# **Cassiodorus'** *Variae* **in the 9th Century**

by Marco Cristini

The fate of Cassiodorus' *Variae* during the Early Middle Ages is largely unknown, since the manuscript tradition begins with the eleventh century, and long quotations taken from them are attested only from that period. However, words or expressions reminiscent of the *Variae* occur more than once in Charlemagne's letters to Byzantium, in the works of Paschasius, and in the *Donation of Constantine*. The author of the epistles sent by Charlemagne to the East Roman emperor was aware of the ideological context of *Variae* I, 1, and the same is true for the *Donation of Constantine*. At the same time, Paschasius used Cassiodorus as a source of elegant words and expressions, thereby treating him like a classical author. It is impossible to ascertain whether Carolingian writers had access to all books of the *Variae,* or only to a substantial selection of letters (similar for instance to the manuscript containing the *Epistolae Austrasicae*), but there is a high degree of likelihood that they knew quite a few epistles of Cassiodorus, and were able to appreciate the political messages conveyed by them.

Early Middle Ages; 9th Century; Charlemagne; Paschasius Radbertus; Cassiodorus' *Variae*; Donation of Constantine.

Marco Cristini, University of Tübingen, Germany, marco.cristini@sns.it, 0000-0002-9629-5119

Referee List (DOI 10.36253/fup\_referee\_list) FUP Best Practice in Scholarly Publishing (DOI 10.36253/fup\_best\_practice)

Marco Cristini, *Cassiodorus'* Variae *in the 9th Century*, © Author(s), CC BY 4.0, DOI 10.36253/978-88-5518- 664-3.09, in Fabrizio Oppedisano (edited by), *Between Ostrogothic and Carolingian Italy. Survivals, revivals, ruptures*, pp. 109-125, 2022, published by Firenze University Press, ISBN 978-88-5518-664-3 (PDF), DOI 10.36253/978-88-5518-664-3

*Abbreviations*

CCCM = Corpus Christianorum, Continuatio Mediaevalis.

CCSL = Corpus Christianorum, Series Latina.

CDS = Cross Database Searchtool.

CSEL = Corpus Scriptorum Ecclesiasticorum Latinorum.

LLT = Library of Latin Texts.

MGH, AA 9 = *Chronica minora saec. IV. V. VI. VII.* (II), ed. T. Mommsen, Berlin 1892 (Auctores antiquissimi, 9).

MGH, AA 11 = *Chronica minora saec. IV. V. VI. VII.* (II), ed. T. Mommsen, Berlin 1894 (Auctores antiquissimi, 11).

MGH, AA 12 = Cassiodorus, *Variae*, ed. T. Mommsen, Berlin 1894 (Auctores antiquissimi, 12).

MGH, Conc. 2, 2 = *Concilia aevi Karolini (742-842). Teil 2 (819-842)*, Hannover-Leipzig 1908 (Concilia, 2, 2).

MGH, Conc. 3 = *Die Konzilien der karolingischen Teilreiche 843-859*, ed. W. Hartmann, Hannover 1984 (Concilia, 3).

MGH, Epp. 3 = *Epistolae Merowingici et Karolini aevi* (I), ed. E. Duemmler, Berlin 1892 (Epistolae [in Quart], 3).

MGH, Epp. 4 = *Epistolae Karolini aevi* (II), ed. E. Duemmler, Berlin 1895 (Epistolae [in Quart], 4).

MGH, Epp. 7 = *Epistolae Karolini aevi* (V), ed. P. Kehr, Berlin 1928 (Epistolae [in Quart], 7).

MGH, Poetae 1 = *Poetae Latini medii aevi Carolini* (I), ed. E. Duemmler, Berlin 1881 (*Poetae Latini medii aevi*, 1).

MGH, SS rer. Germ. 6 = *Annales regni Francorum*, s.a. 814, ed. F. Kurze, Hannover 1895 (Scriptores rerum Germanicarum in usum scholarum separatim editi, 6).

MLW = *Mittellateinisches Wörterbuch bis zum ausgehenden 13. Jahrhundert*, München 1967-. NGML = *Novum glossarium mediae Latinitatis ab anno DCCC usque ad annum MCC*, København 1955-.

PL = Patrologia Latina.

SC = Sources Chrétiennes.

ThlL = *Thesaurus linguae Latinae*, Leipzig 1900-.

#### 1. *Introduction*

The letters of Cassiodorus are one of the most important sources for the history of sixth-century Italy, and were widely used as a model for the drafting of letters during the Late Middle Ages and the Early Modern Period, yet their fate during the Early Middle Ages is largely unknown. They are mentioned in three ninth-century catalogues from Lorsch, and then brief quotations taken from them surface in documents written from 997 onwards in the area around Rome, as recent research carried out by Internullo has shown1 . Apart from two eleventh-century fragments, the *Variae* are again attested from the twelfth century onwards. However, scholars have often looked for explicit mentions of Cassiodorus or, failing these, for whole sentences taken from his correspondence, neglecting the occurrences of single words or expressions2. Thanks to the databases of classical, late antique, and early medieval texts that are now available, it is possible to obtain quite easily a comprehensive overview of the occurrences of the main lexical and stylistic peculiarities of Cassiodorus' letters in earlier and later works, thereby gaining an insight into their diffusion in ninth-century Europe. To this end, selected expressions taken from the *Variae* have been investigated by using three databases (*Library of Latin Texts*, *Monumenta Germaniae Historica*, and *Corpus Corporum*<sup>3</sup> ) and their occurrences have been duly contextualized in order to assess their significance.

