¹ MGH Capitularia I 126–30 no. 45. Halphen, Louis, Charlemagne et l'Empire carolingien (Paris 1949) 136–38; Fichtenau, Heinrich, The Carolingian Empire, tr. Peter Munz (Oxford 1957) 178, 185–86; Ganshof, François-Louis, Frankish Institutions under Charlemagne, tr. Lyon, B. and Lyon, M. (Providence 1968) 16–17; Bullough, Donald, The Age of Charlemagne (London 1966) 197. These accounts tend to be more concerned with the future of the imperial dignity than with the succession as such.Google Scholar

Early planning of successions was not, apparently, an established practice. Charles's father and grandfather divided their realms only in the years of their deaths: Böhmer–Mühlbacher, , Regesten des Kaiserreichs under den Karolinger, 751–918 (2d ed.; Innsbruck 1908) nos. 42a, 106c.Google Scholar

² Classen, Peter, ‘Karl der Grosse und die Thronfolge im Frankenreich,’ Festschrift für Hermann Heimpel III (Göttingen 1972) 109–34. Classen's teaching is endorsed by Ewig, Eugen, ‘Überlegungen zu den merowingischen und karolingischen Teilungen,’ Settimane di studio dell Centro italiano di studi sull' alto medioevo 27 (1981) 240–41, and Beumann, Helmut, ‘Unitas ecclesiae–unitas imperii–unitas regni: Von der imperialen Reichseinheitsidee zur Einheit der Regna,’ ibid. 537–38. It is opposed by Hägermann, Dieter, ‘Reichseinheit und Reichsteilung. Bemerkungen zur Divisio regnorum von 806 und zur Ordinatio imperii von 817,’ Historisches Jahrbuch 95 (1975) 278–307. Hägermann plays the useful part of advocatus diaboli in striving to show that Charlemagne instituted no measures anticipating the act of 817, for he fails to detect any major flaw in Classen's argument.Google Scholar

³ Classen, , ‘Karl der Grosse u. d. Thronfolge’ 109.Google Scholar

⁴ MGH Capitularia I 270–73 no. 136. Halphen, , Charlemagne 236–39; Ganshof, François-Louis, ‘Observations sur l'Ordinatio imperii de 817,’ Festschrift Guido Kisch (Stuttgart 1954) 15–31.Google Scholar

⁵ Classen, , ‘Karl der Grosse u. d. Thronfolge’ 114–17. The kingmaking: Böhmer–Mühlbacher, Regesten ² no. 235b. For the annals, MGH Scriptores I 12; followed by ‘et perrexit ad Romam.’ The date is given as 780, but every previous entry is also off by one year in relation to our reckoning. The entry was considered exaggerated by Eiten, Gustav, Das Unterkönigtum im Reiche der Merovinger und Karolinger (Heidelberger Abhandlungen zur mittleren und neueren Geschichte 18; Heidelberg 1907) 22, and Böhmer–Mühlbacher, Regesten ² no. 235b.Google Scholar

⁶ Einhard, , Vita Karoli 20 (ed. Rau, Reinhold; Quellen zur karolingischen Reichsgeschichte 1 [Berlin 1955] 190); ‘nanus et gibberosus,’ Notker, , Gesta Karoli 2.12 (ed. Rau, R.; Quellen zur karolingischen Reichsgeschichte 3 [Berlin 1960] 402). The deformity is not, I believe, attested before Einhard. See also Fichtenau, , Carolingian Empire 39.Google Scholar

⁷ ‘Karl der Grosse u. d. Thronfolge’ 117–18.Google Scholar

⁸ Charlemagne's planned journey, in connection with the pope's baptizing his second son by Hildegard, born in 777, is referred to by Hadrian, Pope, Codex Carolinus 60 (MGH Epistolae III 586 lines 24–36). But Charles's plans for the Roman visit may have developed between the time when it was first projected and its actual occurrence.Google Scholar

⁹ Bullough, , Age of Charlemagne 60; Ganshof, François-Louis, ‘Une crise dans le règne de Charlemagne, les années 778 et 779,’ in Mélanges d'histoire et de littérature offerts à M. Charles Gilliard (Lausanne 1944) 133–48.Google Scholar

¹⁰ MGH Capitularia I 46–51 no. 20.Google Scholar

¹¹ Classen, , ‘Gregor der Grosse u. d. Thronfolge’ 118 stressed that Charlemagne's plan for the core kingdom was not yet given legal form.Google Scholar

¹² For the details, ibid. 109–10.Google Scholar

¹³ Böhmer–Mühlbacher, , Regesten ² no. 320a.Google Scholar

¹⁴ Supra n. 8. Both Stephen II and Paul I had godfathered children of Pepin's and addressed him as ‘spiritalis compater’: Codex Carolinus 6–11, 14 (MGH Epistolae III 488–504, 511). On the liturgical season for baptism, see de Puniet, P., DACL 2 (1925) 303, 311, 329–30, 331–32, s.v. ‘baptême.’ On the Gospel Book, see Bullough, , Age of Charlemagne 99, and Schreiner, Klaus, “‘Hildegardis regina”: Wirklichkeit und Legende einer karolingischen Herrscherin,’ Archiv für Kulturgeschichte 57 (1975) 9–10.Google Scholar

¹⁵ ‘Karl der Grosse u. d. Thronfolge’ 115–16.Google Scholar

¹⁶ Classen, Peter, ‘Karl der Grosse, das Papsttum und Byzanz,’ in Beumann, Helmut, ed., Karl der Grosse I, Persönlichkeit und Geschichte (Düsseldorf 1965) 557 n. 81. The same idea had been expressed in older literature; for example, Hodgkin, Thomas, Italy and Her Invaders VIII (Oxford 1899) 53, in which the older Pepin's exclusion is attributed to his deformity.Google Scholar

