⁾ The counts of Aquitaine and the Narbonnaise, [8] like their Carolingian successors, led local armed forces into battle and controlled at least some of the fortresses located in the areas they ruled.⁽¹⁶⁾ Slight evidence from Provence suggests that its patricians had similar authority.⁽¹⁷⁾

On the other hand we know nothing about the judicial role played by such counts in Southern France and even less of the court system in existence -- except for some indication that in Provence the patrician could and did act in a legal capacity.⁽¹⁸⁾ What we can say with some assurance is that Visigothic law was in use in the Narbonnaise,⁽¹⁹⁾ and that Roman law survived in both Aquitaine and the Valley of the Rhone.⁽²⁰⁾ Neither seems to have been in use, as far as we can tell, in Gascony.

[9] If our evidence of pre-Carolingian institutions is too slight for our liking, we know more concerning the pre-Carolingian military organization of Southern France. When Charles Martel and Childebrand advanced down the Rhone Valley with their Frankish and Burgundian levies, they found the principal cities of Provence, like Arles and Avignon, protected by walls formidable enough to call for siege operations.⁽²¹⁾ The same seems to have been true of civitates of Septimania like Uzès, Nîmes, Mauguio, Agde, Bèziers, Carcassonne, and especially Narbonne.⁽²²⁾ This region also contained castella located outside fortified cities like the Castellum of Millia which in 678 was mentioned as in existence between Nîmes and Maguelonne.⁽²³⁾

In Aquitaine we learn that St. Didier, about 650, rebuilt the walls of Cahors and reconstructed the castella of the city, as well as the Castrum Mercurio in Cahors itself.⁽²⁴⁾ At about the same time we hear of a Castrum Garnomo near Bordeaux.⁽²⁵⁾ A little later in the eighth century there was a castrum in Velay, probably at LePuy.⁽²⁶⁾ In the early eighth century when Toulouse was attacked by the Moslems, it apparently possessed fortifications.⁽²⁷⁾

Later on about 735 we hear of the Castrum of Blavia near Bordeaux and other castra in the suburbs of the city,⁽²⁸⁾ and when Pepin conquered Aquitaine between 761 and 768 we hear of similar fortifications everywhere. Our accounts make it clear that the major civitates of this region, like Bourges, Limoges, Poitiers, Saintes, Clermont, Périgord, and Angoulême [10] were fortified.⁽²⁹⁾ There were also a number of detached fortresses like Thouars in Poitou, Castelluc in Auvergne, Bourbon in Berry, Turenne in the lower Limousin, Scalas in Quercy, Perrucé in Périgord, and a number of other unspecified castella and roccas.⁽³⁰⁾

There is also ample evidence that the profession of arms was a general one, so much so that in Southern Aquitaine a church council in the late seventh century had to forbid the carrying of arms and the waging of war by priests.⁽³¹⁾ In the Narbonnaise and Aquitaine there was a military class led by counts and principes which furnished opposition to invading Moors and Carolingians alike.⁽³²⁾ The assassination of St. Didier's brother shows us a society in Quercy and the Albigeois which was tumultuous and quarrelsome,⁽³³⁾ just as Abbo's will mentions land, in 739, which he had acquired by conquest.⁽³⁴⁾

On the other hand there does seem to be some indication that pre-Carolingian Southern France did not possess a military class as well organized as those vassi of Frankish origin whom Charles Martel enriched with Church lands⁽³⁵⁾ and who followed him and his successors into battle. When the Carolingians advanced down the Rhone they found that Maurontius, patrician of Provence, had invited Moslem forces to garrison his cities, probably because he could not muster sufficient warriors of his own.⁽³⁶⁾ And in Aquitaine during Pepin's final nine-year campaign, we again and again find mention of Gascons, probably mercenaries, who formed at least a portion of the levies available to Duke Waiffre and his counts.⁽³⁷⁾ On the other hand there is no evidence of the use of such outside auxiliaries in either Septimania or Gascony.

[11] When we turn to the pre-Carolingian Church organization, we find that we can discover little, in no small measure because of the destruction of Church records by both the Moslems and Carolingians in the course of their campaigns in the Midi. For instance we do not know more than the names of any pre-Carolingian abbeys in the Narbonnaise,⁽³⁸⁾ which makes us suspect that, unlike those in Urgell, none survived Moslem occupation.⁽³⁹⁾ In Provence we can only be sure that Saint-Victor of Marseille⁽⁴⁰⁾ and Lérins⁽⁴¹⁾ continued active of the pre-Carolingian monastic establishments of this region. For Aquitaine we possess more information, but still too little for more than generalizations.