#### 2. *Cassiodorus at Aachen: the* Variae *as models for Charlemagne's letters to Constantinople*

Diplomatic letters played a fundamental role in Late Antiquity and the Early Middle Ages, since they integrated and sometimes replaced oral mes-

<sup>1</sup> For an overview, see Michel, *Transmission*. On the Lorsch catalogues, see Cristini, *Liber epistularum Senatoris*. On ninth and tenth century occurrences, see Internullo, *Felix querela*, as well as Internullo's paper in this volume.

<sup>2</sup> The lack of explicit mentions of Cassiodorus' *Variae* in early medieval works has led to a *communis opinio* which could be summarized as follows: «aucun élément ne semble attester de la connaissance des *Variae* parmi les membres de l'école palatine ni de la chancellerie carolingienne» (Michel*, Les Variæ*, p. 90). <sup>3</sup> *Corpus Corporum*: <http://www.mlat.uzh.ch/MLS/>; *Monumenta Germaniae Historica*:

<sup>&</sup>lt;https://www.dmgh.de/>; LLT: <www.brepolis.net.> [Last accessed 28 December 2021].

sages. However, the definition of diplomatic correspondence is by no means straightforward, because it encompassed (and indeed still encompasses) several types of documents which were sent to different addressees. For the purpose of this study, a diplomatic letter is seen as an epistle signed by a sovereign and sent to another sovereign or to a leading secular authority belonging to a political entity situated outside the territories ruled by the sender4. These documents were not at all rare in the ancient and medieval world, for each embassy usually carried one or more letters, possibly supplemented by oral messages, yet they rarely survived, as their usefulness was often limited to a particular situation, after which the preservation of these documents became of secondary importance. The chanceries of the major Post-Roman kingdoms, the Carolingian Empire and Byzantium, probably kept originals or copies of many of them, but the loss of almost all secular archives resulted in the disappearance of most letters concerning foreign affairs, while documents dealing with doctrinal issues or Church properties were preserved by ecclesiastical writers, or in the archives of religious and monastic institutions5 .

Quite a few diplomatic letters written in Ostrogothic Italy and Merovingian Gaul have survived and, although they are by no means complete, they nevertheless allow us to grasp with a good degree of precision the most important features of late antique and early medieval diplomatic epistles6. Unfortunately, the same is not true for the Carolingian world, and especially for its founder, Charlemagne. In fact, only four of Charlemagne's diplomatic letters have survived, two addressed to Offa, king of Mercia, and two sent to Constantinople (one to emperor Nicephorus I and the other to his successor, Michael I)7 .

The small size of the sample is misleading, as Charlemagne's reign witnessed constant contacts with Byzantium, which became crucial after the annexation of the Lombard Kingdom and the subsequent hegemony over most of Italy8. Following the imperial proclamation of 800, the authority of the Frankish sovereigns had to be founded on a new basis, and this often clashed

<sup>4</sup> See most recently Flierman, *Gregory of Tours*. <sup>5</sup> See Gregory the Great, *Epistulae*, IX, 229: the Visigothic ruler Reccared asked the Pope whether the papal archives contained a copy of the treaty signed by Athanagild and Justinian fifty years earlier, but Gregory reported that the documents from Justinian's time had been destroyed by fire. The fact that only half a century later neither the Visigothic nor the Roman chancery (possibly acting as an intermediary) had a copy of such an important document contributes to explain the rarity of early medieval diplomatic letters.

<sup>6</sup> Ostrogothic Italy: Cassiodorus' *Variae* (including 32 diplomatic letters, 19 of which were sent to Constantinople and 13 to Germanic sovereigns or foreign peoples); Merovingian Gaul: *Epistolae Austrasicae* (including at least 44 diplomatic letters). The standard text of the *Variae* is now offered by the six-volume edition directed by Giardina, but see also Mommsen's classical edition. The *Epistolae Austrasicae* have been newly edited by Malaspina, whose work replaced Gundlach's edition.

<sup>7</sup> Edited in Alcuin, *Epistolae*, 87 and 100 (letters to Offa), and in *Epistulae variorum Carolo Magno regnante scriptae*, 32 and 37 (letters to Constantinople). Another letter allegedly written by Charlemagne and addressed to Offa (edited in PL 98, col. 905) is clearly forged, as it has already been shown by Von Sickel, *Acta regum et imperatorum*, 2, pp. 58 and 276. <sup>8</sup> See most recently Kislinger*, Diskretion.*

with the traditional imperial prerogatives, giving rise to a lengthy debate about the title of emperor9. It has already been observed that after 800 Charlemagne had to look for new models, and that he did not hesitate to make use of late antique formulas. The most famous is undoubtedly «Romanum gubernans imperium», which was first used in a few papyri from Ravenna dating back to the time of Justinian and, although in a slightly different form, in the constitution *Deo auctore* of 53010. The letters sent to Constantinople indicate that he may also have used other sixth-century documents.

The first case-study is represented by the letter to Nicephorus I (811)11. Charlemagne reports that he welcomed an envoy from Constantinople, the *spatharius* Arsafius, who had been sent to his son Pippin, but had been unable to carry out the negotiations due to the death of the young *rex Langobardorum* in 810. It is likely that the main goal of the embassy was Pippin's military expedition to Veneto and the Venetian lagoon, a territory which was still formally subject to the authority of Byzantium, although it enjoyed a high degree of independence12. Charlemagne took this opportunity to resume the negotiations with Constantinople, which had been at a standstill for almost a decade, with the aim of achieving the recognition of his imperial title and, more generally, of putting forward a peace agreement13. To this end, he sent several envoys to the East, who are mentioned at the end of the letter.

