# **Lothar's manuscripts, manuscripts for Lothar, manuscripts of Lothar's time**

by Laura Pani

This paper aims to give an account of some of the manuscripts related to Lothar. In its first section an attempt is made at retracing a set of books that could have belonged to Lothar's library, nowadays known only from secondary sources. In the second section some display codices are discussed, either commissioned by Lothar, or dedicated to him, such as Lothar's Gospel Book MS Par. lat. 266 or those traditionally referred to as the *Lothar-Gruppe*, whose actual connection both to Lothar and to each other is questioned here. The third and last part of the paper contains some considerations on the manuscripts produced during the years of Lothar's government in Italy, that essentially coincide with the second quarter of the ninth century.

Middle Ages; 9th century; Carolingian Italy; Verona; Lothar; Pacificus; Carolingian royal libraries; Carolingian manuscripts; Carolingian court school; Carolingian illumination; Carolingian law-books; *Lothar-Gruppe*.

Laura Pani, University of Udine, Italy, laura.pani@uniud.it, 0000-0002-5364-1392

Referee List (DOI 10.36253/fup\_referee\_list)

FUP Best Practice in Scholarly Publishing (DOI 10.36253/fup\_best\_practice)

Laura Pani, *Lothar's manuscripts, manuscripts for Lothar, manuscripts of Lothar's time*, © Author(s), CC BY 4.0, DOI 10.36253/978-88-5518-623-0.03, in Gianmarco De Angelis, Francesco Veronese (edited by), *Networks of bishops, networks of texts. Manuscripts, legal cultures, tools of government in Carolingian Italy at the time of Lothar I*, pp. 13-31, 2022, published by Firenze University Press, ISBN 978-88-5518-623-0, DOI 10.36253/978- 88-5518-623-0

Abbreviations

*C.A.L.M.A.* = *Compendium Auctorum Latinorum Medii Aevi 500-1500*, 2.5, ed. M. Lapidge – C. Leonardi – F. Santi, Firenze 2008.

MGH, DD Lo I / Lo II = *Die Urkunden Lothars I. und Lothars II*., ed. T. Schieffer, München 1979 (MGH, DD Karolinorum, 4).

MGH, Epp. V = MGH, *Epistolarum Tomus V*, ed. E. Dümmler *et al.*, Berlin 1899 (Epistolae Karolini aevi, 3).

MGH, Poetae II = MGH, *Poetae Latini aevi Carolini*, II, ed. E. Dümmler, Berlin 1884.

MGH, Poetae VI/1 = MGH, *Poetae Latini aevi Carolini*, VI/1, ed. K. Strecker, Weimar 1951.

#### 1. *A Lothar library?*

I will start with that which does not exist anymore, that is with the traces of Lothar's library or, at least of the books that Lothar, similarly to other Carolingian sovereigns, owned and presumably read1 . With the exception of the famous display codices discussed in the next section of this paper, only secondary sources, such as letters and inscriptions, or later manuscript copies, actually allow us to retrace what seemed to be more, in any case, a set of books meant to meet the emperor's interests and inclinations, than an organized collection.

Some time between 854 and 855, Lothar wrote to Hrabanus, at that time bishop of Mainz, asking for a lectionary for the Sundays and the other main liturgical feasts, to be read to him during his meals. Lothar's request was that each reading from the Gospel would have to be followed by an explanation by one of the Fathers of the Church; this would mean that he would not have to take with him «omnem commentariorum copiam» on every expedition. For similar, practical reasons, Lothar asked that this homiliary should be contained in just one volume, although at the end of the letter he considered that the work could be divided into two or even three books2.

Hrabanus granted Lothar's wish. A tenth-century copy of the first volume of his homiliary, from Christmas to Holy Week, is now MS Jena, Thüringer Universitäts- und Landesbibliothek, El. f. 32, and includes in the first folios Lothar's letter to Hrabanus and Hrabanus' answer accompanying the work3. The second volume, from Easter to Pentecost, is now missing but was still extant in the seventeenth century, when an edition was made from it in Cologne, whilst the third volume was probably never compiled due to Lothar's death4.

<sup>4</sup> Étaix, *L'homéliaire*.

<sup>1</sup> A first summary on the Carolingian royal collections of books is Lesne, *Histoire*, 4, pp. 446- 452 (p. 450 on Lothar's). As is known, Charlemagne's, Louis the Pious' and Louis the German's libraries were later extensively discussed by Bernhard Bischoff (*The Court Library of Charlemagne*; *The Court Library*; *Bücher*), Charlemagne's library being further questioned by Villa, *La tradizione di Orazio*, and Bullough, *Charlemagne's Court Library*. An account on Charles the Bald's library is McKitterick, *Charles the Bald*. <sup>2</sup> De Jong, *The Empire*, pp. 191-192. The letter is published in MGH, Epp. V, pp. 503-504. 3 A full catalogue description of this manuscript, as well as a digitization, is apparently lacking.