What emerges from what we do know, however, is a picture of a church and a monastic system closely linked to the ruling, landholding classes. This was no innovation of this period, but rather represented a continuation of a state of affairs which had existed since late Roman and early Merovingian times. Thus the example of Aredus and his grandson, who in 572-573 seem to have endowed the abbeys of Saint-Martial and Vigeois in the Limousin with considerable landed property,⁽⁴²⁾ was copied often in the next two centuries. We know this from the record of seventh-century gifts of land to the church of Viviers⁽⁴³⁾ and those given to churches and abbeys in Quercy and the Toulousain dating from the same period. New abbeys were also still being established, as we learn from the case of the abbey of Saint-Gerri established by St. Didier at Cahors⁽⁴⁴⁾ or that of [12] newly founded Moissac which gained in 680 the landed estates belonging to Nizezius.⁽⁴⁵⁾ As late as 739 Abbo left his vast estates in the Middle Rhone region to the newly founded abbey of Novalese in Italy.⁽⁴⁶⁾

The biography of St. Didier, which portrays him acting more like the ruler of Quercy than its bishop again illustrates how closely entwined the Church in the Midi was with the governing system of the region.⁽⁴⁷⁾ So do gifts by patricians of Provence to the abbey of Saint-Victor of Marseille, as revealed at a later period,⁽⁴⁸⁾ and the fact that in 780 the aged bishop of this city bears the same family name as its last patrician.⁽⁴⁹⁾ Except in Gascony there seems to have existed a close relationship and collaboration between bishop and monastery on the one hand and count and leading landowners on the other.

This leads us to a consideration of the economic life of Southern France in the first years of the eighth century, on the eve of Moorish and Carolingian intervention. By this time, it is clear, the outside commerce enjoyed by this region was in a definite state of decline. There is no evidence, for instance, to show the presence of Syrian and Greek merchants in the civitates of the Midi, as had been the case a century earlier.⁽⁵⁰⁾ We do still find a reference to fine silks,⁽⁵¹⁾ however, and, judging from later Carolingian evidence, it is possible that some Jewish merchant colonies still survived.

At the same time one notices an interesting change in the coinage which is in use. During the last years of the seventh century contemporary accounts still mention gold being used in Quercy and the Toulousain,⁽⁵²⁾ a fact which seems confirmed by gold found in coin hoards at Bordeaux⁽⁵³⁾[13] and along the Loire⁽⁵⁴⁾ and the Rhone.⁽⁵⁵⁾ By 718, however, this seems no longer to have been true. Silver seems to have replaced gold as a medium of exchange. It was coined at Narbonne⁽⁵⁶⁾ and in Aquitaine⁽⁵⁷⁾ and is the only metal found in hoards like Plaissac on the Garonne⁽⁵⁸⁾ and Cimiez in Provence.⁽⁵⁹⁾ Such coin hoards also reveal the growing economic localism of the Midi, particularly in the Rhone Valley,⁽⁶⁰⁾ though they do seem to point to a trade route which was still active along the western shores of Aquitaine to Bordeaux and then along the Garonne to Narbonne and Marseille.⁽⁶¹⁾ There is also some evidence that some spices and oriental wares were still reaching Fos in the early eighth century.⁽⁶²⁾ Compared to the earlier Merovingian period, however, by 718 Southern France seems to have been more isolated from the main trade currents of Northern Europe and the Mediterranean than had been the case earlier.

All of this contributed to making land in the Midi its most important source of wealth. And it seems clear that this landed wealth was still held in the early eighth century in much the same way that it had been owned in late Roman and early Merovingian times. It was still largely in the hands of the same sort of Gallo-Roman aristocrats whom we meet at the time of Sidonius and Avitus.⁽⁶³⁾ It must not, however, be thought that such families had a monopoly of land ownership. Side by side with them we find a sizable Frankish element -- numerous in the Limousin, Rouergue, and the Albigeois, and present as far south as Carcassonne.⁽⁶⁴⁾ A Visigothic [14] element is present in the Narbonnaise too in this same period.⁽⁶⁵⁾ One finds this Frankish element mentioned in contemporary accounts, for when Duke Eudes of Aquitaine met and defeated the Moslems near Toulouse in 721, his army is said to have been composed of both Franks and Aquitanians.⁽⁶⁶⁾ Judging by his name Eudes himself was of Frankish origin. Long before the Carolingians, then, a considerable Frankish element had settled in parts of Aquitaine. It also needs to be emphasized that Franks and Burgundians were settled in the Middle and Lower Rhone Valley before the time of Charles Martel. Abbo, patrician of Provence, bore a Frankish name, and we find similar northern Frankish soubriquets among the servants whom he mentioned in his will.⁽⁶⁷⁾

