^∗ For their help in the preparation of this article I would like to thank Professor Otto Maenchen of the University of California, Berkeley, Professor Fred A. Cazel, Jr., of the University of Connecticut, Professor Walter Goffart, University of Toronto, and Miss Jeanne Peterson, University of California, Berkeley.Google Scholar

¹ Certain aspects of the Alan settlement in Gaul have been dealt with by Levison, W., ‘Bischof Germanus von Auxerre …,’ Neues Archiv der Gesellschaft für ältere deutsche Geschichtskunde 29 (1903) 133–140; Ganshof, F. L., ‘Note sur le sens de “Ligeris” au titre xlvii de la Loi Salique et dans le “Querolus,”’ Historical Essays in Honour of James Tait (Manchester 1933) 116–120; and Thompson, E. A., ‘The Settlement of Barbarians in Southern Gaul,’ Journal of Roman Studies 46 (1956) 65–67.Google Scholar

George Vernadsky has written several pieces on the Alans, the latest of which, ‘The Eurasian Nomads and their Impact on Medieval Europe,’ Studi Medievali ³ 4 (1963) 1–34, is a synthesis of his earlier works. On pages 23–31 Vernadsky treats the Alans in Gaul. I am seldom able to agree with Professor Vernadsky's interpretations, mainly because he does not seem to appreciate the attraction which Romania had for the barbarians. This attraction is illustrated in a statement attributed to the Visigothic king, Ataulf, by Orosius, , Historiarum adversum paganos libri VII 7.43 (ed. Zangemeister, C.: CSEL 5 [Leipzig 1882] 560), to the effect that his original intention that Gothia should replace Romania could not succeed since the Goths were too barbaric, and thus he intended to support Romania with Gothic power. Thompson, E. A., op. cit., and Wallace-Hadrill, J. M., The Long-Haired Kings and Other Studies in Frankish History (London 1962), have developed the theme of Rome's attraction for the barbarians in a most illuminating manner.Google Scholar

² Rostovtzeff, M., Iranians and Greeks in South Russia (Oxford 1922) 118–19, and Thompson, E. A., Attila and the Huns (Oxford 1948) 22ff.Google Scholar

³ Courtois, Christian, Les Vandales et l'Afrique (Paris 1955) 39ff. and Jones, A. H. M., The Later Roman Empire (Oxford 1964) I 158, 159, 184, and 185.Google Scholar

⁴ Gregory of Tours, Historiarum libri X 2.9 (edd. Krusch, B. and Levison, W.: MGH, Scr. rer. Merov. 1.1 [2nd ed. Hannover 1937–51] 55): ‘Renatus Profuturus Frigiretus … ait: “Interea Respendial rex Alanorum Goare ad Romanos transgresso, de Rheno agmen suorum convertit… .”’ Google Scholar

⁵ Courtois, , Les Vandales 49ff.Google Scholar

⁶ Although reputable toponomists such as Longnon, Auguste, Les Noms de lieu de la France (Paris 1920–28) 133, no. 534; Vincent, Auguste, Toponymie de la France (Brussels 1937) 30; and Dauzat, A. and Rostaing, Ch., Dictionnaire étymologique de noms de lieux en France (Paris 1961) 8–10, accepted these place-names as being of Alan provenance, historians have found it difficult to establish why the settlements were founded. It seems most likely that the Alans were settled in the above-mentioned places in 407 as a result of Goar's alliance with the empire. Cf. Gröhler, H., Über Ursprung und Bedeutung der französischen Ortsnamen (Heidelberg 1913–35) II 7 and 295.Google Scholar

For modern spelling of the Alan place-names see Dictionnaire des Communes (Paris 1961) 6–7.Google Scholar

Bournatzeff, Alexandre, Noms toponymiques des lieux historiques en France, Oss-Alains (Clamart 1953) 21–22, has suggested that many toponymics are of Alan provenance which heretofore have not been considered as such. For example, he claims that the name of the Arandon river (Isère) is of Alan provenance: aran- meaning frontier and -don, river. Longnon, op. cit. 334, no. 1365, has shown that aran is of Basque provenance and means valley. In general I have been unable to accept Bournatzeff's conclusions where they differ from the standard works cited above. Vernadsky, ‘Eurasian Nomads’ 29–30, accepts Bournatzeff's arguments without critical comment. I have had access to Bournatzeff's work only through Vernadsky's liberal use of it.Google Scholar