The codex is briefly described only by Waitz, *Handschriften*, pp. 694-695, with a focus on the two letters.

The abundance of commentaries mentioned by Lothar in his letter suggests that his library was rich in exegetical books, which is confirmed by what we know from his previous exchanges with Hrabanus himself. Between 840 and 842 Lothar had been the recipient of Hrabanus' Commentary on Jeremiah, begun during his father's reign5 . Moreover, between 842 and 846, Lothar had asked Hrabanus for more «expositiones», namely on the beginning of Genesis, Jeremiah's sermons and Ezechiel, that had not been covered by Bede, Jerome and Gregory the Great respectively. Eventually, Hrabanus had sent to Lothar his commentary on Ezechiel, at the same time explaining why he had not written one on Genesis, and recalling having already sent to him the commentary on Jeremiah6.

In addition to this, the Commentary on the Song of Songs by the monk of Luxeuil Angelomus was written at Lothar's request around 8517 .

Lothar's quite extensive knowledge of exegetical works by different authors, and his eagerness to have as complete a collection of them as possible, as shown in his letter to Hrabanus from 842-846, confirms that exegesis significantly complied with the interests of ninth-century both ecclesiastical people and lay noblemen and royals, the latter not seldom being the first addressees of the Carolingian scholars' works8.

The MS Bamberg, Staatsbibliothek, Class. 30 is a composite manuscript, the first codicological unit of which contains two grammatical treatises: an *Ars grammatica* by Clemens Scottus, an Irish schoolmaster at Louis the Pious' court († after 826), and a *Pauca de barbarismo collecta de multis* by an unknown author; on fol. 70*v* a 9-dystich-long dedicatory poem follows, whose author is Clemens Scottus himself and the recipient, again, Lothar9. According to Bernhard Bischoff, this manuscript was copied in Rheims in the third quarter of the ninth century and is, therefore, a later copy of a book that had been probably presented to Lothar by Clemens as an homage from a master to his pupil10.

<sup>5</sup> De Jong, *The Empire*, p. 208; MGH, Epp. V, pp. 442-444. On the work: Guglielmetti, *Hraba-*

*nus Maurus*, pp. 318-320. 6 De Jong, *The Empire*, pp. 211-212; MGH, Epp. V, pp. 475-476 and 476-478. On the commen-

tary on Ezechiel: Guglielmetti, *Hrabanus Maurus*, pp. 320-322. 7 Guglielmetti, *Angelomus*, pp. 40-42. See also Gorman, *The Commentary*, particularly pp. 563, 567. The prefatory letter, testifying to Lothar's demand, is published in MGH, Epp. V, pp. 625-630.

<sup>8</sup> Hrabanus' exegetical works for Lothar were just a part of the commentaries he dedicated to kings and emperors, and to their wives too: Lothar's wife Irmingard was herself the recipient of Hrabanus' commentary on Esther (De Jong, *The Empire*, pp. 194, 212). See also a recap on the royal recipients of Hrabanus' exegetical works in Guglielmetti, *Hrabanus Maurus*, p. 276. More hints on the exegetical works compiled by Carolingian scholars for Carolingian sovereigns in Gorman, *The Commentary*, pp. 567-568, 589, 601. 9 On Clemens Scottus: *C.A.L.M.A.*, 2.5, pp. 646-647; Ó Corráin, *Clavis*, 2, nn. 517-519. The

poem is published in MGH, Poetae II, p. 670. See also Garrison, *The English*, pp. 99-100. 10 Bischoff, *Katalog*, 1, n. 206, < https://zendsbb.digitale-sammlungen.de/db/0000/sbb00000094/ images/ >.

Since all these books mostly survive only in later copies, we have no clue as to their codicological and paleographical features.

The Vatican MS Reg. lat. 438, containing on fols. 1*r*-30*r* the Martyrology in verse composed by Wandalbert of Prüm and dedicated to Lothar, is also a later copy than the original: whilst the work was composed around 848-849, this extant copy dates to the last decades of the ninth century, and probably comes from Reichenau11. As a matter of fact, the miniatures of this manuscript have recently been attributed to an illuminator working in the Lake of Constance region at the end of the century, and responsible for a number of decorated codices12. We cannot be sure whether the picture with the monk presenting the book to a king on fol. 1*v* of the Vatican manuscript is an original subject – and in this case on who the king is – or a copy from an exemplar showing the same set of miniatures13. Should this exemplar consist in, or have been copied from14, the original one dedicated to Lothar, this would allow us to add a new portrait to the two Lothar portraits known from other luxury codices that were actually produced during his reign15.