¹⁷ ‘Karl der Grosse u. d. Thronfolge’ 115, 117. The three texts are Paul the Deacon's Gesta (n. 93 infra); the Royal laudes at Soissons (PL 138.885–88); and the Confraternity Book of St. Peter's Salzburg (MGH Necrologiae Germaniae II 12). They establish that Pepin continued to rank alongside Charles the Younger as a son of the king, but nothing more. The reference to him in Paul's Gesta involves an explicit contrast to the sons by Hildegard, , as we shall see.Google Scholar

¹⁸ For Charles's, sub-kingdom, see Eiten (n. 5 supra) 46–47. Unless a damnatio memoriae after 792 is assumed, the failure of the sources to record the award of responsible tasks to Pepin, or other marks of distinction (such as a wife), cannot be without significance for his prospects.Google Scholar

¹⁹ Bethmann, Ludwig, ‘Paulus Diaconus Leben und Schriften,’ Archiv der Gesellschaft für ältere deutsche Geschichtskunde 10 (1851) 254–78; Menghini, Evelina, ‘Dello stato presente degli studi intorno alla vita di Paolo Diacono,’ Bolletino della Società pavese di storia patria 4 (1904) 15–100, 231–85, 313–66; Neff, Karl, Die Gedichten des Paulus Diaconus (Quellen und Untersuchungen zur lateinischen Philologie des Mittelalters 3, H. 4; Munich 1908). On his relations to Benevento, Krüger, Karl Heinrich, ‘Zur “beneventanischen” Konzeption der Langobardengeschichte des Paulus Diaconus,’ Frühmittelalterliche Studien 15 (1981) 19–21. The commonly accepted date for the beginning of his northern sojourn is late 782. It presupposes that his famous petition to Charlemagne on behalf of his brother (Neff, , Gedichten no. 11, May 782) was composed in Italy and sent to Charlemagne from there. If, as seems probable, Paul drew up the petition when already at the Frankish court, his journey may be set back by as much as a year.Google Scholar

²⁰ The full title is Libellus de numero vel ordine episcoporum qui sibi ab ipso praedicationis exordio in Mettensi civitate successerunt (ed. Pertz, G. H.; MGH Scriptores II 261–68). The text closes with a brief address to Angilram (n. 37 infra). See also Paul the Deacon, Historia Langobardorum 6.16 (edd. Bethmann, L. and Waitz, G.; MGH Scriptores rerum Langobardicarum 170): ‘in libro quem de episcopis eiusdem civitatis conscripsi flagitante Angilramno, viro mitissimo et sanctitate praecipuo, praefatae ecclesiae archiepiscopo.’ Google Scholar

²¹ Werner, Karl Ferdinand, ‘Bedeutende Adelsfamilien im Reich Karls des Grossen,’ in Karl der Grosse I 119 n. 133; Fleckenstein, Josef, ‘Karl der Grosse und sein Hof,’ in ibid. 35. Fulrad was inconspicuous from 780 onward; Angilram, overshadowed by Alcuin, never equalled his predecessor's influence.Google Scholar

²² For no stated reason, the date ‘about 785’ has lately been attached to Paul's Gesta : Oexle, Otto Gerhard, ‘Die Karolinger und die Stadt der heiligen Arnulf,’ Frühmittelalterliche Studien 1 (1967) 274; Jäschke, Kurt Ulrich, ‘Die Karolingergenealogien aus Metz und Paulus Diaconus,’ Rheinische Vierteljahresblätter 34 (1970) 207. Composition of the Gesta in 785 and no earlier seems crucial to the interpretation of Karl Hauck, ‘Die Ausbreitung des Glaubens in Sachsen und die Verteidigung der römischen Kirche als konkurrierende Herrscheraufgaben Karls des Grossen,’ Frümittelalterliche Studien 4 (1970) 147–60; no justification is given for the date. That Paul wrote the Gesta then rather than earlier is plausible if Paul is assumed to have returned to Italy in Charlemagne's entourage late in 786, as proposed, though only as a possibility, by Bethmann, , ‘Paulus Diaconus’ 266; Neff, , Gedichten 132; Hodgkin, Thomas, Italy and Her Invaders V (2d ed.; Oxford 1916) 78. But this chronology has little in its favor. The last event mentioned in the Gesta is Charlemagne's marriage to Fastrada (before October 783); no reference is made to Angilram's becoming archchaplain; Paul's presence in the north after Charles's remarriage cannot be substantiated (it is unlikely that the Gesta could have been completed earlier than 784). A papal letter of ca. 785–86 (Codex Carolinus 89, MGH Epistolae III 626) indicates that, some time before, Paul transmitted to the pope a request of Charlemagne's; see Vogel, Cyrille, ‘La reforme liturgique sous Charlemagne,’ in Bischoff, B., ed., Karl der Grosse II, Das geistige Leben (Düsseldorf 1965) 224–25. Paul was definitely at Monte Cassino from 787 onwards: Neff, Gedichten 143. Either Paul prolonged his northern sojourn without leaving any concrete trace or, as seems more likely, he returned to Italy as early as 784, and no later than the spring of 785.Google Scholar

²³ Paul's simplifying distortion of the Carolingian genealogy is generally overlooked; for example, Oexle (n. 22 supra) 272, and Sot, Michel, ‘Historiographie episcopate et modele familial en Occident au ix^e siècle,’ Annales: Économies, Sociétés, Civilisations 33 (1978) 439–41.CrossRefGoogle Scholar