It seems highly probable, however, that by the early eighth century such Frankish and Burgundian elements had been largely absorbed into the prevailing Gallo-Roman population of Provence⁽⁶⁸⁾ and Aquitaine, just as in the Narbonnaise the Visigothic element had become a part of the essentially Gallo-Roman society of Septimania.⁽⁶⁹⁾ Only in Gascony does one find what may represent an unmixed stock, and even here there seems to be some evidence that in Couserans and the Bordelais, Gascons were still advancing into what had earlier been Gallo-Roman areas.⁽⁷⁰⁾

Whether its owner was of Gallo-Roman, Frankish, Burgundian, or Visigothic origin, however, the villa was still the prevailing unit of landholding, as had been the case since late Roman times. St. Didier, for example, gave villas as an endowment to the abbey he founded in Quercy [15] and to the churches which he restored there and in the Albigeois.⁽⁷¹⁾ Our sources make clear that the estates of Nizezius located between the Tarn and the Garonne were organized as villas in 680.⁽⁷²⁾ We find villas mentioned in Vivarais as well as mansi.⁽⁷³⁾ In Provence, around Marseille, the villa, made up of colonicas, seems to have prevailed,⁽⁷⁴⁾ while as far north as Dauphiny, Abbo's will shows the same pattern of land used in the Rhone Valley.⁽⁷⁵⁾ We have every reason to believe that the same thing was true in the Narbonnaise also, though this belief rests on conjecture alone.

Some landowners in this period controlled only a villa or two. More typical, however, seems to have been a class of large estate owners, controlling vast areas -- men such as St. Didier and Nizezius in Aquitaine, or others like them in Velay, or magnates like Abbo who possessed so much property in Dauphiny. Nor were men alone in possessing such estates. We find women like Bobila, the senatrix Romana of Quercy, who owned property in their own right.⁽⁷⁶⁾ We also find a large class of servi or coloni living on such villas as they had in late Roman time, though a group of freedmen also seems to have existed.⁽⁷⁷⁾ It is worth noting that the prevailing system of landholding for the upper class was allodial and that landowners [16] could and did freely sell or will their estates or give them to abbeys and churches.

There can be little doubt that in this period the allod was the prevailing type of landholding to be found. But it is important to note that we have evidence of another sort of tenure -- the precarium or beneficium.⁽⁷⁸⁾ In the biography of St. Didier we find a reference to beneficia which were given out by nobles and magnates of the region and something called beneficiendo or cohemendo which seems somewhat similar to Carolingian commendation.⁽⁷⁹⁾ In a document from Provence from later Carolingian times we hear how Metrannus, patrician of Provence, about 700 A.D. gave a villa to the abbey of Saint-Victor pro beneficio and that his successor Abbo did the same in beneficium.⁽⁸⁰⁾More explicit is information concerning landholding found in Abbo's will, dealing as it does with land over a wider area in the Valley of the Rhone and in Dauphiny. This will mentions land which "Australdus habet in beneficio" or which "Bartona libertus noster in benefitium habet," and others as well. There is also a reference to "fidele meo Protadio."⁽⁸¹⁾ In a rare document from Auvergne dating from 756-757 we also find Waiffre, princeps of Aquitaine, giving a villa to a certain Gédéon, as a life precarium, in return for another villa and two pounds of silver -- this some five years before the final Carolingian offensive against Aquitaine.⁽⁸²⁾ Though we have no surviving charters from pre-Carolingian Septimania, from what we know of Visigothic Spain during this period we can fairly assume that such precaria are to be found in the Narbonnaise too.⁽⁸³⁾

[17] Unfortunately the examples of precaria and beneficia dating from this period in the Midi are so few that it is difficult to be certain about their exact nature or the conditions under which they were given. In the case of Gédéon it is clear that the precarium granted was for a lifetime only and that he had to make a definite payment in both land and money for it.⁽⁸⁴⁾ Concerning the others we know little, except that in certain cases they were not given in return for military service and were not the monopoly of the upper classes. Several of Abbo's precaria granted to freedmen make this clear.⁽⁸⁵⁾ What seems more important, however, is the evidence which can be drawn from these examples that a precarium frequently was a private grant by an allodial landholder who chose this method of disposing of rights to his property. The precarium, then, was a right to use an allod under certain conditions prescribed by its owner.⁽⁸⁶⁾ As in Visigothic Spain or Lombard Italy,⁽⁸⁷⁾ then, there existed in the Midi conditional grants of land called beneficia or precaria before the period of the Carolingians.