### 2. *Lothar's display codices and the* Lothar-Gruppe *reconsidered*

The first portrait of Lothar is found in the famous Gospel Book Par. lat. 266, also known as Lothar's Gospel Book («Évangiles de Lothaire»)16. This manuscript comes from the scriptorium of Tours, of which it shows the wellknown hierarchy of scripts (including half-uncial beside square capital and uncial, and of course caroline minuscule), and the style of rich ornamentation. Lothar's portrait, sitting on the throne and flanked by two soldiers, is found on fol. 1*v*, preceding an inscription in verse that mentions the making of

noël, *Évangiles de Lothaire*, in Laffitte – Denoël, *Trésors carolingiens*, pp. 102-103, n. 12.

<sup>11</sup> On Waldalbert of Prüm Brunhölzl, *Histoire de la littérature*, pp. 67-70, 504-505. The Martyrology is published in MGH, Poetae II, pp. 567-622. On the Vatican MS Reg. lat. 438 (digitised in < https://digi.vatlib.it/view/MSS\_Reg.lat.438 >): Wilmart, *Codices Reginenses Latini*, 2, pp. 559-560 («Saec. IX-X»).

<sup>12</sup> Utz, *The Master*, pp. 43-53. 13 Bernhard Bischoff dated the Vatican manuscript to a few years after the mid-ninth century (Bischoff, *Katalog*, 3, n. 6678: «Bodenseegebiet, IX. Jh., kurz nach der Mitte»); therefore he identified the king as Louis the German (Bischoff, *Bücher*, p. 189). Utz, *The Master*, p. 43 mentions Louis the German, or Charles III, or Arnulf of Carinthia. The figure in Reg. lat. 438 as portrait of Lothar continue to be taken for granted by other scholars (e.g. Sot, *Références*, p. 18; Poilpré, *Le portrait*, p. 325). 14 According to Haubrichs, *Neue Zeignisse*, p. 3, the MS Reg. lat. 438 shares with two more

copies of the Martyrology a common archetype, that probably dated around the mid-ninth century or a few years later.

<sup>15</sup> It is known that the portraits of sovereigns in books appear for the first time under Charlemagne's heirs. Particularly in the fifth decade of the ninth century, under Lothar and Charles the Bald, they are for the first time found in liturgical or biblical books: Kessler, *A Lay Abbot*, pp. 653-654. Moreover, these representations seem to tend to present a realistic portrayal of the emperor's features: Laffitte, *«Portraits»*, p. 31. 16 Bischoff, *Katalog*, 3, n. 3980, < https://gallica.bnf.fr/ark:/12148/btv1b8451637v >; C. De-

the codex at Lothar's request and under the supervision of a Sigilaus (fol. 2*r*), and an illustration of Christ in Majesty on fol. 2*v*17. This manuscript has been traditionally dated to the time-span 849-851, as a celebration of Lothar's and Charles the Bald's reconciliation at Péronne (and before Irmingard's death, mentioned as alive in the dedicatory inscription)18. In more recent years, it has been backdated to right after 842, as a practical sign of the emperor's closeness to the *confraternitas* of monks in S. Martinus of Tours19. It could therefore be one of the first portraits of a sovereign in a Biblical manuscript, together with the other Lothar portrait in Lothar's Psalter in London, one of the five luxury manuscripts of the so-called *Lothar-Gruppe*.

According to traditional scholarship – which has generated a vast literature to which the following pages will refer just in basic terms – this group of display codices was produced at Lothar's court between the fifth and sixth decades of the ninth century, and allegedly continued and renewed the artistic tradition of the so-called Charlemagne's court school, that seemed to have suffered kind of setback under Louis the Pious20. More generally, they perpetuate the habit of producing luxury manuscripts for both royals and eminent personalities of the Carolingian establishment, as either a supply for their chapels, or books for personal devotion, or gifts to ecclesiastical institutions21.

The *Lothar-Gruppe* includes five complete manuscripts, namely:


The term of *Lothar-Gruppe* was used for the first time in 1927 in a study on the Padua Sacramentary, with reference to the latter together with the

<sup>17</sup> Poilpré, *Maiestas Domini*, particularly pp. 265-267; Sot, *Références*, pp. 23-25. The dedicatory poem is edited in MGH, Poetae II, pp. 670-671.

<sup>18</sup> For example, McKitterick, *The Carolingians*, p. 156. <sup>19</sup> Wagner, *Die liturgische Gegenwart*, particularly chapter 3 (Kaiser Lothar I in seinem Evangeliar für St. Martin von Tours [Paris, Bibliothèque nationale de France, Fonds latin 266]), pp. 163-203.