²⁴ Vinay, Gustavo, ‘Paolo Diacono e la Poesia. Nota,’ Convivium n.s. 1 (1950) 97–113; Bognetti, Gian Piero, ‘Processo logico e integrazione delle fonti nella storiografia di Paolo Diacono,’ in Bognetti, , L'Età longobarda III (Milan 1967) 159–84; Sestan, Ernesto, ‘La storiografia dell'Italia Longobarda, Paolo Diacono,’ Settimane di studio dell Centro italiano di studi sull'alto medioevo 17 (1970) 357–86; Brunhölzl, Franz, Geschichte der lateinischen Literatur des Mittelalters I (Munich 1975) 257–68.Google Scholar

²⁵ The Vita must be consulted in the edition of Walter Stuhlfath, Gregor I. der Grosse: Sein Leben bis zu seiner Wahl zum Papste nebst einer Untersuchung der ältesten Viten (Abhandlungen, Heidelberger zur mittleren und neueren Geschichte 39; Heidelberg 1913) 98–108.Google Scholar

²⁶ As an expansion of Eutropius, ed. Droysen, H., MGH Auctores antiquissimi II; in its own right, ed. Crivellucci, Amadeo, Fonti per la storia d'Italia 51 (Rome 1914). Because Paul signs himself ‘exiguus et supplex’ (as he would in later life), he seems to have already been a monk of Monte Cassino. The work is assumed, probably rightly, to date from before the fall of the Lombard kingdom (774).Google Scholar

²⁷ I shall fully discuss this and other features of Paul's narratives elsewhere. The most characteristic examples of bipartite exposition are: Historia Romana 12.1–8/9–17; 13.1–8/9–18; 15.1–10/11–20; 16.1–10/11–23; Vita Gregorii Magni 1–9/10–17; Historia Langobardorum 1.1–15/16–22 3.1–15/16–35, 4.1–29/30–51. It is arguable that the unfinished Historia Langobardorum reaches mid-point at the end of book four.Google Scholar

²⁸ A string of brief references to Britain in the Romana seems to be designed as miniature foreshadowings of what, in the next book, happens in the heart of the Roman world. In this way, Hist. Rom. 13.17 anticipates book 14.Google Scholar

²⁹ Bullough, , Age of Charlemagne 104; Löwe, Heinz in Wattenbach–Levison, Deutschlands Geschichtsquellen im Mittelalter. Vorzeit und Karolinger H. 2 (Weimar 1953) 217–18; cf. Jäschke, (n. 22 supra) 207; Sot, Michel, Gesta episcoporum, gesta abbatum (Typologie des sources du moyen âge occidental 31; Turnhout 1981) 33–34; Grundmann, Herbert, Geschichtsschreibung im Mittelalter (2d ed.; Göttingen 1965) 39. A special link between Angilram and the Liber pontificalis was suggested by Buchner, Max, ‘Zur Überlieferungsgeschichte des Liber pontificalis und zu seiner Verbreitung im Frankenreich im 9. Jahrhundert,’ Römische Quartalschrift 34 (1926) 153–54.Google Scholar

³⁰ Liber pontificalis, ed. Duchesne, Louis (Paris 1886). For Tours, see Gregory of Tours Historiae 10.31 (edd. Krusch, Bruno and Levison, Wilhelm; MGH Scriptores rerum Merovingicarum I [2d ed.] 526–36). If Gregory's deacon who witnessed Gregory the Great's elevation to the Holy See brought home an early version of the papal Liber, as he might, Gregory of Tours would have a rather better claim than Paul to composing the first northern derivative. For editions of the other early episcopal and abbatial gesta, see Sot (n. 29 supra) 13–14 nn. 4–7.Google Scholar

³¹ Cf. Bullough, , Age of Charlemagne 104.Google Scholar

³² Versus de episcopis Mettensibus, MGH Poetae Latini I 60–61; Neff, , Gedichten 188–90. On their attribution, Löwe (n. 29 supra) 218 n. 166; Jäschke, Kurt Ulrich, ‘Zu Metzer Geschichtsquellen der Karolingerzeit,’ Rheinische Vierteljahresblätter 33 (1969) 11–13; Brunhölzl, , Gesch. d. lat. Lit. I 261. The Versus differ in significant ways from Paul's Gesta. It is easy to believe that Paul did not compose them, less so to accept that Paul's prose was their source. I find merit in Neff's idea (p. 186) that Angilram caused the verse catalogue to be written and that it was Paul's source, but am open to other possibilities.Google Scholar

³³ Gesta epp. Mett. (262 lines 1–3, 264 lines 4–5, 267 lines 39–41 Pertz); Sigibald was Chrodegang's predecessor. The longest of these minor expansions concerns a miracle, perhaps quite recent, at the grave of Rufus and Adolfus, the ninth and tenth bishops (262 lines 8–20).Google Scholar

³⁴ Gesta epp. Mett. (264 lines 35–36 Pertz, ): ‘quod ita coelesti esset iudicio definitum, ut universa Gallia barbarorum foret infestationi tradenda.’ Google Scholar

³⁵ The Liber pontificalis has nothing comparable to Paul's many unadorned names. Even the briefest entries (for example, Linus, Cletus, Anecletus, Evaristus) are miniature counterparts of long ones, with the same formulaic structure.Google Scholar