A final word concerning land in Southern France during these years [18] seems in order. How much vacant or uncultivated land was there? We know that there was a great deal of it in later Carolingian times in the Narbonnaise, in Roussillon, in the Rhone Valley, in Aquitaine, and in Catalonia. Most historians have tended to explain this fact by calling attention to the six decades of disorder in the Midi between 718 and 778. In other words they have given the blame to invading Moslem and Carolingian forces for a devastation to which they ascribe such a condition.

Such views seem exaggerated, to say the least. Except for two razzias, one which reached Autun about 722⁽⁸⁸⁾ and another Poitiers in 732,⁽⁸⁹⁾ Moslem raids were confined to the Narbonnaise, Rouergue, the Albigeois, Velay, and the Lower Rhone Valley, and even here they seem to have lasted for only a few decades. Carolingian conquest of the Rhone Valley between 736 and 739 was a swift affair and Carolingian activities in Septimania seem to have been confined to the same period and to the years 752-760, when the subjugation of this region and nearby Southern Aquitaine was finally completed. Pepin's conquest of the rest of Aquitaine, though destructive, took place over an eight-year period between 761 and 768. Gascony was not entered by Carolingian forces until 769 and then again in 778 during the campaign against Saragossa. Thus neither the military activities of the Moslems nor of their Carolingian rivals really explain the vast amount of uncultivated land found in the Midi after 778, though they help explain some of it.

Thus we must look for another explanation. Perhaps the most satisfactory one lies in the fact that the society of the Midi -- and this includes Catalonia -- had not in the period before 718 begun to use land very effectively. The villa system, which still seems to have been the dominant method of landowning, simply did not provide a way of settling new land which the coloni who dwelt on such villas might well have desired to cultivate for themselves. Nor do we have any evidence that the Church in the Midi was any more enterprising in this respect.⁽⁹⁰⁾ Not until the Carolingians established a new land policy were the waste places of the [19] Midi to be put to the plow or brought into effective use, and this in a new and different era.

Such then was the pattern of Southern French society on the eve of Moslem and Carolingian intervention. It was a region which consisted of large estates (except perhaps in Gascony), which were generally cultivated by serflike coloni and which were run by an aristocracy of allodial landholders, who made some use of precaria as well. It was a region whose Church, except in Gascony, seems to have been closely linked to the same landholding class and whose government, divided into four great regions, was in the hands of counts and similar officials. Throughout its area, again with the exception of Gascony, Roman and Visigothic law had survived, and the aristocracy, who were frequently fighting men, relied in part on castles and other fortresses to defend themselves. The society was one, however, in which neither government nor military service was well organized and in which the economy was increasingly local. Its two stronger neighbors, Moslem and Carolingian, were soon to intervene in its affairs and bring it to a new and different era.

1. E. Duprat in Bouches du Rhône: Encyclopédie départementale, II, 248-250; G. de Manteyer, La Provence du I^er au XII^e siècle, pp. 100-154.

2. The basic work on the Narbonnaise in the pre-Carolingian period remains A. Dupont, Les cités de la Narbonnaise première dupuis les invasions germaniques.

3. Julian of Toledo, Historia Excellentissimi Wambae Regis, ch. 5, in España sagrada, vol. VI; Chronologia et series regum Gothorum, in Hist. Gén. de Lang., II, col. 19-20; Divisio terminorum episcopatum provinciae Narbonensis, in ibid., II, col. 20-21.

4. On Basques on the French side of the Pyrenees see J. Jaurguin, La Vasconie, I, 80 ff. On the fact that these Basques "trans Garronam et circa Pireneum montem habitant" see Annales Regni Francorum, ed. G. Kurze, anno. 816, p. 141.

5. On the Basque-inhabited lands in Northern Spain see J. Lacarra, Vasconia medieval: Historia y Filología, pp. 14-20.

6. On this change to Islamic control in Northeastern Spain see R. de Abadal i de Vinyals, "El paso de Septimania del dominio godo al franco a través de la invasión sarracena," in Cuadernos de Historia de España, XIII, 9-15.

7. Divisio terminorum episcopatum provinciae Narbonensis, col. 20-21. By the seventh century these bishoprics had been completely integrated into the Visigothic kingdom and their bishops were regularly represented at Church councils held at Toledo (Hist. Gén. de Lang., I, c. 7, no. 16).