⁷ Grenier, Albert, Manuel d'archéologie gallo-romaine I (Paris 1931) 398ff.Google Scholar

⁸ Stein, Ernst, Histoire du Bas-Empire I, tr. Palanque, J. R. (Paris 1959) 264.Google Scholar

⁹ Olympiodorus, frg. 17 (ed. Müller, C.: Fragmenta historicorum Graecorum 4 [Paris 1868] 61. Stein, , Bas-Emp. I 264, 558, and Grégoire, Henri, ‘Où en est la question des Nibelungen?’ Byzantion 10 (1935) 219. Cf. Stroheker, K. F., ‘Studien zu den historisch-geographischen Grundlagen der Nibelungendichtung,’ Deutsche Vierteljahrschrift für Literaturwissenschaft und Geistesgeschichte 32 (1958) 216–240.Google Scholar

¹⁰ Stein, , Bas-Emp. I 264.Google Scholar

¹¹ Paulinus of Pella, Eucharisticos 383–5 (ed. Brandes, W.: CSEL 16 [Vienna 1888] 306): ‘Vallanturque urbis pomeria milite Alano, / acceptaque dataque fide certare parato / pro nobis, nuper quos ipse obsederat hostis.’ Google Scholar

Levison, , ‘Bischof Germanus’ 135–36, argues that Goar was the leader of the Alan band with which Paulinus dealt at Bazas. He writes: ‘Ein anderer Alanenführer in Gallien ist aus dieser Zeit nicht bekannt, und der Schluss ist kaum abzuweisen, dass der römisch gesinnte Freund des Paulinus eben Goar gewesen ist.’ See also Vita Germani episcopi Autissiodorensis auctore Constantio (ed. Levison, W.: MGH, Scr. rer. Merov. 7 [Hannover and Leipzig 1920] 271 n. 5). Seeck, Otto, ‘Goar,’ RE 7.2.1547, only hesitatingly admits the possibility that Goar was the leader of the Alans at Bazas. Courtois, Les Vandales 47 n. 3, completely rejects Levison's position.Google Scholar

Levison's position is based in part upon an argument from the silence of the evidence and in part upon the assumption that the Alans of Gaul were a single group under a single leader, i. e., Goar. Vernadsky, ‘Eurasian Nomads’ 23ff., regards the Alans in Gaul as a unified group far greater in size and strength than the evidence would seem to permit. Thus he accepts Bournatzeff's questionable place-name attributions (see note 6 above) and regards the Romano-Alan agreement of 414 as responsible for driving the Visigoths from Gaul. The Visigothic withdrawal from Gaul is discussed much more convincingly by Thompson, ‘Barbarians in Southern Gaul’ 67ff.Google Scholar

If Goar and his followers had left northeastern Gaul and settled in southern Gaul (see note 12 below), then the places in which the Alans had settled in the northeast would not have retained their Alan names. It is possible that Goar did go south with some of his followers, but an argument from the silence of the sources may also be used to contend that he was not the leader of the group at Bazas. By 414 Goar was well known in Gaul. He had joined the Romans in 407 and had supported Jovinus in 411. If Paulinus, who dealt with the Alans, had negotiated with Goar, it seems probable that he would have mentioned the chief by name. It would seem that the Alans who accompanied Ataulf at Bazas were one of the many Alan groups which accompanied the Visigoths after the coming of the Huns, and not Goar's band. For example, Marcellinus, Ammianus, Rerum gestarum libri 31.12.17 (ed. Clark, C. U. [Berlin 1910–15] 589), notes that at the battle of Adrianople in 378 a band of Alan cavalry fought alongside the Visigoths. For other instances of Visigothic-Alan connections, see Ganshof, ‘Note sur le sens de “Ligeris”’ 118.Google Scholar