³⁶ Although Paul freely adapts the formulary of the Liber pontificalis, the similarity is recognizable at first glance, the more so in contrast to Paul's different style of exposition until then. Beginning: ‘Chrodegangus … ex pago Hasbaniensi oriundus, patre Sigramno, matre Landranda, Francorum ex genere primae nobilitatis progenitum’; cf. John, II, ‘Iohannes, natione Syrus, de provintia Antiocha, ex patre Cyriaco …’ (366 Duchesne). Early life: ‘hic in palatio …’; cf. Benedict, II, ‘Hic ab ineunte aetate …’ (363). Talents and virtues: ‘Fuit autem omnino clarissimus …’ (a characteristic opening in the Liber, but unusual in this context); cf. Leo, II (359). Achievements as bishop: 'Hic clerum adunavit… Hic fabricare iussit… Construxit etiam … Construxit etiam … Expetiit denique (more variations of verbs than normal in the Liber pontificalis); cf. Sergius, , repetitively, ‘Hic fecit…’ (374–75). Paul's second list of Chrodegang's virtues is not paralleled by the formulary of the Liber. Ordinations: ‘Hic consecravit episcopos quam plurimos per diversas civitates …’; cf. Sergius, , ‘Hic ordinavit per diversas provincias episcopos …’ (376). Paul then numbers Chrodegang's regnal years, information that the Roman formulary places in the first sentence. The latter also omits the date of death (which Paul provides), replacing it with the date and place of burial (in Paul, only the place).Google Scholar

³⁷ Gesta epp. Mett. (268 Pertz): ‘Hic iam, pater sanctissime Angilramne, narrationis serie vestram beatitudinem locus expectat. Sed ego meae tenuitatis non inmemor, adtemptare minus idonee non audeo, quae de vestrae vitae cursu laudabili maiori stilo promenda sunt.’ Cf. Eutropius, , Breviarium 10.18.3 (182 Droysen): ‘quia autem ad inclitos principes venerandosque perventum est, interim operi modum dabimus. nam reliqua stilo maiore dicenda sunt. quae nunc non tam praetermittimus quam ad maiorem scribendi diligentiam reservamus.’ Eutropius' lines were incorporated into Paul's Historia Romana. Google Scholar

³⁸ Origo gentis Langobardorum (ed. Waitz, G.; MGH Scriptores rerum Langobardicarum 2–6). It contains a spare chronicle of the Lombard kings (mere names in c. 2 and partly in c. 6), together with two developed tales: migration from ‘Scadanan’ (c. 1) and Alboin's murder (c. 5). The alternation of story and genealogical transition in Genesis is as follows (chapter and verse are indicated only for the transitional passages): Creation, Adam and Eve; Adam's descendants, Gen. 4.15–5.31; Noah; his descendants, Gen. 10.1–32; the tower of Babel; Shem's descendants, Gen. 11.10–32; Abraham, ; Ishmael's descendants, Gen. 25.12–18; Isaac, and Jacob, ; Esau's descendants, Gen. 36.1–43; Jacob, and Joseph, .Google Scholar

³⁹ Gen. 27.1–40. Rebecca's trick in the prototype turns in Paul's Gesta into a question of merit: the younger son, by his obedience and charity, earns preferment over the elder. Sot (n. 23 supra) 440 observed the biblical parallel; see also Oexle (n. 22 supra) 275.Google Scholar

⁴⁰ Gesta epp. Mett. (261 lines 1–33 Pertz); Peter, ‘ad eam quae totius tunc mundi caput erat, hoc est urbem Romuleam, tota alacritate contendit’; for the doctores accompanying Clement, see n. 42. The specific tale concerning Metz begins only at line 15.Google Scholar

⁴¹ Sot (n. 23 supra) 439 claims to find in episcopal lists a parallel to pagan divinities in royal genealogies.Google Scholar

⁴² Cf. Hauck, (n. 22 supra) 152, and Brunhölzl, Gesch. d. lat. Lit. I 261. Clement is not mentioned as belonging to Peter's consortium, from which the Italian church founders emanated. Peter is qualified as Clement's praeceptor only as an afterthought (when Clement is already at Metz). We first encounter Clement as a ‘vir egregius et mentis probatum,’ chosen by Peter for the mission and ordained bishop by him, ‘cum quo pariter, sicut antiqua tradit relatio, ad eas quae praecipuae erant Galliarum urbes verbi fidei obtinendas, alii quoque religiosi doctores ab eodem apostolorum principe missi sunt’ (Gesta epp. Mett., 261 lines 17–21 Pertz). A decretal of Innocent I's asserts the Roman evangelization of all the western churches as reason why these churches must observe Roman usages (JK 311). The Carolingian court, owing to its part in propagating Roman usages, had reason to stress Petrine origins even without special reference to Metz. For the wider context of foundation legends, see Émile Amman in Fliche, A. and Martin, V., Histoire de l'Église VII (Paris 1948) 179–81.Google Scholar

⁴³ ‘Ostensor iustitiae,’ Gesta epp. Mett. (261 lines 28–29 Pertz, ). When Clement reached Metz, , ‘in cavernis, ut ferunt, amphitheatri quod extra eandem urbem situm est, hospitium habuit; in quo etiam loco oratorium Domino construens … etc.’ (lines 21–24); ‘Denique asseverant qui eiusdem loci cognitionem habent, quod in amphitheatro, ubi primitus adveniens habitavit, usque in praesentem diem nec serpens consistere queat, sed et omnino noxiae pestes locum ilium refugiant, unde olim verae salutis emanarunt insignia’ (lines 30–33).Google Scholar

⁴⁴ For the northern metropolises, Dorothea Ewert, K., ‘Italia sacra,’ in Poole, Reginald Lane, Historical Atlas of Modern Europe (Oxford 1901) no. 69. In Paul's time, several southern sees in Byzantine territory had metropolitan status, but Brindisi (which became an archbishopric only after the Norman conquest) was Beneventan and presumably claimed foundation by Peter's disciple, Lucius.Google Scholar