8. On the bishops of Provence see Bouches du Rhône, II, 248-249.

9. Hist. Gén. de Lang., II, no. 3, col. 40-41, and C. Higounet, Le comté de Commignes de ses origines à son annexation à la Couronne, I, 12-15. The above charter mentioning the Council of 673 contains among its witnesses a bishop who calls himself the bishop of the Basques. See also mention of bishoprics in Auvergne, and at Bourges, Rodez, Agen, Angoulême, Périgord, and Cahors, in Vie de Saint Didier, ed. R. Poupardin, VIII, 27-28.

10. See the mention of counts in the Narbonnaise in the late seventh century in Julian of Toledo, Historia Excellentissimi Wambae Regis, ch. 5. On the Gothic Count Misemundus, who was the ally of Pepin the Short in the Narbonnaise see Chronicon Ucenense, in Hist. Gén. de Lang., II, col. 25-27 (hereafter cited as Chron. of Uzès). On the Gothic count of Maguelonne who was the father of St. Benedict of Anime see Vita Benedicti Anianensis, ed. E. Waitz, bk. I, in Mon. Ger.
Hist. Scriptores, XV, 201.

11. On the money of Narbonne coined as late as the reign of Witiza (701-711) see Hist. Gén. de Lang., VII, 324.

12. Vie de Saint Didier, I-II, 1-9, and V, 21.

13. Fredegarius, Chronicon quod dicitur Fredegarii continuatio, ed. R. Krusch, ch. 125, 129-130, 134, in Mon. Ger. Hist. Scriptores renum Merov., II, 187, 189-190, 192 (hereafter cited as Fredegarii cont.).

14. See account of testimony given to a later Carolingian court (in 780) by the aged Bishop Maurontius concerning the power exercised by pre-Carolingian patricians of Provence over the abbey of Saint-Victor of Marseille, in Cartulaire de l'abbaye de Saint-Victor de Marseille, ed. B. Guéraud, no. 31, pp. 43-46 (hereafter cited as Cart. de Saint-Victor).

15. One of the earliest references to such duces or principes of Gascony is in 673 when a document refers to a certain "Lupo duce" in Hist. Gén. de Lang., II, no. 3. See also reference to this same Lupo at about the same period in a document which says that he controlled a number of cities in Aquitaine (Miracula Sancti Martialis Lemovicensis, ed. D. Bouquet, bk. II, ch. 3, in Recueil des Historiens des Gaules et de la France, III, 580). Later in 769, Lupo dux of the Gascons, submitted to Charlemagne (Annales Laurissense, in Mon. Gen. Hist. Scriptores, I, 148; Einhard, Vie de Charlemagne, ed. L. Haiphen, ch. 5, p. 445; and Astronomus, Vita Hludovici imperatoris, ed. G. Pertz, ch. 2, in Mon. Gen. Hist. Scriptores, II, 608 [hereafter cited as Astronomus, Vita Hludovici]).

16. Count Misemundus or Ansemundus in 752 turned over to Pepin fortresses at Nîmes, Maguelonne, Agde, and Béziers (Chron. of Uzès, col. 25-26; Annales Anianenses, in Hist. Gén. de Lang., II, col. 5-6 [hereafter cited as Annals of Aniane]; and Chronicon Moisiacense, ed. G. Pertz, in Mon. Gen. Hist. Scniptores, I, 293-294 [hereafter cited as Chron. of Moissac]). See also the account of the independent authority in Aquitaine exercised in 761 by Umbert, count of Bourges, and Bladino, count of Auvergne, according to Fredegarii cont., ch. 125, p. 187.

17. Though the Chron. of Moissac, p. 291, says that the Moslems seized Arles in 734, another contemporary source says that in 737 the patrician Maurontius turned Avignon over to them (Fredegarii cont., ch. 19, pp. 177-178). This would imply that at this date he still controlled this city and its castra. Another local source agrees with the Chron. of Moissac (Annals of Aniane, cols. 3-5). Duprat in Bouches du Rhône, II, 131-132, A. Molinier in "Sur les invasions Arabes dans le Languedoc," in Hist. Gén. de Lang., II, 550-552, and H. Zotenburg in ibid., pp. 557-558, follow the Fredegarius version.

18. See account of court of 780 already mentioned (Cart. de Saint-Victor, no. 31).

19. The Visigoths of Narbonne only surrendered this city to Pepin in 759 when he assured them they could continue to use their own Visigothic laws (Chron. of Moissac, p. 294, and Annals of Aniane, col. 6-7).