This document includes some lexical and stylistic peculiarities that bring it close to Cassiodorus' *Variae*. First of all, the title *fraternitas,* referring to the *basileus,* is striking. When Frankish kings wrote to Constantinople, they usually addressed the Eastern emperor as *dominus* or *pater*, not as *frater* or *fraternitas*14. Interestingly, Einhard remarks that Charlemagne sent several embassies and letters to Byzantine rulers, in which he called them brothers15. This information was considered as noteworthy, since it is one of the few remarks concerning Charlemagne's relationship with Constantinople which Einhard included in his work.

The use of the vocabulary of kinship in diplomatic communications has always been an important element of the correspondence between ancient

<sup>9</sup> This is the so-called *Zweikaiserproblem*, on which historians have been debating for over a century, see e.g. Ohnsorge, *Das Zweikaiserproblem*; Muldoon, *Empire and Order*, pp. 46-51; most recently Ančić, *The Treaty of Aachen*. <sup>10</sup> Classen, *Romanum gubernans imperium*; see also Herrin, *Ravenna*, p. 378. <sup>11</sup> I henceforth use the text edited by Duemmler in *Epistulae variorum Carolo Magno regnante* 

*scriptae*, 32 (pp. 546-548). On the background of the two letters sent to Constantinople, see Lounghis, *Les ambassades byzantines*, pp. 158-162. The sources mentioning the two legations are listed by Nerlich, *Diplomatische Gesandtschaften*, pp. 265-267. <sup>12</sup> See Kislinger*, Diskretion, pp. 286-289.* <sup>13</sup> The negotiations led to the so-called Treaty of Aachen, see most recently Ančić, *The Treaty of* 

*Aachen*; Džino*, From Justinian to Branimir*, pp. 151-152. <sup>14</sup> See *Epistolae Austrasicae*, 18 («dominus»), 19 («dominus et pater»), 20 («dominus et pater»),

<sup>25 («</sup>dominus […] pater»), 26 («dominus»).

<sup>15</sup> Einhard, *Vita Caroli*, 28: «mittendo ad eos crebras legationes et in epistolis fratres eos appellando».

sovereigns, but other expressions were usually employed in this period16. A comparison with the *Epistolae Austrasicae* confirms the rarity of *fraternitas*, which occurs only in a letter of Bishop Mapinius to Bishop Nicetius (*Epistolae Austrasicae*, 11, 4) in a religious context. This represented the main usage of *fraternitas* within epistles written during late antiquity17. The term was not part of the technical vocabulary which was used in the chancery of the Merovingian courts when addressing foreign sovereigns. Charlemagne, on the other hand, used *fraternitas* not only in his letters to Byzantium, but also in a letter to Offa, king of Mercia, dated to 79618. In this case, it is likely that the letter was drafted by Alcuin, who may have applied a term which was typical of letters written by clerics in the correspondence between sovereigns, but he may have borrowed an expression then in use in the chancery at Aachen, or have been inspired directly by a late antique letter collection, such as that of Cassiodorus19.

What is certain is that Cassiodorus is one of the very few authors who employed the term *fraternitas* in letters that are not addressed to the clergy, as is shown by *Variae*, III, 2 (to the king of the Burgundians), and *Variae*, V, 1 (to the king of the Varni), both written on behalf of King Theoderic20. *Fraternitas* conveys here a precise political message, as it places the addressee and the sender on an equal footing, which may be, depending on the circumstances, a *captatio benevolentiae* towards a sovereign who was clearly less illustrious than the sender, or an implicit claim to a degree of authority which has not yet been fully accepted by the addressee.

Of course, the choice of the term *fraternitas* by Charlemagne might be considered not so much a reminiscence from Cassiodorus as an expression with a precise political meaning, aimed at stressing the equivalence of Charlemagne's position, especially since previous Frankish kings usually addressed the Eastern emperor by using *pater* or *dominus*, which implies a relationship

magne's letters, see Louis the German, *Epistula ad Basilium*. <sup>17</sup> See ThlL VI, 1, col. 1259, ll. 7-14; MLW 4, coll. 466-467. Gregory the Great often employs *fraternitas* when addressing members of the clergy, see O'Donnell, *The Vocabulary*, p. 178. On the other hand, Symmachus uses it in a letter to a friend (*Epistulae*, IV, 21, 2). <sup>18</sup> Alcuin, *Epistolae*, 100 (p. 145, l. 10): «Relectisque vestre fraternitatis epistolis». On this letter

<sup>16</sup> See Dölger, *Die "Familie der Könige"*; Krautschick, *Die Familie der Könige*; Nerlich, *Diplomatische Gesandtschaften*, pp. 73-78; Brandes*, Die »Familie der Könige«*. There are indeed a few occurrences of *frater*, but in different geographical or chronological contexts. The Visigothic king Sisebut calls *frater* the Lombard ruler Adaloald (*Epistolae Wisigoticae*, 9, p. 671, l. 19); the same term occurs in a letter sent by Emperor Michael II to Louis the Pious (*Concilium Parisiense a. 825*, pp. 475 and 478, often in the expression *spiritalis frater*), whereas Louis the German calls Basil I both *frater* and *fraternitas*, possibly following the example set by Charle-

see Musca, *Carlo Magno*, pp. 54-62. <sup>19</sup> Wallach, *Charlemagne and Alcuin*, has argued that Alcuin took part personally in the

drafting of Charlemagne's correspondence.

<sup>20</sup> Cassiodorus, *Variae*, III, 2, 3-4 («Et ideo illum et illum legatos ad fraternitatem tuam credidimus destinandos. (...) Quapropter fraternitas vestra adhibito mecum studio eorum nitatur reparare concordiam»), and V, 1, 1 («spathas nobis etiam arma desecantes vestra fraternitas destinavit»).

of subordination. However, these two interpretations are by no means mutually exclusive. It is likely that Charlemagne, or rather the actual writer of his letters, decided to turn to a few late-antique documents because he needed models offering expressions suitable for a relationship between sovereigns who should have been regarded as equals.