⁴⁵ N. 42 supra. Google Scholar

⁴⁶ Capitulare Heristalense 1 (MGH Capitularia I 47).Google Scholar

⁴⁷ Lesne, Émile, La hiérarchie épiscopale. Provinces, métropolitains, primats en Gaule et en Germanie, 742–882 (Paris 1905) 30–79; Eugen Ewig in Jedin, H. and Dolan, J., Handbook of Church History III (Montreal 1969) 71–72.Google Scholar

⁴⁸ The miracles, announced at Gesta epp. Mett. 262 lines 25–27 Pertz, , begin to be related only forty-six lines later.Google Scholar

⁴⁹ Gesta epp. Mett. ‘ea tempestate, quando non solum Gallia sed universus pene occidens barbarorum saevientium est perpessus saevitiam’ (262 lines 24–25 Pertz); ‘Attila rex Hunnorum, omnibus belluis crudelior’ (line 28). Paul's basis for this account was Gregory of Tours, Historiae 2.5–6 (45–48 Krusch–Levison), specifically stressing the unalterability of God's decree. Paul's report of Attila's advance embodies details from his own Historia Romana 14.5 (202 Droysen).Google Scholar

⁵⁰ Seven times in twenty-nine lines (Gesta epp. Mett. 262 lines 23–52 Pertz, ). There is nothing of the sort in Gregory of Tours.Google Scholar

⁵¹ On aquatic imagery, see Goffart, Walter, ‘Rome, Constantinople, and the Barbarians,’ American Historical Review 86 (1981) 283–84. Fredegar, Chronicon 2.37 (ed. Krusch, B.; MGH Scriptores rerum Merovingicarum II 62 line 31) substituted ‘inundatio gentium’ for Jerome's word ‘pestilentia’ as cause for the destruction of the army of Marcus Aurelius.Google Scholar

⁵² To the basic story of Gregory of Tours, Hist. 2.6, Paul added the reference to the protomartyr's blood, an account of the miraculous collapse of the city's walls (ascribed to the Vandal invasion of Gaul by Fredegar Chronicon 2.60, 84 Krusch), and details of the miracle that saved the church from destruction. Gregory of Tours, In gloria martyrum 34 (ed. Krusch, B.; MGH Scriptores rerum Merovingicarum I 508) knew of relics of St. Stephen's blood, but only at Bourges.Google Scholar

⁵³ Auctor's sole response to the news that the walls had collapsed was to order the children's baptism (Gesta epp. Mett. 262 lines 46–48 Pertz, ). His liberation of his fellow-captives and mending of the altar stone are the two miracles Paul announced at the beginning of the narrative.Google Scholar

⁵⁴ Cf. Sot (n. 23 supra) 441.Google Scholar

⁵⁵ Boniface, , Epistolae 50 (MGH Epistolae selectae I 82); Actus pontificum Cenomannis in urbe degentium , edd. Busson, G. and Ledru, A. (Archives historiques du Maine 2; Le Mans 1901) 244–64; Laporte, Jean, ‘Les monastères francs et l'avènement des Pippinides,’ Revue Mabillon 30 (1940) 1–30; Ewig, Eugen, ‘Milo et eiusmodisimiles,’ in Sankt Bonifatius: Gedenkgabe zum zwölfhundertsten Todestag (Fulda 1954) 430–39; Ewig (n. 47 supra) 8–9.Google Scholar

⁵⁶ In Paul's account, the Carolingians descend from Arnulf's son Anschisus (properly, Ansegiselus), ‘cuius Anschisi nomen ab Anchise patre Aeneae, qui a Troia in Italiam olim venerat, creditur esse deductum. Nam gens Francorum, sicut a veteribus est traditum, a Troiana prosapia trahit exordium’ (264 lines 38–40 Pertz, ).Google Scholar

⁵⁷ Grundmann, , Geschichtsschreibung ² 39, suggested that several continuations of Paul's Gesta were composed. See, to the contrary, Hermann, Hans-Walter, ‘Zum Stande der Erforschung der früh- und hochmittelalterlichen Geschichte des Bistums Metz,’ Rheinische Vierteljahresblätter 28 (1963) 134–35: there was no continuation; a new episcopal history was undertaken in the twelfth century. Grundmann's statement (on the same page) that Paul's Gesta served as ‘Vorbild für viele andere Bistumsgeschichten’ has yet to be proved true in even a single instance. The influence on the Actus pontificum Cenomanensium intimated by Sot (n. 29 supra) 34, is on the basis of circumstantial rather than textual evidence.Google Scholar

⁵⁸ Gesta epp. Mett. (264 lines 13–15 Pertz, ). There is no inherent reason why an episcopal historian should not summarize or include extracts from existing material; many later ones did little else. Paul himself quotes Gregory of Tours and his own funerary verse. Vita s. Arnulfi , ed. Krusch, Bruno (MGH Scriptores rerum Merovingicarum II 432–46).Google Scholar

⁵⁹ Fredegar, , Chronicon, Continuationes 6 (ed. Wallace-Hadrill, J. M. [London 1960] 86), Brühl, Carlrichard, ‘Königspfalz und Bischofsstadt in fränkischer Zeit,’ Rheinische Vierteljahresblätter 23 (1958) 240, indicates that Drogo's sons Arnulf and Hugo may have been buried with him, but there is no support for this view in Mühlbacher, E., ‘Zur Genealogie der älteren Karolinger,’ Forschungen zur deutschen Geschichte 19 (1879) 455–64.Google Scholar

⁶⁰ N. 48 supra; unlike Auctor, Arnulf is not shown performing even one miracle (he does, of course, in the Vita).Google Scholar