¹² Paulinus, , Euch. 377–8 (p. 305): ‘Concurrit pariter cunctis ab sedibus omnis / turba Alanarum armatis sociata maritis.’ Also concerning the Alans, lines 395–8 (p. 306): ‘Cuius non sero secuti / exemplum et nostri, quos diximus, auxiliares / discessere, fidem pacis servare parati / Romanis, quoquo ipsos sors oblata tulisset.’ See Lot, F., ‘Du régime de hospitalité,’ Revue beige de philologie et d'histoire 7 (1928) 1007 n. 6.Google Scholar

¹³ For Alan (Haute-Garonne) and Alaigne (Aude) see the toponomical works cited in note 6 above. Neither of these towns is very near Bazas, and it is clear from the text cited in note 12 above that the Alans did not remain in the immediate area of Bazas.Google Scholar

¹⁴ Chronica Gallica 124 (ed. Mommsen, Th.: MGH, Auct. ant. 9 [Berlin 1892] 660); ‘Deserta Valentinae urbis rura Alanis, quibus Sambida praeerat partienda traduntur.’ Bury, J. B., The Later Roman Empire I (London 1923) 293, seems to imply that this colony was in the Orléanais.Google Scholar

¹⁵ Dauzat, and Rostaing, , Dict, étymologique 10. Vernadsky, , ‘Eurasian Nomads’25, claims that Alaigne (Aude) was a part of the colony established at Valence.Google Scholar

¹⁶ Chron. Gallica 127 (p. 660): ‘Alani, quibus terrae Galliae ulterioris cum incolis dividendae a patricio Aetio traditae fuerant, resistentes armis subigunt et expulsis dominis terrae possessionem vi adipiscuntur.’ For the location of this settlement see Levison, , ‘Bischof Germanus’ 136; Ganshof, , ‘Note sur le sens de “Ligeris”’ 116; and Stein, , Bas-Emp. I, 580. Hodgkin, Thomas, Italy and her Invaders, 2nd ed., II (Oxford 1892) 111, confuses the settlement of the Orléanais with that at Valence. Seeck, Otto, Geschichte des Untergangs der antiken Welt VI (Stuttgart 1910) 128, follows Hodgkin.Google Scholar

For the Alan settlements in this area see the toponomical works cited in note 6 above, and Jordanes, , Getica 37 (ed. Mommsen, Th.: MGH, Auct. ant. 5.1 [Berlin 1882] 108).Google Scholar

¹⁷ Vita Germani 28 (p. 272). If Goar came to this new area with all of his followers and completely deserted the settlements in northeastern Gaul, then it is unlikely that the Alan place-names in that area would have survived.Google Scholar

¹⁸ Thompson, , Attila 125ff.Google Scholar

¹⁹ Thompson, , ‘Barbarians in Southern Gaul’ 71.Google Scholar

²⁰ The Tractus Armoricani (Armorica) was a huge area encompassing five provinces and not merely the coast of Brittany. See Notitia Dignitatum 37.24–29 (ed. Seeck, Otto [Berlin 1876] 205): ‘Extenditur tamen tractus Armoricani et Nervicani limitis per provincias quinque: per Aquitanicam primam et secundam, Lugdunensem Senoniam, secundam [Lugdunensem] et tertiam.’ See also Wallace-Hadrill, , Long-Haired Kings 29.Google Scholar

²¹ Vita Germani 28 (pp. 271–73). For the date see Thompson, E. A., ‘A chronological note on St. Germanus of Auxerre,’ Analecta Bollandiana 75 (1957) 136–37.Google Scholar

²² Thompson, , ‘Barbarians in Southern Gaul’ 71.Google Scholar

²³ Jordanes, , Getica 37 (p. 108): ‘Sangibanus … rex Alanorum … pollicetur … Aurelianam civitatem Galliae, ubi tunc consistebat in eius [Attilae] iura transducere’; see also 38 (p. 109): ‘dextrum itaque cornum cum Vesegothis Theoderidus tenebat, sinistrum Aetius cum Romanis, conlocantes in medio Sanguibanum, quem superius rettulimus praefuisse Alanis… . ’ Google Scholar