20. See Chapter 3, Section II, on the survival of Roman and Visigothic law in Aquitaine and the Valley of the Rhone in later Carolingian times. For a more contemporary example of the use of Roman private law see the will of Abbo, dating from 739, in Cartulaire de l'église Cathédrale de Grenoble, ed. J. Marion, pp. 34-49, and the comments on its Roman legal content in G. Chevrier, "L'évolution de l'acte à cause de mort en Dauphiné du VIl^e à la fin du X^e siècle," in Recueil de la société historique de droit et des institutions des pays du droit écrit, I (1948).

21. Chron. of Moissac, p. 292. Fredegarii cont., ch. 20, p. 178, mentions a siege of Avignon and its castra in 737 by Childebrand, the brother of Charles Martel.

22. On Charles Martel's destruction of the castra and walls of the civitates of the eastern Narbonnaise after he retreated from before Narbonne in 738 or 739 see Annals of Aniane, col. 4-6; Chron. of Moissac, p. 292; and Fredegarii cont., ch. 20, pp. 177-178. On Pepin the Short's long siege of Narbonne, which did not fall until 759 see Chron. of Moissac, p. 294; Annals of Aniane, col. 6-7; and Chron. of Uzès, col. 26-27. Obviously the Narbonnaise was heavily fortified during this period, as it had been in Visigothic times.

23. Divisio terminorum episcopatum provinciae Narbonensis, col. 21.

24. Vie de Saint Didier, II-III, 16-19, and VIII, 50.

25. Hist. Gén. de Lang., II, no. 3.

26. Hist. Gén. de Lang., II, no. 208, contains references to the castle of Viviers, which date from the seventh century.

27. Chron. of Moissac, p. 290; Annals of Aniane, col. 2-3; and Chron. of Uzès, col. 25.

28. Fredegarii cont., ch. 15, p. 176.

29. Ibid., ch. 126, 129, pp. 187-189, Annales Laurissenses, pp. 142-144. On walls of Limoges which were destroyed by Pepin see a later account in "Ex historia monasteril usercensis," in Cartulaire de l'abbaye d'Uzerche, ed. J. Champéval, pp. 13-14.

30. Fredegarii cont., ch. 25, 125-126, 129, pp. 180, 187-189, and Annales Laurissenses, pp. 142, 144, 146.

31. Hist. Gén. de Lang., II, no. 3.

32. See Chapter 2, Section I, on the warlike nature of this society.

33. Vie de Saint Didier, II, 15-16.

34. Cart. de Grenoble, pp. 34-38. It is interesting to note that Abbo's will mentions no fortresses within or near his domains.

35. A. Boretius and V. Krause, Capitularia regum Franconum, in Mon. Gen. Hist. Capitularia, I, nos. 10-12, pp. 29-32 (hereafter cited as Boretius, Capitulania).

36. See Note 17. Local authorities and sources mention Moslem conquest rather than a collaboration between Maurontius and the Moslems.

37. Ademar de Chabannes, who may be drawing upon an unknown source from Aquitaine, says that as early as 718 Eudes, duke of Aquitaine, used Gascons in his wars with Charles Martel (Ademar de Chabannes, Chronique, ed. A. Chavanon, I, 51, pp. 51-52 thereafter cited as Ademan de Chabannes]). The more contemporary Fredegarii cont., ch. 10, agrees. In 742 Duke Hunald had Gascon allies (ibid., ch. 25). In 761 Count Bladino of Clermont had Gascons in his garrisons (ibid., ch. 125). So did Count Humbert of Bourges in 762 (ibid., ch. 126). During Duke Waiffre's last years of struggle in 764-766, he had Gascon assistance (ibid., ch. 130-131).

38. They are Saint-Basil of Nimes, Saint-André of Agde, and Saint-Gilles, probably located in the Rhone delta (Hist. Gén. de Lang., I, ch. 6, no. 70).

39. On the survival of Visigothic abbeys in Urgell see R. de Abadal i de Vinyals, Els Primers Comtes Catalans, pp. 115-120, and "La batalla del adopcionisme en la destintegración de la Iglesia visigoda," in Real Academia de Buenas Letras de Barcelona.

40. See Cart. de Saint-Victor, no. 31, and Introduction to this cartulary by B. Guéraud, pp. i-xxii.

41. See Cartulaine de l'abbaye de Lénins, ed. H. Mons and E. Blanc, and ibid., I, 293, no. 290.

42. Cartulaire de l'abbaye de Vigeois, ed. M. de Montegut, no. CLI, pp. 95-97.

43. Hist. Gén. de Lang., II, no. 208.

44. Vie de Saint Didier, VI, 22-23.

45. Ibid., VII, 25-26, for mention of abbey of Moissac. On land acquired by Moissac see Hist. Gén. de Lang., II, no. 4, col. 42-45.