<sup>20</sup> Mütherich, *Book Illumination*, particularly pp. 594-595. <sup>21</sup> Caillet, *Caractères*, particularly pp. 11-13; McKitterick, *Royal Patronage*, pp. 116-117. <sup>22</sup> Bischoff, *Katalog*, 1, n. 445. <sup>23</sup> Bischoff, *Katalog*, 1, n. 449. 24 Bischoff, *Katalog*, 3, n. 6811, < https://digi.vatlib.it/view/MSS\_Urb.lat.3 >. 25 Bischoff, *Katalog*, 2, n. 2406, < http://www.bl.uk/manuscripts/FullDisplay.aspx?ref=Add\_

MS\_37768 >. 26 Bischoff, *Katalog*, 3, n. 3892; a monograph on this codex was published in 2005: *Liber Sa-*

*cramentorum Paduensis*. See also *I manoscritti miniati*, pp. 63-69.

Gospel Book Berlin260 and the Psalter Add3776827. Wilhelm Koehler's and Florentine Mütherich through their research led to extending the group to the five codices mentioned above28. In 1990, Rosamond McKitterick added to the group one folio from another Gospel Book, containing the beginning of Matthew's Gospel, which she had examined right before it was auctioned in 198829.

The grouping of these manuscripts mainly rested on art historical criteria, and particularly on the Franco-Saxon style of their initials – either in gold, and/or in gold outlined with minium and/or with brown fillings, and/or on a coloured background, as in the London Psalter30. Furthermore, specific similarities have been pointed out between Berlin260 and UrbLat3 as regards the structure of the canon tables31.

As for the two manuscripts containing full-page illustrations, that is Berlin260 and Add37768, the existence of common models, or the following of a common style, seems more problematic: whilst the representations of the four Evangelists in Berlin260 can be related to the style of Rheims as shown, for example, in the famous Vienna Coronation Gospels, the portraits of Lothar himself, David and Jerome on fols. 4*r*, 5*r*, 6*r* of Add37768 can be traced to different models, the one of Lothar's possibly going back to the Late Antiquity32.

Since two of these manuscripts – Add37768 and PaduaD47 – contain an explicit reference to Lothar, not only does the whole group continue being referred to as the production of Lothar's court school, but also Aachen is frequently pointed out as the physical place where they were allegedly copied and illuminated33.

In the following paragraphs I will try to present some considerations on both the actual consistency of this group, and its connection to Lothar, and the idea of a Lothar's court school.

First of all, the connection to Lothar is undeniable only for the Psalter Add37768: it contains on fol. 4*r* a portrait of the emperor34 and a dedicatory poem in *capitalis rustica* on fol. 3*v* mentioning a Byzantine embassy to Lothar that probably took place in 842 – so giving a valuable *terminus post* 

<sup>27</sup> Mohlberg, *Die älteste erreichbare Gestalt*, pp. xxxiii-xxxiv: «Wir haben eine Gruppe liturgischen Handschriften vor uns, die wir Lothargruppe nennen können».

<sup>28</sup> Koehler – Mütherich, *Die Hofschule Kaiser Lothars*. 29 McKitterick, *Carolingian Uncial*. 30 Koehler – Mütherich, *Die Hofschule Kaiser Lothars*, pp. 17-19. 31 *Ibidem*, pp. 16-17. <sup>32</sup> *Ibidem*, pp. 20-23. A full account on Berlin260 and Add37768 will soon be published by Lawrence Nees in his book *Frankish Manuscripts 7th-10th century*, now in press. I am very grateful to Professor Nees for kindly sending me the pre-proof versions of his entries for the two manuscripts, as well as for fruitfully sharing with me his ideas and papers on schools and networks (see also below, notes 52 and 53).

<sup>33</sup> For example, Laffitte, *Les manuscrits impériaux*, p. 87; Crivello, *L'arte e l'immagine*, pp. 86-87.

<sup>34</sup> Kessler, *A Lay Abbot*, p. 653.

*quem* for its dating35. Whether this Psalter was made either for Lothar or for one of his sisters, as suggested by the prayer added on fols 2*v*-3*r*, it can undoubtedly be related to Lothar's court and Lothar's books.

An explicit mention of Lothar is present in the Sacramentary PaduaD47 too, namely in the last folio and lines of the Sacramentary itself (136*v*), inside the text of the *Exultet*: «Memorare Domine famulum Hlotharium imperatorem»36. In this case, it can only be said that this manuscript was produced during Lothar's time, and particularly between 840 and 850, after Louis the Pious' death and before Louis II's coronation, when Lothar was the only emperor in charge37. Although we are undisputably dealing with a luxury manuscript produced for a high-ranking person, it is not certain that Lothar was its recipient. As a matter of fact, the Italian court of Louis II has been suggested38.