²⁴ Gregory of Tours, Hist. 2.7 (p. 50) and Additamenta ad Prosperum Hauniensia, a. d. 453 (ed. Mommsen, Th.: MGH, Auct. ant. 9 [Berlin 1892] 302). It is unlikely that Thorismund brought the city of Orleans or the Alan settlements to the north of it under his control. If Orleans had fallen to the Visigoths at this time either a contemporary chronicler or Gregory of Tours would surely have mentioned it. It seems that there were some Alan settlements on the left bank of the Loire and that these were the places which Thorismund brought under Visigothic control. Jordanes, , Getica 43 (p. 115), contrasts ‘Alanorum partem trans flumen Ligeris considentem’ to the settlements of the Orléanais. See also Hodgkin, , Italy and Her Invaders II 169–70, and Levison, , ‘Bischof Germanus’ 136 n. 7.Google Scholar

²⁵ Apollinaris, Sidonius, Carmina 5.475 (ed. Lütjohann, C.: MGH, Auct. ant. 8 [Berlin 1887] 199).Google Scholar

²⁶ Jordanes, , Getica 45 (p. 118); Hodgkin, , Italy and Her Invaders II 440.Google Scholar

²⁷ Vernadsky, , ‘Eurasian Nomads’ 23, implies that Goar intended to set up an Alan empire in Gaul with Jovinus as a figurehead to give it legitimacy. It is clear, however, that the Alans were only one of several groups which supported Jovinus. See Gregory of Tours, Hist. 2.9 (p. 56), again quoting Renatus Profuturus Frigiretus: ‘… nuntii veniunt, Iovinum … cum Burgundionibus, Alamannis, Francis, Alanis … inminere… .’ Even if they wanted to, the Alans in Gaul were too few in number and too disunited to build an empire. This may be seen from the extent of their settlement as indicated by place-names, unless of course Bournatzeff's claims are accepted.Google Scholar

²⁸ Vita Germani 28 (p. 272): ‘… medioque interprete… .’ After having lived in Gaul for almost four decades, Goar still had not learned to speak Latin.Google Scholar

²⁹ Chron. Gallica 127 (p. 660): ‘Alani, quibus terrae Galliae ulterioris cum incolis dividendae a patricio Aetio traditae fuerant, resistentes armis subigunt et expulsis dominis terrae possessionem vi adipiscuntur.’ In this passage resistentes admits of two interpretations: either the remaining landholders were driven out by the Alans, or the resisting landholders were driven out. If the former interpretation is accepted, then the Alans seem to be guilty of greedily violating Aetius' arrangements; but if the latter interpretation is accepted, the landholders illegitimately resisted Aetius' arrangements and the Alans merely carried out the directions of the Roman authority by force of arms. The latter interpretation seems more reasonable since the essence of the land division indicated by the Chron. Gallica, loc. cit., must be understood as a sharing of the estates in which no land-holders would have been forced to leave initially; thus the reading of remaining land holders makes no sense since there can only be a remainder when there is a previous departure. Lot, Both, ‘Hospitalité’ 1010, and Ganshof, , ‘Note sur le sens de “Ligeris”’ 117, accept the former interpretation. But against them see Stein, , Bas-Emp. I 331.Google Scholar

³⁰ Jordanes, , Getica 38 (p. 109).Google Scholar

³¹ Ibid. , 37 (p. 108).Google Scholar

³² Gregory of Tours, Hist. 2.7 (p. 48), and Vita Aniani episcopi Aurelianensis 7–11 (ed. Krusch, B.: MGH, Scr. rer. Merov. 3 [Hanover 1896] 112–117).Google Scholar

³³ Carmina 7.328–56 (pp. 211–12).Google Scholar

³⁴ Epistolae 8.15.1 (p. 147). Thompson, , Attila 140, accepts Jordanes' version of the story.Google Scholar

³⁵ Groups of Alans fought in the Roman army under Gratian, Theodosius I, and Stilicho. See the discussion of these in Jones, , Later Roman Empire I 158, 159, and 183.Google Scholar