Another echo of Cassiodorus can be found at the end of the epistle, in the expression «propter quod nihil morantes (...) legatos nostros praeparavimus ad tuam amabilem fraternitatem dirigendos». If the use of the verb *dirigere* is widespread in documents of this kind, the gerundive and – above all – the noun *legati* are much less so, especially in the *Epistolae Austrasicae*, in which *legatus* occurs just twice (*Epistolae Austrasicae*, 8, 1, and 18, 1, written respectively by Bishop Nicetius and Theodebald), compared to twenty-three occurrences of *legatarius*, while *dirigere* is never used in the gerundive. In a similar way, the ambassadors are called *missi*, not *legati*, in the second letter to Offa21. On the other hand, the expression «legatos [ad aliquem] dirigendos» preceded by a perfect indicative in the first person plural occurs three times in Cassiodorus, always within diplomatic letters: first in an epistle sent to the Visigothic king Alaric II, then in one addressed to the Frankish sovereign Clovis and, finally, in the first letter of the young Athalaric to emperor Justin22. As far as the latter two documents are concerned, a further parallel with Charlemagne's letter is represented by the use of the adverb *quapropter* to introduce the sentence containing the mention of the embassies. Of course, these are common statements in diplomatic letters, but it should be noted that the expression «legati dirigendi» before the year 900 appears within a diplomatic letter only in Cassiodorus and in the epistle to Nicephorus I23.

Turning now to the letter to Michael I (813), this document aimed to promote peace between the Carolingian Empire and Byzantium, and concord between their respective Churches. Charlemagne announced that he had sent two envoys to Constantinople, Amalarius of Metz, archbishop of Trier, and Peter, abbot of Nonantola, who were tasked with concluding peace negotiations between the two empires. It seems that the emperor of Constantinople had agreed to sign a formal peace treaty following the previous embassy, and had sent a draft of it to Charlemagne, who signed it and sent it back to Byz-

<sup>21</sup> Alcuin, *Epistolae*, 100 (p. 145, l. 11). The first letter does not mention any envoy. On the terms used to refer to envoys, see Nerlich, *Diplomatische Gesandtschaften*, pp. 103-106. <sup>22</sup> Cassiodorus, *Variae*, III, 1, 4: «Et ideo salutationis honorificentiam praelocuti legatos no-

stros illum atque illum ad vos credidimus esse dirigendos»; III, 4, 4: «quapropter ad excellentiam vestram illum et illum legatos nostros magnopere credidimus dirigendos, per quos etiam ad fratrem vestrum, filium nostrum regem Alaricum scripta nostra direximus»; VIII, 1, 5: «quapropter ad serenitatem vestram illum et illum legatos nostros aestimavimus esse dirigendos, ut amicitiam nobis illis pactis, illis condicionibus concedatis».

<sup>23</sup> As is shown by the search string «legat\* dirigend\*» on *Corpus Corporum* (works written before 900) and on LLT and CDS (up to ten words between the two terms). There is an occurrence in *Acta Concilii Carthaginensis a. 525* (p. 256): «Epistula ergo quam beatissimo sancto fratri et consacerdoti meo seni missori, primati provintiae Numidiae, pro dirigendis legatis transmisimus, ab officio recitetur».

antium for the emperor to sign as well. In the end, Amalaric and Peter were supposed to bring back to Aachen a copy of the agreement translated into Greek and bearing the signature of the *basileus*24.

A close analysis of the letter reveals striking similarities with the *Variae*, especially with the first letter, *Variae*, I, 1. The expression «quaesitam (...) pacem» at the beginning of Charlemagne's letter brings to mind the incipit of the epistle sent by Theoderic to Anastasius I: «Oportet nos, clementissime imperator, pacem quaerere»25. This is a significant analogy not only because the political goals of the two documents are very similar, but also because the expression *pacem quaerere/quaesita pax* is rare in classical Latin, and is mostly used by Augustine and other ecclesiastical writers in religious contexts, not to refer to political issues26. More traditional expressions such as *pacem petere*, *orare*, *postulare*, *exposcere* or *rogare*27 would have placed the sender in a clearly subordinate position, comparable to that of a supplicant, and very similar to the traditional image of a defeated barbarian. Therefore, Cassiodorus, writing on behalf of Theoderic, chose an ambiguous expression to maintain the balance between a formal deference to Constantinople and the wish to claim the independence of the Goths and the quasi-imperial status of their sovereign.

Although in a radically different geopolitical context, Charlemagne's concerns were quite similar. The Frankish king wished to obtain recognition of his imperial dignity and a peace agreement that would put an end to the clashes with Byzantium, but at the same time it was advisable not to offend the *basileus*, who was traditionally reluctant to share the title of *imperator* with other sovereigns. For these reasons, the reign of Theoderic and, more specifically, the vocabulary used in *Variae*, I, 1, represented a perfect model.

Charlemagne's letter contains other traces of expressions reminiscent of Cassiodorus, as is shown by the sentence «praesentes legatos nostros (...) ad tuae dilectae fraternitatis gloriosam praesentiam direximus»28. While it is not

<sup>24</sup> This eventually happened in 814, see *Annales regni Francorum*, s.a. 814 (p. 140): «Leo imperator, qui Michaheli successerat, dimisso Amalhario episcopo et Petro abbate, (...) descriptionem et confirmationem pacti ac foederis misit».