The Gospel Book Berlin260 is the third manuscript of the group that is considered to be closely related to Lothar. Currently, it is also known as the Evangeliary of Prüm, and identified with the Gospel Book presented by Lothar to the abbey of Prüm, when he entered it as a monk in the last years of his life. As it is known, in 852 Lothar gave the abbey two manuscripts – a Gospel Book with a precious binding, and a decorated Bible –, several relics and some liturgical objects39. The 1003 inventory of the Treasure of Prüm lists «Evangelia IIII cum eo quod dominus Lotharius dedit, ex quibus unum interius et exterius aureum»40. The identification of Berlin260 with Lothar's gift to Prüm dates back to 1967, and to a careful reconstruction by Herman Knaus of the history of what had been to that day known as the Evangeliary of Clèves41. In point of fact, we know that two luxury Gospel Books now in the State Library of Berlin come from Prüm, Berlin260 and Theol. Lat. Fol. 733; the latter was made in Tours in Vivian's time (mid-ninth century)42 and includes in the last folios (233*v*-234*v*) some later notes concerning the abbacy of Prüm, among which is one about Lothar entering it in 852 and offering some unspecified gifts43. Which of these two Evangeliaries of Prüm was actually Lothar's gift remains under discussion.

<sup>40</sup> Gottlieb, *Ueber mittelalterlichen Bibliotheken*, p. 65, n. 163. 41 Knaus, *Rheinische Handschriften*; Fingernagel, *Die illuminierten lateinischen Handschriften* (1991), 1, pp. 73-76; 2, ill. 225-234. 42 Bischoff, *Katalog*, 1, n. 470; Schillman, *Verzeichnis*, pp. 94-100. 43 Fol. 234*r*: «Anno dominice incarnationis DCCCLII, indictione XV, adveniens Lotharius im-

<sup>35</sup> Edited in MGH, Poetae VI/1, p. 163. 36 An image in *Liber Sacramentorum Paduensis*, Fig. 8. As it is known, PaduaD47 is a composite manuscript, the Sacramentary being its first codicological unit (ff. 1-138), the second one being a Martyrology (ff. 139-148).

<sup>37</sup> Crivello, *Origine*, p. 59. 38 Koehler – Mütherich, *Die Hofschule Kaiser Lothars*, p. 12. 39 MGH, DD Lo I / Lo II, n. 122, pp. 280-281: «evangelium scilicet cum ebore, cristallo atque auro gemmisque compositum; bibliothecam cum imaginibus et maioribus carachteribus in voluminum principiis deauratis».

perator Prumiam monasterium quod est constructum in honore Domini et salvatoris nostri Iesu Christi nec non et ostriles eiusdem Dei et domini nostri beatissimi quoque Iohannis Babtistę et Pauli ceterorumque apostolorum, Stephani quoque prothomartiris cunctorumque sanc-

In any case, Berlin260 has ever since the beginning of the last century been grouped together with Lothar's Psalter and the Padua Sacramentary, due to their many similarities not only in the decoration but also in the *mise en page* and the script, by the same hand in the three manuscripts according to Bernhard Bischoff. Actually, PaduaD47 and Add37768 are closer in size, whereas Berlin260 is decisively bigger, but Berlin260 and PaduaD47 show similar dimensions of the writing frame; the writing frame itself consists of a similar, though very common, pattern in the three codices, with a double line on the left and on the right of the written surface (and a further framing in Berlin260), 22 lines for the writing in Berlin260 and 21 in PaduaD47 and Add3776844. As for the writing, which is traced in gold in the three manuscripts, I would not take for granted either the identity of the hands responsible for the copy of every manuscript, nor the presence of just one hand in each of them45. Figg. 1-3 could conceivably prove both morphological differences and, at the same time, the stylistic similarities in the scripts of the three samples: a caroline minuscule quite irregular in the alignment, with letters that are basically wider than high, compliant with the canon, although apparently written by not thoroughly confident hands46.

According to Koehler and Mütherich, the MS UrbLat3 was also copied by the same scribe as Berlin26047. But its script is a different, rather stiff and somehow unnatural caroline minuscule, with at least one change on fols. 130- 131 to an equally artificial and at the same time hesitant hand – as regards, for example, the incline of the axis (Fig. 4). The uncial of the display scripts shows a similar lack of confidence. Bischoff himself dated the script of this manuscript to the third quarter of the ninth century, thus implicitly excluding its belonging to the *Lothar-Gruppe*48.

As for the Gospel Book Berlin3, some eccentricities of which had already been noticed by Koehler and Mütherich, it has been more convincingly withdrawn from the group at the end of the 20th century on art historical grounds,

torum martirum, Martini etiam et Benedicti venerabilissimorum confessorum cunctorumque sanctorum, anno imperii in Italia XXXII et in Francia XIII et optulit hęc mente devota sancto Salvatori et omnibus prefatis sanctis pro remedio animę suę et coniugis defunctę prolisque et omnium predecessorum suorum pro statu regni».


<sup>44</sup> Koehler – Mütherich, *Die Hofschule Kaiser Lothars*, pp. 38, 48, 53.

<sup>45</sup> I make these observations on the basis of a personal analysis of Berlin260 and of the digitization of Add37768. Unfortunately, I could not examine Padua D47 due to the closure of the Chapter Library.