³⁶ Marcellinus, Ammianus 31.2.17–25 (pp. 560–562).Google Scholar

³⁷ See notes 12, 14 and 16 above, and Thompson, , ‘Barbarians in Southern Gaul’ 66. Ganshof, , ‘Note sur le sens de “Ligeris”’ 117, follows Lot, ‘Hospitalité’ 1010, in considering the Alans ‘trop sauvages’ to take part in the host arrangement. To support their arguments both writers rely upon the characterization of the Alans by Salvian, , De gubernatione Dei 4.14 (ed. Halm, C.: MGH, Auct. ant. 1.1 [Berlin 1877] 49): ‘rapacitas Alani,’ and Vita Germani 28 (p. 272): ‘… quae ille aviditate barbaricae cupiditatis inhiaverat.’ Yet these passages indicate, if indeed they are more than the moralizing rhetoric of churchmen, that the Alans were greedy rather than ‘anti-social’ or ‘wild.’ Ganshof's phrase, op. cit. 117 n. 3, ‘sur la sauvagerie des Alains …,’ seems to indicate that he is using ‘sauvage’ to mean anti-social, but rapacious Alans rather than anti-social Alans seems to be a better interpretation of the sources.Google Scholar

³⁸ Jordanes, , Getica 37 (p. 108): 'Sangibanus … rex Alanorum … Aurelianam civitatem Galliae ubi tunc consistebat.Google Scholar

³⁹ Vita Germani , 28 (p. 272): ‘Iam progressa gens fuerat totumque iter eques ferratus impleverat… .’ Google Scholar

⁴⁰ For the Roman cataphracts see Marcellinus, Ammianus 31.2. 21, 22 (pp. 561ff.), and for the Gallo-Roman magnates see Apollinaris, Sidonius, Epistulae 3.3.3, and 5.12.1 (pp. 41 and 86). Cf. Rattenbury, R. M., ‘An Ancient Armored Force,’ Classical Review 56 (1942) 113–16.Google Scholar

⁴¹ Jordanes, , Getica 50 (p. 125): ‘… Alanum gravi … armatura aciem strui.’ Google Scholar

⁴² Ibid. 38 (p. 109). Thompson, , Attila 142 n. 1, conservatively estimates Attila's army at 30,000 warriors. Aetius' forces could not have numbered much less, and the Alans holding the center of the line must have been at least a quarter if not a third of the entire force; somewhere between 4,000 and 7,000 Alan warriors is a minimum estimate.Google Scholar

⁴³ Lot, , ‘Hospitalité’ 977, shows that the Burgundians, who were settled by Aetius at about the same time that he settled the Alans, received a share in the slaves as well as a share in the land. We know more about the terms of the Burgundian settlement because their law code has survived. In the Orléanais, where the Alans drove out some of the landowners, the agricultural dependents of these men were probably taken over by the Alans.Google Scholar

⁴⁴ Salvian, , De gub. Dei 4.14 and 7.15 (pp. 49 and 95), and Vita S. Germani 28 (p. 272).Google Scholar

⁴⁵ Salvian, , De gub. Dei 4.14 (p. 49), and Vita S. Germani 28 (p. 272).Google Scholar

⁴⁶ Vita Goaris Confessoris Rhenani 1 (ed. Krusch, B.: MGH, Scr. rer. Merov. 4 [Hanover 1902] 411): ‘In diebus Childeberti regis Francorum, filio Chlodoveo, erat vir venerabilis nomine Goar, homo Aquitanicus, cuius pater vocatur Georgius et mater eius Valeria.’ There has been some discussion concerning the adoption of Roman names by barbarians. Fustel de Coulanges, ‘De l'analyse des textes historiques,’ Revue des questions historiques 41 (1887) 12, argues for such adoption, while Kurth, G., Études franques I (Paris 1919) 253–54, argues against it. This discussion, however, was limited to the Frankish use of Roman names and does not vitally concern the problem being studied here, which concerns Iranians and not Germans. Ammianus Marcellinus 31. 12. 6 (p. 586) notes that the fourth-century cavalry commander with the Roman name Victor was a Sarmatian. If Goar's parents had not been Christian, the author of his Vita would surely have noted it.Google Scholar