<sup>25</sup> Cassiodorus, *Variae*, I, 1, 1. See *Epistolae variorum Carolo Magno regnante scriptae*, 37 (p. 556, ll. 7-9): «in diebus nostris diu quaesitam, et semper desideratam pacem inter Orientale atque Occidentale imperium stabilire». On the revival of Cassiodorus, *Variae* I, 1, in the Late Middle Ages, see Fabrizio Oppedisano's paper in this volume (note 61).

<sup>26</sup> See Livy, *Ab Urbe condita libri*, XLII, 50, 11 («de bello et pace quaeri»), which is the occurrence most similar to those of Cassiodorus, but the context is completely different. «Quaesita pax»: see Cicero, *De officiis*, I, 80; Justin, *Historiae Philippicae*, II, 4; *Historia Augusta*, *Gallieni duo*, 5, 5; Orosius, *Historiae*, I, 15, 3. For occurrences in religious works, see e.g. Tertullian, *Adversus Marcionem*, II, 19; Cyprian, *De Ecclesiae Catholicae unitate*, 24; Jerome, *Epistulae*, 125, 93; Augustine, *Epistulae*, 220, 12, and *Enarrationes in Psalmos*, 33, ser. II, 19; Gregory the Great, *Homiliae in Ezechielem*, X, 44. There are only few occurrences of the expression with a political meaning after Cassiodorus, and mainly in poems which hardly influenced Charlemagne's chan-

cery; see Corippus, *Iohannis*, IV, 377; Theodulf, *Carmina*, 27, 101. <sup>27</sup> ThlL X, 1, p. 876, ll. 56-69. See NGML (Passabilis - Pazzu), col. 820, ll. 16-52. <sup>28</sup> *Epistolae variorum Carolo Magno regnante scriptae*, 37 (p. 556, ll. 17-20).

the case to dwell further on *fraternitas* and *legati*, which have already been examined, the expression «praesentes legatos» preceded by *direximus* deserves a brief comment, as it is similar to a passage of *Variae*, IX, 5, 2, namely «praesentes direximus portitores». The two expressions are not identical, but a search of the occurrences of *direximus* followed or preceded by *praesentes* indicates only one case similar to those examined, namely a letter of Pope Paul I to King Pippin, in which he writes «direximus praesentes nostros fidelissimos missos»29. Clearly, this similarity could be a mere coincidence, but in the light of the presence of not a few analogies in both Charlemagne's letters and the *Variae*, it is likely that the person in charge of drafting the royal correspondence was inspired either by Cassiodorus' letters or by chancery formulas containing some expressions taken from them.

The letter to Michael includes another late antique linguistic borrowing that is quite significant in terms of political communication. We find it once again in the sentence «diu quaesitam et semper desideratam pacem inter Orientale atque Occidentale imperium»30. In fact, the expressions «Orientale imperium» and «Occidentale imperium» are quite rare, and occur mostly during Late Antiquity, more precisely in the sixth century31. They started to be used to define the two parts of the Roman Empire in the fifth century, and the occurrences show a fairly consistent pattern: out of nine authors, six use both expressions, and only three limit themselves to using one of them, an indication that they employed them consciously and not just casually.

The first occurrences are found in the *Historia Augusta* and in Orosius, who exerted a strong influence on the authors of the following centuries, and served as a model for Paul the Deacon. It cannot be ruled out that the author of the letter to Michael I was inspired by Orosius, or simply by Paul the Deacon, but it is noteworthy that half of the occurrences date back to the sixth century,

<sup>29</sup> *Codex Carolinus*, 17 (p. 514, l. 5). The occurrences have been found by using Brepolis (LLT and CDS), searching for *direximus* in association with *praesentes* (up to ten words between the terms). There is another occurrence in Iohannis VIII papa, *Epistolae*, 181 (p. 145, l. 21: «praesentes misso nostros direximus»), but it is a letter sent by Pope John VIII to Wigbod, bishop of Parma, in 879.

<sup>30</sup> As it has already been noted, although very briefly, by Classen, *Karl der Grosse*, p. 95, note 355: «Mit den Begriffen *imperium occidentale* und *orientale* wird spätrömischer Sprachgebrauch aufgenommen».

<sup>31</sup> «Orientale imperium»: *Historia Augusta*, *Aurelianus*, 22, 1, possibly also *Triginta Tyranni*, 30, 11; Orosius, *Historiae*, VII, 36, 2 (in all likelihood the source of Paul the Deacon, *Historia Romana*, XIII, 9); *Chronica Gallica a. 452*, pars posterior, 11 (p. 646); Prosper, *Epitoma Chronicon*, continuatio II, 13 (p. 489); Cassiodorus, *Chronica*, 1328 (p. 159); Cassiodorus, *Historia Ecclesiastica Tripartita*, 9, 4 (titulus); Marcellinus Comes, *Chronicon*, praefatio; Jordanes, *Getica*, 244, and *Romana*, 339; *Laterculus imperatorum ad Iustinum I* (p. 422, l. 36, and p. 423, l. 24); Paul the Deacon, *Historia Romana*, XV, 7. «Occidentale imperium»: Orosius, *Historiae*, VII, 37, 1 (transcribed by Paul the Deacon, *Historia Romana*, XII, 9); Prosper, *Epitoma Chronicon*, 1286 (p. 470); Cassiodorus, *Chronica*, 1209 (p. 155); Marcellinus Comes, *Chronicon*, a. 392, 1; Jordanes, *Getica*, 236. These occurrences are the result of looking for the search strings «occidental\* imperi\*» and «oriental\* imperi\*» on *Corpus Corporum* (http://www.mlat.uzh. ch/MLS/, works written before 850) and on LLT and CDS (http://www.brepolis.net/, up to ten words between the terms).

almost all of them in chronicles, many of which were present in ninth-century Carolingian libraries and scriptoria, as the manuscript tradition indicates32. Cassiodorus himself used both expressions in his *Chronica* and, in all likelihood, he was behind the composition of the *Historia Tripartita*, which was written under his supervision, whereas the *Getica* is based on his *Gothorum Historia*. Thus, a quarter of the occurrences can be traced back directly or indirectly to Cassiodorus and his circle. Nor should we overlook Marcellinus Comes, who was very close to Justinian and accepted to define the political entities that emerged after the division of the empire with the expressions «Occidentale imperium» and «Orientale imperium»33.