<sup>46</sup> As far as I know, the only palaeographical description of the script of the *Lothar-Gruppe* was made for the Padua Sacramentary: Mohlberg, *Die älteste erreichbare Gestalt*, pp. XVIII-XXII;

Lazzarini, *Un sacramentario*, pp. 37-39; Martini, *Descrizione*. 47 Koehler – Mütherich, *Die Hofschule Kaiser Lothars*, p. 15. <sup>48</sup> Bischoff, *Katalog*, 3, n. 6811: «etwa Nordostfrankreich, IX Jh. 3. Viertel».

Fig. 1. Berlin, Staatsbibliothek zu Berlin – Preußischer Kulturbesitz, Ms. theol. lat. fol. 260, fol. 182*r*.

Fig. 2. © [2022] Biblioteca Capitolare di Padova, D 47, fol. 93*r.*

Fig. 3. © British Library Board. London, British Library, Add. 37768, fol. 10*r.*

Fig. 4. Città del Vaticano, Biblioteca Apostolica Vaticana, Urb. Lat. 3, fol. 130*r.*

which connect it more with the style of the so-called Charlemagne's court school than with the one shown by the other codices of the *Lothar-Gruppe*49.

Taking into account all these considerations, the criteria for inclusion of the manuscripts in the *Lothar-Gruppe* might seem elusive and notably grounded in the authority of the tradition and of the scholars who dealt with them.

On the one hand, the similarities between the five manuscripts do not seem to be exclusive of the group, as regards both the widespread – in North-Eastern France in the central decades of the ninth century – Franco-Saxon style of the initials and the artistic features of the illustrations, that have in part been attributed to the school of Rheims, as well as a number of other display codices of different content and provenance50. On the other hand, some differences in style, script and features of the manuscripts of the *Lothar-Gruppe*, of which Albert Boeckler in 1930 was already aware51, make it difficult to locate their origins in a specific place or at a specific writing centre. Furthermore, Lothar's patronage or possession can only be proved certain for the London Psalter.

All of this will not surprise if we consider, as recent scholarship tends to do, the so-called court school not as a physical place but rather as «neworks of men and books»52, clusters of illuminators and perhaps scribes working together on common projects, the illuminators not seldom moving from and to

<sup>51</sup> Boeckler, *Abendländische Miniaturen*, p. 30. 52 Nees, *Imperial Networks*, particularly p. 92.

<sup>49</sup> Fingernagel, *Die illuminierten lateinischen Handschriften* (1999), 1, pp. 133-135; 2, ill. 360- 363, 396.

<sup>50</sup> Not by chance the second part of Koehler – Mütherich, *Die Hofschule Kaiser Lothars* (pp. 71-115) is about other single manuscripts from Lotharingia.

different centres, the full-page illuminations themselves possibly being sent to complete codices copied elsewhere53.

In this sense, the *Lothar-Gruppe* can become just a fraction of a wider manuscript production for a royal or, more generally, a high-profile patronage that took place in Lothar's time; and Lothar could have been only one among possible recipients of books, or perhaps the recipient of books that no evidence can connect to him anymore.

## 3. *Manuscript production in Lothar's Italy*

Which books were copied and circulated in Northern Italy during the years of Lothar's government, which essentially coincide with the second quarter of the ninth century?

When it comes to Carolingian Italy, identifying manuscripts of documented origin is rather difficult. On the one hand, apart from some well known ecclesiastical libraries, the surviving manuscripts are often scattered across different institutions with no traces left of their previously belonging to other collections54. Therefore, not only reconstructing the activity and production of specific writing centres is more difficult, but any attempt to do so can also have an unexpected, frustrating outcome in terms of (not) shared graphic tendencies55. Bernhard Bischoff's many uncertainties in proposing an Italian origin for ninth century manuscripts confirm this difficulty. Moreover, due to praxis and tradition or simply by chance, and so quite differently from the situation beyond the Alps, most of the presumably Italian manuscripts lack any explicit data concerning their origin or provenance apart from, sometimes, the disappointingly common scribe's name.

I recently investigated the bishops' libraries in the Early and High Middle Ages, on the basis of the manuscript evidence, that is of both library inventories and booklists, and *ex libris*, colophons and notes mentioning the bishops as owners, patrons or donors. Whilst, for the regions beyond the Alps, these kind of data allow one to piece together significant sets of manuscripts that the bishops of the ninth century owned, commissioned or presented to their cathedrals or other ecclesiastical institutions, the Italian situation appears