46. Cart. de Grenoble, pp. 34-48.

47. Vie de Saint Didier, II-III, 16-19.

48. Cart. de Saint-Victor, no. 31, mentions a series of gifts to this monastery by patricians of Provence.

49. Ibid.

50. Dupont, Les cités de la Narbonnaise, pp. 210-212.

51. The abbey of Moissac included pallios worth 200 solidi in the purchase price they paid to Nizezius for certain large tracts of land in 680 in the Toulousain region (Hist. Gén. de Lang., II, no.4, col. 45).

52. The same charter mentions 700 solidi (ibid.).

53. P. Le Gentilhomme, Mélanges de Numismatique Merovingienne, pp. 9-10. See also comments by J. Lafaurie in Settimane de studio del Centro di studi sull'-alto medioevo, VIII (1960), 254-257, and in Revue Numismatique, XXIII (1959-1960), 153-210.

54. Le Gentilhomme, Mélanges de Numismatique Merovingienne, pp. 120-125, on hoard discovered at Baugisière.

55. Ibid., pp. 125-130, on hoard discovered at Buis in the Rhone Valley.

56. Hist. Gén. de Lang., VII, 324.

57. Ibid., p. 367.

58. Le Gentilhomme, Mélanges de Numismatique Merovingienne, pp. 126-128.

59. A. Morel-Fatio, Catalogue raisonné de la collection des deniers merovingiens des VIl^e et VIll^e siècles de la trouvaille de Cimiez.

60. See A. Lewis, "Le commerce maritime et la navigation sur les côtes de la Gaule atlantique du V^e au VIll^e siècle," in Le Moyen Age, LIX (1953), 277.

61. Ibid., pp. 270-276.

62. L. Levillain, Examen des chartes de Corbie, p. 220.

63. E. Salin, La Civilisation Merovingienne, I, 25-30.

64. M. Brouens, "Le peuplement germanique de la Gaule entre Méditerranée et l'ocean," in Annales du Midi, XLVIII (1956), 17-37. On this large Frankish element in Rouergue, which, in part, may date from Carolingian times, see C. Higounet, "Observations sur la seigneurie rurale et l'habitat en Rouergue du IX^e au XIV^e siècle," in ibid., LXII (1950), 121-134. On the similar situation in the Limousin see G. Tenant de la Tour, L'homme et la terre de Charlemagne à Saint Louis, pp. 87 ff.

65. Dupont, Les cités de la Narbonnaise, pp. 200-201. Visigothic elements seem to have settled in some numbers near Béziers, Perpignan, Narbonne, and Carcassonne -- that is to say in Western Septimania. Some of this Gothic population, however, may represent later Carolingian aprisio holders.

66. Annals of Aniane, col. 2; Chron. of Uzès, col. 25; and Chron. of Moissac, p. 290.

67. Abbo's will mentions men who bear such names as Astruald, Siguald, and Merobert, and who hold benefices from him (Cart. de Gnenoble, pp. 34-48). This seems to show a Frankish and Burgundian element present in Dauphiny in 739. On Clothair II's absorption of Provence into the Merovingian realm in the early seventh century see H. Adelson, "Early Medieval Trade Routes," in American Historical Review, LXV (1960), 2-3.

68. Again it is the will of Abbo mentioned earlier which shows how romanized Abbo and other Franks had become in this part of France in the early eighth century. On this see also Chevrier, "L'evolution de l'acte"

69. Dupont, Les cités de la Narbonnaise, p. 201.

70. On this Gascon penetration and mixing with a non-Gascon population see Higounet, Le comté de Commignes, I, 13-15.

71. Vie de Saint Didier, IX, 31-38. On the continuation into the ninth century of this villa pattern in this region see Higounet, "Observations sur la seigneurie rurale en Rouergue," pp. 17-23. The rare charter of 756-757, issued by Duke Waiffre in Aquitaine only mentions villas in Auvergne (Cartulaire de Brioude, ed. H. Doniol, no. 26, pp. 47-48). Ademar de Chabannes also mentions that during the course of his war with Duke Waiffre between 761 and 768 Pepin gave two villas to the canons of Saint-Martial and Saint-Etienne of Limoges (Ademar de Chabannes, I, 58, pp. 60-61).