Although it is not possible to come to any definitive conclusion, it seems likely that the authors of Charlemagne's letters decided to use a few expressions taken from the late antique political vocabulary when it came to defining formally his relations with Constantinople. Faced with the *basileus* claiming the uniqueness and indivisibility of the imperial title, Charlemagne and the intellectuals of his court made use of works written in the fifth and sixth centuries, which in all likelihood included the *Variae*, to demonstrate that the coexistence of two empires, one in the West and one in the East, was by no means impossible34. The *Epistolae Austrasicae* and, more generally, the letters written by the Merovingian sovereigns no longer constituted a valid model because of both the problematic relationship between Charlemagne and the previous dynasty, and his radically different attitude towards Constantinople. It was necessary to find new models, and Cassiodorus' *Variae*, written three centuries earlier under a sovereign who considered himself the legitimate heir of the Western emperors, represented an excellent alternative. This could contribute to explain the interest of the Carolingian sovereigns

<sup>32</sup> See the comments by the editors at: Cassiodorus, *Chronica*, p. 117 (the archetype is a Reichenau manuscript written at the beginning of the ninth century, see Gatti – Stoppacci, *Cassiodorus Senator*, p. 87); Cassiodorus, *Historia Tripartita*, p. XVII (manuscript C, written at Corbie at the beginning of the ninth century); Jordanes, *Getica*, pp. XIII (diffusion in the Carolingian period), XXVIII-XXIX (e.g. manuscript V, written in the ninth century and coming from the abbey of Saint-Amand, and manuscript H, written in eighth/ninth century and coming from Fulda; it is likely that manuscript P, possibly from Lorsch, was written in the first half of the ninth century and not in the tenth century; see also Tischler, *Remembering the Ostrogoths*, p. 72, note 26). The circulation of the *Laterculus* and Marcellinus' *Chronicon* in Carolingian Europe is not attested. *Laterculus*, pp. 48-50, reports that one of the archetypes of the latter work, manuscript T of Oxford, can be dated to the sixth century. It possibly originated from Vivarium, see Troncarelli*, Il teatro delle ombre*, p. 85. The location of the manuscript in the ninth century is unknown, but it was in southern France in the fifteenth century.

<sup>33</sup> See also the *Laterculus*, which was probably written in the same period, as is argued by Zec-

chini, *Ricerche*, p. 71, and Van Hoof – Van Nuffelen, *Clavis Historicorum*, p. 683. <sup>34</sup> I find unconvincing Ančić, *The Treaty of Aachen*, p. 32, according to whom the word *imperium* means that «Charlemagne and Michael have the highest authority in *one world empire*, whose prime function is to provide the peace and security necessary for the proper functioning of the Church and the means of salvation. In this world of ideas there is no place for *two different empires*» (italics of the author). The intertextuality with late antique sources indicates instead that Charlemagne intended to refer to two distinct empires, although united by the common faith and the same Roman origins.

in the mythical and historical figure of Theoderic, which is attested by the equestrian statue of the Amal king that was brought to Aachen, as well as by the famous poem *De imagine Tetrici* by Walahfrid Strabo35.

### 3. *Cassiodorus and Paschasius Radbertus*

So far the discussion has focused exclusively on diplomatic correspondence. Further light on the issue of the circulation of Cassiodorus' *Variae* in the ninth century may be shed by turning to the monasteries of the Frankish Kingdom. In fact, the works of Paschasius Radbertus show at least two significant similarities with Cassiodorus, which could be the result of a direct knowledge of the *Variae*.

The most important one is represented by the verb *pennesco*, which in classical and early medieval times occurs only in these two authors36. Cassiodorus uses it in a letter in which a young Goth is released from the guardianship of his uncle. As it often happens, Cassiodorus inserts a metaphor taken from the animal world in a bureaucratic document, comparing the transition of the Goths to adulthood, which was determined by their ability to handle weapons, to young eagles that procure food on their own after having taken on adult plumage37. Paschasius uses the term in an allegorical context to allude to the desire for glory, but he also refers to wings and thus to the idea of flying38.

The relevance of this similarity emerges with greater clarity if we broaden the search for occurrences up to the thirteenth century. Although the number of extant Latin works increases dramatically, there are only two other occurrences, both in Saba Malaspina's *Liber gestorum regum Sicilie*39. In view of the extreme rarity of the verb *pennesco*, the similar (though not identical) context in which it occurs in Cassiodorus and Paschasius, and the fact that the other occurrences of this verb in the pre-humanistic period undoubtedly

<sup>35</sup> See most recently Licht's edition of Walahfrid Strabo, *De imagine Tetrici*, as well as Herrin,

*Ravenna*, pp. 378-381. <sup>36</sup> See ThlL X, 1, col. 1096, ll. 64-68; Du Cange *et al*., *Glossarium*, 6, col. 258a: <http://ducange. enc.sorbonne.fr/PENNESCERE>; NGML (Pea - Pepticus), col. 230, ll. 3-6.

<sup>37</sup> Cassiodorus, *Variae*, I, 38, 2: «pullos suos audaces aquilae tamdiu procurato cibo nutriunt, donec paulatim a molli pluma recedentes adulta aetate pennescant: quibus ut constiterit firmus volatus, novellos ungues in praedam teneram consuescunt: nec indigent alieno labore vivere, quos captio potest propria satiare».