<sup>53</sup> Nees, *Early Carolingian Manuscripts*, p. 172: «We should pay more attention to who produced books, and for whom they were produced, and perhaps pay less attention to where they might have been produced, shifting emphasis to scribes, artists, patrons and readers and away from abstracted localized centers of production which are in many instances as much or more a product of modern imagination as of medieval evidence». On the reconsideration of the idea of court school see also Caillet, *La classification*, and again Nees, *Networks or Schools?*. 54 I discuss this issue in Pani, *Manuscript Production*, p. 277. 55 I refer here, for example, to the significant yet unpublished work by Rosso, *I manoscritti*,

especially pp. 34-79 on ninth-century manuscripts from Bobbio and the difficulties in pointing out graphic similarities and common tendencies as expected for a scriptorium traditionally regarded as one of the most organized and productive writing centres of Carolingian Italy.

very incomplete56. As a matter of fact, the Vatican MS S. Maria Maggiore, 43, a Gregory the Great's *Regula pastoralis* copied at the request of the bishop of Priverno Martinus around 861, is the only such case for the ninth century (and is later than Lothar's years)57. There are of course groups of books that can be attributed to the action of specific bishops – as Miriam Tessera's paper in this volume shows – but this fact does not rest on explicit data, at least as far as Lothar's years are concerned. Therefore, any attempt to investigate the Italian Carolingian manuscript production must deal with the common sensation of a lack of "anchors".

There are, though, both single manuscripts datable on internal grounds, and groups of manuscripts that can quite unmistakably be related to the activity of specific and well-defined (and definable) writing centres, such as as Verona or Nonantola.

Some law books, for example, contain on their last folios additions that offer not only the *termini post quem* and *ante quem* for their original parts, but also samples of precisely datable scripts.

The famous MS Sankt Paul im Lavanttal, Stiftsbibliothek, 4/1, a book of *leges* and capitularies, was copied after 818-819, perhaps outside Italy – excerpts from the *Capitula per se scribenda* and the *Capitula legibus addenda* being the latest among the texts copied by the main scribe – but before 825, when the *Capitulare Olonnense mundanum* was added on fols 182*r*-183*v*58.

The MS Ivrea, Biblioteca Capitolare, XXXIV (5) is a slightly later collection of Carolingian capitularies and Lombard laws, datable to the time-span 825-832: whilst the *Capitulare Olonnense mundanum* and the contemporary *ecclesiasticum* one belong to the original core of the book, copied by different, more or less cursive hands (e.g. fols 21*r*, 27*r*), Lothar's *Capitulare Papiense* is added by a different – and younger – hand on fols 166*v*-167*v*59.

Lastly, the Vatican MS Lat. 5359, another collection of Lombard laws and Italian capitularies from many different, unsteady and unskilled hands, offers just the *terminus post quem* of 832 for its copy, as it contains Lothar's *Capitulare Papiense* on fols 142*v*-145*v*. In any case, it was presumably copied right in the following years, with a little chance of a later dating as proposed by Bischoff60.

What strikes one in the scripts of both the original parts and the additions of these manuscripts – all datable to the second quarter of the ninth century, as shown above – is their remoteness from the canons of the caroline

<sup>56</sup> Pani, *The Bishops' Libraries*, in print. 57 < https://digi.vatlib.it/view/MSS\_S.Maria.Magg.43 >. <sup>58</sup> Mordek, *Bibliotheca*, pp. 685-695; Bischoff, *Katalog*, 3, n. 5934, < http://www.leges.uni-koeln.de/en/mss/codices/st-paul-abs-4-1/ >. <sup>59</sup> Mordek, *Bibliotheca*, pp. 178-185; Bischoff, *Katalog*, 1, n. 1561, < http://www.leges.uni-koeln.

de/en/mss/codices/ivrea-bc-xxxiv-5/ >. <sup>60</sup> Mordek, *Bibliotheca*, pp. 881-883; Bischoff, *Katalog*, 3, n. 6903; < http://www.leges.uni-ko-

eln.de/en/mss/codices/vatikan-bav-vat-lat-5359/ >. On this manuscript see also the recent Garipzanov, *Magical Charaktêres*, pp. 287-308.

Fig. 5. Sankt Paul im Lavanttal, Stiftsbibliothek, 4/1, fol. 182*r.*

minuscule. All these hands, whether skilled or unskilled, seem still to adhere more to the early-medieval, pre-caroline graphic tradition than to a renewed graphic system (Figg. 5-7).

It must be noticed that such manuscripts as the law books discussed here can perhaps be referred to non-ecclesiastical contexts of book production – such as counts' or notarial offices –, whose existence, though elusive, is more and more often suggested for the Early Middle Ages as regards the copying of at least certain types of books61.

If these ill-identified "lay" centres show a tardiness in the reception of the new graphic models, due to a graphic education rooted in a different epoch and in different models, and addressing different purposes than the copying of books, what can then be said about some ecclesiastical institutions located in what is supposed to have been the core of the Carolingian domination in Italy, notably Verona or Nonantola?