⁶¹ See the next note.Google Scholar

⁶² Gesta epp. Mett. (264 lines 15–36 Pertz, ): ‘Unum tamen eius admirabile factum referam, quod satis miratus sum quo ordine praeterierit is, qui eius contexuit vitam. Hic denique cum poenitentiam pro aliquibus excessibus ageret, contigit ut per Mosellae fluminis pontem transiret. Cumque subterfluentium aquarum profundas nec visu penetrabiles gurgites cerneret, non dubiae spei fidutiam mente gerens, extractum digito anulum in illam aquarum profunditatem proiecit: “Tunc” inquiens, “me putabo culparum nexibus absolutum, quando istum quem proicio recepero anulum.” Post aliquot vero annos cum ad episcopatus ascendisset officium, ei die quadam piscator quidam piscem attulit, quem ille sibi, quia carnibus abstinebat, ad refectionem vespertinam iussit praeparari. Cumque eum minister eius officii more solito exenterasset, eundem anulum intra ipsius piscis intestinum repperit. Ille factum admiratus, sed rem nesciens, beato Arnulfo detulit. Quem ille ut vidit, statim recognovit, ac Deo omnipotenti, iam de remissione delictorum fidus, gratias retulit, ac deinceps non remissius vitam duxit, sed maiori se potius abstinentia coartavit. Nec dissimiliter pater hic venerabilis quam olim Gedeon ille signum a Domino poposcit. Ille etenim indicium per vellus in area positum capere voluit, utrum in bello victoriam de inimicis habere potuisset; iste anulum in profundissimum fluminis gurgitem proiciens, experiri cupiit, utrum de inimicis iam victoriam cepisset. Fortes quidem erant quos ille devicerat, sed hii quos iste superaverat fortiores. Haec ego non a qualibet mediocri persona didici, sed ipso totius veritatis assertore, praecelso rege Karolo, referente cognovi; qui de eiusdem beati Arnulfi descendens prosapia, ei in generationis linea trinepos extabat.’ Trinepos is one generation too many. Even if slightly wrong, this detail looks ahead to the coming genealogy and suggests that Paul was writing a coherent story rather than detached scenes.Google Scholar

⁶³ Vita Arnulfi 7–8 (435 Krusch, ), generosity and asceticism; 22 (442 Krusch), the saint on his deathbed asserts his sinfulness; 21 (441 Krusch, ), he gave away his property to the poor, without obstacle, only after resigning the bishopric.Google Scholar

⁶⁴ Paul's statement needs to be plain fact in the interpretation of Hauck (n. 22 supra) 154–60; also, less crucially, Sot (n. 23 supra) 440. But a less prosaic reading is also possible. Paul does not claim to quote Charles in the way he quotes Gregory of Tours (Gesta epp. Mett. 263 lines 9–20 Pertz, ).Google Scholar

⁶⁵ Herodotus, Historiae 3.39–43, 120–26.Google Scholar

⁶⁶ Frye, Northrop, The Secular Scripture: A Study of the Structure of Romance (Cambridge, Mass. 1976) 147–49; Künzig, Johannes, ‘Der im Fischbauch wiederaufgefundene Ring in Sage, Legende, Märche und Lied,’ in Schewe, Harry, ed., Volkskundliche Gaben, Johan Meier dargebracht (Berlin-Leipzig 1934) 85–103. Because Künzig makes much of the Solomon legend as prototype for Paul's version, it is worth pointing out that Solomon's ring has magic associations with power (losing it, Solomon becomes a beggar), whereas Arnulf's does not. At first sight, Arnulf's ring seems entirely neutral.Google Scholar

⁶⁷ Doyle, Arthur Conan, ‘Silver Blaze,’ in Baring-Gould, William S., ed., The Annotated Sherlock Holmes (2d ed.; London 1973) II 261–81 at 277.Google Scholar

⁶⁸ For someone like Arnulf, who passed directly from being a layman to episcopacy: Vita Arnulfi 7 (434 Krusch, ).Google Scholar

⁶⁹ The facts are stated in proper order by the Vita Arnulfi 5 (433 Krusch, ).Google Scholar

⁷⁰ Miraculous punishment of the calumniator: Vita Arnulfi 13 (437 Krusch); the saint's ascetic practices, 8 (435 Krusch).Google Scholar

⁷¹ His ‘virtus belligerandi’ is mentioned, with rhetorical emphasis but without details: Vita Arnulfi 4 (433 Krusch). His promotions as domesticus and eventually primas palatii (7, 434–35 Krusch) involve no stated leadership in war.Google Scholar

⁷² Böhmer–Mühlbacher, , Regesten ² no. 261b. For a full account of Hildegard, including the late medieval legends about her, see the study of Schreiner (n. 14 supra) 1–70. For the sake of simplicity, I assume that Charles's ephemeral union to a Lombard royal daughter was not a full-fledged marriage.Google Scholar

⁷³ Marriage to Fastrada, Böhmer–Mühlbacher, , Regesten ² no. 264a (before 6 October). On Charles's later children, see the table in Werner, Karl Ferdinand, ‘Die Nachkommen Karls des Grossen bis um das Jahr 1000,’ in Braunfels, W. and Schramm, P. E., Karl der Grosse IV, Das Nachleben (Düsseldorf 1967), with commentary on p. 443. Two mistresses bore Charles three sons between 801 and 807, and two daughters in the same period. On Fastrada, mother of Charles's two known children between 783 and 800, see also Fichtenau, , Carolingian Empire 40.Google Scholar

⁷⁴ N. 56 supra. Paul's opening reference to Arnulf's sons lists Anschisus first (Gesta epp. Mett. 264 lines 37–38 Pertz, ). He later specifies that Chlodulf was the elder (264 lines 43–44).Google Scholar