<sup>38</sup> Paschasius Radbertus, *De fide, spe et charitate*, Spes, 5: «Celsa igitur spes gloriae, quae omnibus illustratur bonis, et virtutum pennescit alis, ut semper ad altiora attollat animam possidentis».

<sup>39</sup> Saba Malaspina, *Liber gestorum regum Sicilie*, I, 6 (p. 107), and IV, 3 (p. 181): the chronicler first, referring to Manfred, writes: «volat audax aquila, que nondum etate plene ceperat adulta pennescere, et rapaces ungulas assuefacit ad predas»; then, he describes Conradin: «catulum dormientem et pullum aquilae, qui nondum etate ceperat adulta pennescere». In both cases, it is clear that Saba draws inspiration from the passage of Cassiodorus, as has already been noted by the editors of *Liber gestorum*, see Saba Malaspina, *Liber gestorum regum Sicilie*, p. 107, note 96.

derive from the *Variae*, it is likely that Paschasius was also inspired by them, either directly or indirectly. In fact, the choice of *pennesco* would be quite difficult to explain had Paschasius not intended to imitate Cassiodorus, since he could have employed a much more common synonym, *plumesco*, found in several authors, including Jerome, Augustine (in the *Confessions*) and Gregory the Great (in the *Moralia*), as well as in the *Bible*40.

A further echo of Cassiodorus can be found in the sixth book of the *Expositio in Matheo*, composed by Paschasius after 849-853, when he had to leave the office of abbot41. The nexus «iniusta praesumptio» occurs only in this work and in the *Variae*42. This parallel strengthens the conjecture that Paschasius knew either the letters of Cassiodorus or a Carolingian collection of formulas including passages taken from them.

# 4. *Cassiodorus and the* Constitutum Constantini

Expressions reminiscent of the *Variae* are also present in one of the most famous medieval forgeries, namely the *Donation of Constantine*, or *Constitutum Constantini*. As is well known, the genesis of this document has been the subject of a long debate and scholars are still far from reaching unanimous conclusions about its author, dating and the existence of several versions43. It is not possible here to offer a complete and exhaustive overview of the most recent bibliography, suffice it to say that in recent years the *Constitutum* has been traced back to the monastery of Corbie, where much of the process of drafting the *Pseudo-Isidorian Decretals* took place around 83044. According to Johannes Fried, the donation in its present form is ascribable to Wala and Paschasius Radbertus, with the collaboration of Hilduin of Saint-Denis45. This reconstruction is mainly based on the manuscript tradition, since the *Constitutum* was transmitted almost exclusively through manuscripts containing the *Pseudo-Isidorian Decretals*46, and does not necessarily exclude a Roman origin of the document, which might have been subsequently modified in Corbie.

<sup>40</sup> See ThlL X, 1, col. 2458, and e.g. Jerome, *Commentarii in Matheum*, XXIV, 28; Augustine, *Confessiones*, IV, 16; Gregory the Great, *Moralia*, XIX, 48, and XXX, 35; *Biblia Vulgata*, *Job*, XXXIX, 26.

<sup>41</sup> De Jong, *Epitaph for an Era*, p. 43. <sup>42</sup> See Cassiodorus, *Variae*, IV, 13, 1; XI, 7, 5, and Paschasius Radbertus, *Expositio in Matheo*, VI (p. 583). On Radbert's use of his sources, see the introduction to Pascasius Radbertus, *Expo-*

*sitio in Matheo*, p. XXI. <sup>43</sup> For an overview of the different reconstructions, see Gandino, *Falsari Romani*; Muresan*, Le 'Constitutum Constantini'.*

<sup>44</sup> On the origin of the *Pseudo-Isidorian Decretals* at Corbie, see Zechiel-Eckes, *Fälschung als* 

*Mittel*. More prudent Knibbs, *Pseudo-Isidore's Ennodius*. <sup>45</sup> Fried, *Donation*. Interestingly, Saint-Denis seems to be the source of the Frankish interpolations of the *Liber pontificalis*, thereby indicating that Carolingian writers were willing to modify and use previous works to pursue political goals, see McKitterick, *Rome and the Invention* 

*of the Papacy*, pp. 216-218. <sup>46</sup> The manuscript Paris Lat. 2777 is the only exception.

Fried remarks that the conclusion of the sixteenth paragraph of the *Donation*, in which Constantine allows Sylvester to wear a tiara during processions, contains an unusual expression, namely *ad imitationem imperii nostri*. In the body of the text, Fried reports that the sentence is «new and without precedent»47, but in a footnote he admits the existence of an analogy with the first epistle of the *Variae*, in which Theoderic, addressing Emperor Anastasius, writes that «regnum nostrum imitatio vestra est, forma boni propositi, unici exemplar imperii»48. Although, by his own admission, the concept of *imitatio imperii* expressed in such explicit terms does not occur in other authors, Fried believes that the *Variae* were not used as a model for the *Donation of Constantine*49. On the other hand, Muresan comes to the opposite conclusion after re-examining the issue in a recent paper, and argues that the author of the *Donation*, or at least of this passage, intentionally imitated Cassiodorus50.

Muresan's argument finds confirmation in another passage of the *Constitutum*, namely paragraph 19, which contains the sentence «si quis autem, quod non credimus, in hoc temerator aut contemptor extiterit». The expression «si quis autem (...) temerator extiterit» occurs only here and in two letters of Cassiodorus, *Variae*, IX, 13-1451; similar forms are attested in other sources, but the wording «si quis autem» seems to be a peculiarity of Cassiodorus in this context52. The *Donation* contains two other expressions typical of late antique Latin, namely «amplissimus senatus» and «serenitas nostra». Although they are also attested in other authors, there is the possibility, especially with regard to the former, that the author of the *Constitutum* had the *Variae* in mind when drafting the corresponding passages53.