72. On the nineteen villas of Nizezius acquired by the abbey of Moissac in 680 see Hist. Gén. de Lang., II, no. 4. See also an analysis of these estates in C. Higounet, "L'occupation du sol du pays entre Tarn et Garonne au moyen âge," in Annales du Midi, LXIV (1952), 312-314.

73. Hist. Gén. de Lang., II, no. 208.

74. See mention of the villas of Caladio and Alpheus owned by the abbey of Saint-Victor of Marseille in Cart. de Saint-Victor, no. 31, pp. 43-45. On the prevailing villa system in Provence see R. Latouche, "Quelques aperçus sur le manse en Provence au X^e et au XI^e siècles," in Recueil des travaux offerts à M. C. Brunel, II.

75. Cant. de Grenoble, pp. 34-48.

76. Vie de Saint Didier, IX, 33.

77. See mention of coloni of the villa of Rusticiago in Rouergue at the time of St. Didier (Vie de Saint Didier, VIII, 27). See also the serfs of five villas owned by Nizezius in the Toulousain in 680 (Hist. Gén. de Lang., II, no. 4, col. 42-43).

78. Ganshof has noted the lack of any satisfactory study of the precarium of Merovingian times. This seems especially true of pre-Carolingian Southern Gaul (F.Ganshof, "L'origine des rapports féodo-vassiliques," in I problemi della Civiltà Carolingia, p. 48 n.).

79. Vie de Saint Didier, IX, 32.

80. Cart. de Saint-Victor, no. 31.

81. Cart. de Grenoble, pp. 34-48. It is worth noting that while humble freedmen like Bartone are given benefices, the mention of a fidelis Protadio may show a more honorable relationship between Abbo and some of those who held his land.

82. Cart. de Brioude, no. 26, pp. 47-48.

83. C. Sánchez-Albórnoz has made it clear that various types of tenure of this sort, both military and nonmilitary in nature, were common in Visigothic Spain, and thus also in Septimania which was part of this realm. See his "España y el feudalismo Carolingia," in I problemi della Civiltà Carolingia, pp. 111-124, and El "stipendium" hispano-godo y los origines del beneficio pre-feudal. A canon of the thirteenth council of Toledo seems to imply that much land was held in this fashion by magnates in Septimania, who as vassals of King Wamba had their land confiscated for disloyalty (Canon I, XIIIth Council of Toledo, in España sagrada, VI, 29). On Wamba's law of 683 which forbade churches giving out their land as precaria or sub stipendium see Lex Visigothorum, IV, 5-6, in Mon. Ger. Hist. Leges, I, 202.

84. Cant. de Brioude, no. 26, pp. 46-47. Note that when Pepin in 743 or 744 in theory returned to the Church the estates he had given his vassals, he ordered those who held such lands to pay a cens to the Church owner and receive from it a charter of precarium (Boretius, Capitularia, I, nos. 10-12, pp. 29-32). Ganshof notes that in 779 these were called "precaria verbo regis" (Capitulary of Heristal, anno 779, in Boretius, Capitularia, I, no. 20, art. 13, p. 32). Such precaria seem similar to that in Auvergne which Duke Waiffre gave to Gédéon. Perhaps Waiffre was copying Charles Martel and Pepin in an attempt to raise forces to protect his realm against Carolingian attacks.

85. In the eighth century, before Charles Mattel attracted important magnates into ties of vassalage by distributing Church lands to them in 743, vassalage in Frankish domains generally was reserved for men of humble condition (Ganshof, "L'origine des rapports féodo-vassiliques," pp. 35-42). This remained true also under Pepin the Short and even under Charlemagne (see Boretius, Capitularia, I, nos. 15, 16, 104, pp. 38, 44, 215). This seems generally true also of all Southern France north of Septimania during this period.

86. Ganshof reports the same thing for precaria and benefices in general (Ganshof, Qu'est-ce que la féodalité? pp. 22-24, and "L'origine des rapports féodo-vassiliques," p. 49).

87. On pre-Carolingian Italian feudalistic institutions and customs that parallel those of Visigothic Spain and Merovingian Southern France see P. Leicht, "Il Feudo in Italia nell'età Carolingia," in I problemi della Civilta Carolingia, pp. 71-107.

88. Annals of Aniane, col. 2-3.

89. Ibid., col. 3-4; Chron. of Moissac, pp. 291-292; and Fredegarii cont., ch. 13, p. 175.

90. Here Higounet's study of the cultivation of Moissac's land located between the Tarn and the Garonne in this period and in later Carolingian times is of vital importance. He shows that only old used land was cultivated. New lands were not exploited until much later (Higounet, "L'occupation du sol du pays entre Tarn et Garonne au moyen âge").