⁷⁵ Liber historiae Francorum 1 (ed. Krusch, Bruno; MGH Script. rer. Merov. II 241–42). The seventh-century version in Fredegar, Chronicon 1.5, 2.4–6, 8–9, 3.2 (21, 45–47, 93 Krusch), gives no prominence to the' primitive' association with the future Romans; Aeneas is almost an afterthought, whereas the close companions of the Frigii–Franci are the Macedonians and Turks.Google Scholar

⁷⁶ Vita Arnulfi 14 (437 Krusch, ); the context is different.Google Scholar

⁷⁷ Gesta epp. Mett. (264 line 45 to 265 line 3 Pertz, ): ‘at vero minor filius, id est Anschisus, fidens de Christi pietate sibi pluriora condonari, ad omnia quae pater vellet, se libenter obedire promittit. Agit venerandus pater gratias filio, et praedicit ei, pluriora eundem quam reliquerat habiturum; insuper benedixit eum eiusque cunctam progeniem nascituram in posterum. Factumque est. Nam et pluriores Anschiso quam reliquerat divitiae accesserunt, et ita in eo paterna est constabilita benedictio, ut de eius progenie tam strenui fortesque viri nascerentur, ut non inmerito ad eius prosapiam Francorum translatum sit regnum.’ Google Scholar

⁷⁸ Ibid. 267 lines 29–33: ‘… Chlodulfus cuius supra mentionem fecimus, beati patris Arnulfi genitalis filius, Anschisi quoque, a quo semen propagatum est regium, germanus, ad episcopale culmen ob paternae sanctitatis gloriam tricesimus atque secundus ascendit, de quo nihil ad nos amplius, praeter quod a tali radice exortus est, fama perduxit.’ Google Scholar

⁷⁹ Ibid. 265 lines 4–12.Google Scholar

⁸⁰ See n. 89 infra. Sot (n. 23 supra) 439–41 fails to realize that the Carolingian genealogy looks ‘episcopal’ only because Paul made it so.Google Scholar

⁸¹ N. 69 supra; the Vita Arnulfi merely refers to the sons without naming them. Paul avoids placing Arnulf's renunciation of property at the definite time that the Vita specifies, i.e., when Arnulf resigned the bishopric (n. 63 supra).Google Scholar

⁸² On inalienability, see de Clercq, C., La législation religieuse franque I (Louvain–Paris 1936) 29, 34, 60, 64. Of course, partition was out of the question in cases of episcopal succession.Google Scholar

⁸³ Whatever the status of Charles's relationship to Himiltrude may have been, the son of this union would not have been called after Charles's father unless meant to be his heir; see Classen, , ‘Karl der Grosse u. d. Thronfolge’ 118–19. Hildegard's fourth son was Louis' twin; born in 778 and named Lothar, he died in 780 (ibid. 114).Google Scholar

⁸⁴ Notes 3, 12, 13, 18 supra. Google Scholar

⁸⁵ Classen, , ‘Karl der Grosse u. d. Thronfolge’ 109; Brühl, Carlrichard, ‘Fränkischer Krönungsbrauch und das Problem der “Festkrönungen”,’ Historische Zeitschrift 194 (1962) 307–12, 322.Google Scholar

⁸⁶ Gesta epp. Mett. (264 lines 44–45 Pertz, ): ‘Tunc maior filius, id est Chlodulfus, se hoc facere, id est ut portionem sibi debitam patri largiretur, omnimodis denegavit.’ Google Scholar

⁸⁷ N. 17 supra. Google Scholar

⁸⁸ On the ending of comedies, Frye, Northrop, The Anatomy of Criticism (Princeton 1957) 163–64.Google Scholar

⁸⁹ ‘Anschisus genuit Pippinum, quo nihil umquam potuit esse audatius; Pippinus genuit Karolum viris omnino fortissimis conferendum, qui inter cetera et magna bella quae gessit, ita praecipue Sarracenos detrivit, ut usque hodie gens illa truculenta et perfida Francorum arma formidet. Hic itaque genuit Pippinum, sapientia nihilominus et fortitudine satis clarum, qui inter reliqua quae patravit, Wascones iamdudum Francorum ditioni rebelles cum Waifario suo principe felicitate mira debellavit et subdidit. Huius item filius magnus rex Karolus extitit, qui Francorum regnum, sicut numquam ante fuerat, dilatavit’ (Gesta epp. Mett. 265 lines 4–12 Pertz). For full modern genealogies, see the tables opposite Böhmer–Mühlbacher, , Regesten ² p. 250, and Hlawitschka, Eduard, ‘Die Vorfahren Karls des Grossen,’ in Karl der Grosse I 73.Google Scholar

⁹⁰ ‘Denique inter plura et miranda quae gessit, Langobardorum gentem bis iam a patre devietam, altero eorum rege cui Desiderius nomen erat capto, alteroque, qui dicebatur Adelgisus et cum genitore regnantem suo, Constantinopolim pulso, universam sine gravi praelio suae subdidit dicioni, et, quod raro fieri adsolet, clementi moderatione victoriam temperavit. Romanos praeterea, ipsamque urbem Romuleam, iampridem eius praesentiam desiderantem, quae aliquando mundi totius domina fuerat, et tunc a Langobardis depressa gemebat, duris angustiis eximens, suis addidit seeptris; cunctaque nihilominus Italia miti dominatione potitus est. De quo viro nescias, utrum virtutem in eo bellicam, an sapientiae claritatem omniumque liberalium artium magis admireris peritiam’ (Gesta epp. Mett. 265 lines 12–21 Pertz, ). Hauck, (n. 22 supra) 152 is not alone in seeing an omen imperii in these lines. The passage may have more bearing on Paul's attitude toward the papacy than on his ethnic feelings. Menghini, , ‘Stato presente’ 238–41, shows greater astuteness on this point than Sestan, ‘Storiografia dell'Italia longobarda’ 370–71. The historical annotations that Paul makes to the genealogy are precisely echoed in the first of his verse epitaphs, that of Rothaid, daughter of Pepin III (Gesta epp. Mett. 265 line 35–266 line 2 Pertz, ).Google Scholar