With about 70 surviving manuscripts datable to the ninth century, all clearly recognizable in terms of codicological and especially graphic features, the scriptorium of the Cathedral library of Verona could be regarded as a sort of prime example of the Italian book production in Lothar's years, especially because it would basically coincide with the so-called "Pacificus age", a phase of intense activity for the copying of books, under the supervision of the archdeacon Pacificus († 846).

<sup>61</sup> See, for example, McKitterick, *Some Carolingians Law-Books*, pp. 13-14, 22-23; Nicolaj, *Ambiti di copia*, pp. 354-357.

Fig. 6. Ivrea, Biblioteca Capitolare, XXXIV (5), fol. 7*r.*

In this regard, I would note that not only Pacificus' role must be – and actually has been – downplayed62, but also that the "Pacificus' age" might in reality exceed Pacificus' life (and so Lothar's epoch) in paleographical terms. In other words, that the graphic features normally related to the book production under Pacificus persisted throughout the ninth century, and therefore it is possible/likely that some of the manuscripts dated to Pacificus' age might have been, in reality, copied later63.

In any case, what is traditionally defined as the "Veronese caroline minuscule" – and therefore a testimony to an early adoption of the new Carolingian book script by the Veronese scriptorium64 – can hardly be considered a Caroline minuscule: with the persistence of other cursive ligatures than &, *ct* and *st*, and of half-uncial letters such as *a* and *g*, it is a typised script, used across the whole ninth century by a high number of scribes, much more inspired by the old half-uncial script and its Alemannic filiation than by the reformed Carolingian handwriting65.

Caroline minuscule is absent from Verona until the mid-ninth century («sembrerebbe […] plausibile che di carolina, a Verona, si possa concretamente parlare non prima della metà del secolo

<sup>62</sup> La Rocca, *Pacifico*, is a well-known essential work on this topic. <sup>63</sup> The MS Verona, Biblioteca Capitolare, XLVII (45) (Bischoff, *Katalog*, 3, n. 7042) is a significant example. It contains John the Deacon's *Vita Gregorii*, a text composed around 873-875. It is copied, arguably in the last decades of the ninth century, by several hands, some of which still show the graphic features normally related to Pacificus, in terms of both general appearance of the script and the shape of the single letters. Furthermore, many of the *marginalia* in Veronese codices normally attributed to Pacificus' hand (by Bischoff too) are by different hands, and therefore cannot *a priori* count as *termini ante quem* for the dating. 64 This is still Paolo Cherubini's and Alessandro Pratesi's opinion: Cherubini – Pratesi, *Paleo-*

*grafia latina*, pp. 385-386. <sup>65</sup> On this topic Santoni, *Scrivere documenti*, especially pp. 207-211. According to Santoni the

Fig. 7. Città del Vaticano, Biblioteca Apostolica Vaticana, Vat. Lat. 5359, fol. 143*v.*

In 2009, in a fundamental multi-author monograph on the famous MS Cesena, Biblioteca Malatestiana, S.XXI.5, of Isidore's *Etymologiae*, Anna Bellettini wrote: «L'impressione (…) è che si tratti del prodotto di uno scrittorio importante ma non al centro del rinnovamento carolingio, in cui è in uso una scrittura volta al passato, se si giudica che il manoscritto dovrebbe essere stato scritto in anni in cui altrove la minuscola carolina era già diffusa e affermata»66. Bellettini dated the manuscript to the first half of the ninth century and convincingly suggested Nonantola as the place of its origin. Actually the other manuscripts that can be arguably attributed to Nonantola and the first half of the ninth century show graphic features which, not so unlike those from Verona, reproduce a local pre-Caroline type rather than adhere to the Caroline canon67.

A random check of manuscripts dated to the second quarter of the ninth century, of presumed Italian origin, would confirm the strong presence of pre-Caroline features such as cursive ligatures and a pre-Caroline form of the letters.

If the Caroline minuscule is supposed to have spread across Italy in Lothar's years, the ways and channels of its penetration are yet to be discovered.

IX», p. 211). As I stated above, I believe the end of the century a more convincing chronological term, if not for the appearance of the Caroline minuscule, at least for the persistence in the use of the Veronese type of minuscule.

<sup>66</sup> Bellettini, *Testi e scritture*, p. 22 («My impression is that this manuscript was copied in a *scriptorium* that was not at the core of the Carolingian reform, in which a script was used, that looked back to the past. We must consider that this manuscript was copied in a time when elsewhere the Caroline minuscule was spread and used»).

<sup>67</sup> Palma, *Manoscritti nonantolani*, p. 105, and *passim*; on the Nonantola-type also Cherubini – Pratesi, *Paleografia latina*, pp. 284-285.

#### **Works cited**


in *Committenti e produzione artistico-letteraria nell'alto medioevo occidentale*, Spoleto 1992 (Settimane, 39), pp. 647-675.


Laura Pani Università degli Studi di Udine laura.pani@uniud.it