<sup>50</sup> Muresan, *Le 'Constitutum Constantini'*, pp. 187-189. <sup>51</sup> Cassiodorus, *Variae*, IX, 13, 3 («si quis autem iussionum nostrarum inprobus temerator exstiterit»), IX, 14, 6 («si quis autem saluberrimi constituti temerator extiterit»).

<sup>52</sup> I have looked for the search string «temerator exstiterit» and «temerator extiterit» on LLT and CDS (ten words between the terms, works written before 820) and *Corpus Corporum* (before 900). See *Codex Iustinianus*, X, 26, 3, 1: «sin vero quisquam temerator horreorum extiterit»; Bonifatius, *Epistolae*, 43 (p. 291, l. 21): «nam qui temerator exstiterit»; *II Concilium Toletanum*, 5: «si quis ergo huius decreti temerator exstiterit». This last occurrence is quite similar to that of Cassiodorus, but there is «ergo» instead of «autem» and the authenticity of the acts of the Second Council of Toledo is not certain. The expression «si quis autem (...) temerator extiterit» occurs in *Concilium Romanum a. 826*, 17 (p. 575, ll. 5-6), as well as in *Concilium Romanum a. 853*, 17 (p. 322, l. 33). The last occurrence is a transcription of the previous one, which refers to the Roman Council of November 826, presided over by Pope Eugene II, who had travelled to France in 824 to meet Louis the Pious. It seems likely that the passage in question is the result of an imitation of either the *Variae* or (most likely) the *Donation*, which Eugene II or some member of his retinue may have seen during his stay in Gaul.

<sup>53</sup> «Amplissimus senatus» occurs twice in Cassiodorus, *Variae*, IX, 16, shortly after the two letters including «si quis autem (...) temerator extiterit».

<sup>47</sup> Fried, *Donation*, p. 45 (with note 140). <sup>48</sup> Cassiodorus, *Variae*, I, 1, 3. <sup>49</sup> It seems that there are no occurrences outside the *Variae* and the *Constitutum Constantini*. I have looked for the search string «imitatio\* imperi\*» on LLT and CDS (ten words between the terms, works written before 1500).

Taken individually, these analogies might seem too small and uncertain to support the hypothesis that the author of the *Donation of Constantine*, or at least of its final version, was familiar with the *Variae*, but they should be assessed as a whole and together with the occurrences of other Cassiodorean expressions within ninth-century texts. Paschasius Radbertus most likely knew some of the letters contained in the *Variae,* and possibly played a role in the drafting of the *Constitutum*. Moreover, *Variae*, I, 1, was almost certainly used as a model for the second letter that Charlemagne sent to Constantinople. It is therefore unsurprising that this same letter, and some other epistles of Cassiodorus, were taken into consideration a few years later when drafting or, more likely, reworking a document aimed at drastically reducing the prestige of the Eastern Empire, since it retrospectively deprived it of the monopoly of imperial authority and symbols from its very conception by Constantine.

## 5. *Conclusions*

Words or expressions reminiscent of Cassiodorus' *Variae* occur more than once in Charlemagne's letters to Byzantium, in the works of Paschasius and in the *Donation of Constantine*. Evidently, one cannot rule out that such analogies are due to formularies used in the Carolingian chancery, which included a few expressions taken from Cassiodorus, yet this does not seem to be the most likely explanation. It is possible that the first document of the *Variae* became a kind of epistolary model, especially after Charlemagne (or rather the actual writers of his epistles) imitated it when writing to Byzantium. However, *pennesco* is taken from a quite unimportant letter, and the other Cassiodorean passages which were imitated by Carolingian authors show no clear pattern. If ninth-century chancery formularies really included a few passages by Cassiodorus, we should expect to find more substantial analogies, such as the transcription of whole paragraphs or sentences.

The intertextuality with the *Variae* indicates a more complex situation. The author of Charlemagne's letters to Constantinople seems to have possessed some awareness of the ideological context of *Variae* I, 1, therefore it is likely that he had some basic knowledge of sixth-century history. The same is true for the *Donation of Constantine*. On the other hand, Paschasius used Cassiodorus as a source of elegant words and expressions, thereby treating him like a classical author, whose writings could provide early medieval scholars with models of style. It is impossible to ascertain whether Carolingian writers had access to all books of the *Variae* or solely to a substantial selection of letters (similar for instance to the manuscript containing the *Epistolae Austrasicae*), but the case studies which have been examined so far suggest that they knew quite a few epistles of Cassiodorus, and that they were able to appreciate the political messages conveyed in them.

#### **Works cited**


Prosper, *Epitoma Chronicon*, ed. T. Mommsen, Berolini 1892 (MGH, AA 9, pp. 341-499).

Saba Malaspina, *Liber gestorum regum Siciliae (Die Chronik des Saba Malaspina)*, ed. W. Koller – A. Nitschke, Hannover 1999 (MGH, SS 35).

Symmachus, *Epistulae*, ed. O. Seek, Berlin 1883 (MGH, AA 6, 1, pp. 1-278).

Tertullian, *Adversus Marcionem*, ed. C. Moreschini, Milano-Varese 1971.

Theodulf, *Carmina*, ed. E. Duemmler, Berlin 1881 (MGH, Poetae I, pp. 445-581).


T. Von Sickel, *Acta regum et imperatorum Karolinorum digesta et enarrata*, vol. 2, Wien 1867.


Marco Cristini Universität Tübingen marco.cristini@sns.it