⁹¹ Sot (n. 23 supra) 441 ascribes to the Gesta statements about Chrodegang, and Stephen, Pope II that it does not contain.Google Scholar

⁹² Paul's cut-and-dried account lacks the nuances stressed by modern historians; for example, Bullough, Age of Charlemagne 168. The conceit urbs Romulea recurs in the Vita Gregorii Magni 1 (98 Stuhlfath), and in Paul's dedication to Charlemagne of an epitome of Festus De uerborum significatu (text in Neff, Gedichten 123–25); in the latter, the phrase appears in a context similar to the Gesta's.Google Scholar

⁹³ Gesta epp. Mett. (265 lines 22–28 Pertz): ‘Hic ex Hildegard coniuge quattuor filios et quinque filias procreavit. Habuit tamen, ante legale connubium, ex Himiltrude nobili puella filium nomine Pippinum. Natorum sane eius quos ei Hildegard peperit, ista sunt nomina; primus dictus est Karolus, scilicet patris ac proavi vocabulo nuncupatus; secundus item Pippinus, fratri atque avo aequivocus; tertius Lodobich qui cum Hlothario, qui biennis occubuit, uno partu est genitus; ex quibus iam Deo favente minor Pippinus regnum Italiae, Lodobich Aquitaniae tenent.’ Google Scholar

⁹⁴ Oexle (n. 22 supra) 270.Google Scholar

⁹⁵ On Paul's disregard of notable males buried at St. Arnulf's, see supra n. 59.Google Scholar

⁹⁶ Neff, , Gedichten 119.Google Scholar

⁹⁷ Partitions: Böhmer–Mühlbacher, , Regesten ² nos. 42a, 106c (cf. Einhard Vita Karoli 3); the Carlomans leave the scene: nos. 52a, 130a; their sons: nos. 53e, 158g.Google Scholar

⁹⁸ On the premise that Paul alludes to this figurally in Arnulf's resolve to give ‘omnes suas facultates ad usus pauperum’ (Gesta epp. Mett. 264 line 43 Pertz, ).Google Scholar

⁹⁹ Classen, , ‘Karl der Grosse u. d. Thronfolge’ 115. The name returned initially with the oldest son of Louis the German, whose first public actions suggest that he was born near 830: Böhmer–Mühlbacher, , Regesten ² no. 1091b.Google Scholar

¹⁰⁰ The capacity of sons of Friedelehe to inherit is stressed by Fichtenau, , Carolingian Empire 38–41. The situation was perhaps not so clear cut. Paul's statement that Pepin was born ‘ante legale connubium’ is either incidental or deeply meaningful; its import depends on the view that the court took of canonical marriage as a condition for the birth of full-fledged heirs. The case of Grifo, adduced by Fichtenau, , has one face in the lifetime of Charles Martel and a very different one after 741.Google Scholar

¹⁰¹ N. 7 supra. Google Scholar

¹⁰² Gen. 27.28, ‘Cui Esau: Num unam, inquit, tantum benedictionem habes, pater?’ Isaac weeps.Google Scholar

¹⁰³ Böhmer–Mühlbacher, , Regesten ² nos. 672b, 773a. Halphen, , Charlemagne 251–53, 259, noted that the Ordinatio of 817 had been drawn up ‘comme si cet empereur de quarante ans à peine … avait fait vœu de chasteté jusqu'à la fin de ses jours.’ Charlemagne was closer to thirty-five than to forty at Hildegard's death.Google Scholar

¹⁰⁴ Fichtenau, , Carolingian Empire 41, points out the dark implications of this coincidence.Google Scholar

¹⁰⁵ N. 78 supra. Google Scholar

¹⁰⁶ Bullough, ,Age of Charlemagne 79,94; Duchesne, Louis, Fastes épiscopaux de l'ancienne Gaule III (Paris 1915) 57–58.Google Scholar

¹⁰⁷ Simson, Bernhard, Jahrbücher des fränkischen Reichs unter Ludwig dem Frommen (Leipzig 1874–76) I 196–97, II 126–27.Google Scholar

¹⁰⁸ Historia. Romana 14 (200–207 Droysen) occupies the dramatic center of Paul's continuation of Eutropius.Google Scholar

¹⁰⁹ N. 36 supra. Google Scholar

¹¹⁰ Fulrad, is not attested as carrying out the duties of his office after 780 (n. 21 supra). Conceivably, Angilram was in temporary charge of the capella before Fulrad's death and his own appointment.Google Scholar

¹¹¹ Notes 90–92 supra. Google Scholar

¹¹² In addition to the passages cited in n. 49 supra, see also Gregory of Tours, Hist. 2.7: the battle with the Huns is briefly mentioned after three long evocations of the Roman apostles.Google Scholar

¹¹³ Oddly, Paul makes no reference to the papal role in the kingship of Pepin or the crownings of 781; Pepin is already king when he chooses Chrodegang to escort Stephen II northward (Gesta epp. Mett. 268 lines 2–4 Pertz, ). Paul's genealogy draws no sharp line between descendants of Arnulf who were kings and those who bore lesser titles.Google Scholar

¹¹⁴ Böhmer–Mühlbacher, , Regesten ² nos. 649a–50, and n. 4 supra. Ewig (n. 2 supra) 244–45; Beumann, (n. 2 supra) 540–45.Google Scholar