⁴⁷ Fortunatus, Venantius, In laudem sanctae Mariae 289–91 (ed. Leo, F.: MGH, Auct. ant. 4.1 [Berlin 1881] 378). On the genuineness of this poem, edited by Leo as spurious, see the literature cited by Dekkers, E., Clavis patrum Latinorum (2nd ed. Sacris erudiri 3 [1961]) no. 1044a. - In an inscription from Visigothic Spain dedicated by St. Martin of Braga to St. Martin of Tours and dating from the second third of the sixth century the Alans are included (line 14) in a list of peoples who are said to have been converted to Christianity by St. Martin of Tours (Vives, José, Inscripciones cristianas de la España romana y visigoda [Barcelona 1942] 119–20; more recently, Barlow, C. W., ed., Martini ep. Bracaren. opera omnia [Papers and Monographs of the American Academy in Rome 12; New Haven 1950] 282). Both this list and the one made by Fortunatus which is discussed above may be considered fanciful. Despite this putative fancy, however, these lists do contain a kernel of truth. Neither Martin of Braga nor Fortunatus would have included people in a list of those who were converted to Christianity if they were still known to be pagans, because if they were still pagans and it was known, the saint would not get the credit for their conversion. Thus it may be concluded that any of the peoples mentioned in these lists who were known either to the authors or to their contemporaries in Spain or Gaul were Christians. It is of course impossible for Martin of Tours († 397) to have converted the Alans, but by the mid-sixth century the Alans in both Gaul and Spain must have been Christians. The reference to Vives and the warning concerning the fanciful nature of both lists I owe to Professor Maenchen.Google Scholar

⁴⁸ Dauzat, A., Dictionnaire étymologique des noms de famille et prénoms de France (Paris 1951) 4–6 and 295.Google Scholar

⁴⁹ Procopius, , History of the Wars 5.12.12ff. (trans. Dewing, H. B., Loeb Classical Library, Procopius III [London 1919] 121): ‘… the Arborychi had become soldiers of the Romans. And the Germans, wishing to make this people subject to themselves, since their territory adjoined their own and they had changed the government under which they had lived from of old, began to plunder their land and, being eager to make war, marched against them with their whole people. But the Arborychi proved their valour and loyalty to the Romans and shewed themselves brave men in this war, and since the Germans were not able to overcome them by force, they wished to win them over and make the two peoples kin by intermarriage.’ Google Scholar

For the Arborychi as the Armoricans, see the works cited by Lot, F., ‘La conquête du pays d'entre Seine et Loire par les Francs,’ Revue historique 165 (1930) 241–42. Yet Lot himself, op. cit. 244, believes that Arborychi means aborigines.Google Scholar

There are several interpretations of the conquest of the area between the Seine and the Loire. Kurth, G., Clovis (2nd ed. Paris 1901) I 257–59, seems to feel that Frankish and other military colonists controlled the region and went over to Clovis. Lot, op. cit., also feels that the area was infiltrated by Franks before Clovis. I think both these writers overestimate the number and strength of the Franks in this area before Clovis. Van de Vyver, A., ‘Clovis et la politique méditerranéenne,’ Études d'histoire dediées à la mémoire de Henri Pirenne … (Brussels 1937) 370, 372, and Wallace-Hadrill, , Long-Haired Kings 165, remain closer to Procopius' description of events and seem more convincing.Google Scholar

⁵⁰ Hist, of the Wars 5.12.14 (p. 121).Google Scholar

⁵¹ Ibid. , 5.12.15 (p. 121).Google Scholar

⁵² Van de Vyver, , ‘Clovis’ 370, 273, and Wallace-Hadrill, , Long-Haired Kings 165.Google Scholar

⁵³ Gregory of Tours, Hist. 2.37 (p. 85): ‘ut nullus de regione illa aliud quam herbarum alimenta aquamque praesumeret.’ Google Scholar

⁵⁴ This problem is discussed by White, Lynn Jr., Medieval Technology and Social Change (Oxford 1962) 8–9.Google Scholar