Between Southern Italy and Dalmatia : Missal MR 166 of the Metropolitana Library , Zagreb Emanuela Elba

The Missal MR 166 from the Metropolitana Library, Zagreb, written in Beneventan script and dating back to the twelfth–thirteenth centuries, has long been considered a Dalmatian product, similar to the coeval illuminated manuscript in Beneventan script preserved in the Trogir Cathedral and originating in Zadar. Nevertheless, later studies — specifically based on the textual features of the manuscript — showed that it is undoubtedly a Southern Italian product, and a significant testimony of the uninterrupted book circulation that existed on both sides of the Adriatic for three centuries, roughly from the eleventh to the thirteenth, thus influencing the activity of the Benedictine scriptoria on the Dalmatian coast. On the basis of the study that makes it possible to define more closely the group of manuscripts that make up the “corpus of the illuminated manuscripts from Dalmatia”, the paper aims to support the Southern Italian origin of the Missal by means of a critical analysis of the theories put forward so far about the “typically Dalmatian” features of its Initialornamentik.

Recent studies have stressed the importance of the Adriatic sea for the exchange of artistic and cultural forms between its eastern and western sides. 1 This paper deals with a manuscript which testifies specifically to the close links between the Eastern Adriatic area and the Southern Italy territories.
With regard to the same issue, research has been conducted on the illuminated manuscripts in Beneventan script from Dalmatia, for the first time gathered in a unified and systematic "corpus" that in a detailed description of the decorative sets, aims to emphasize the essential features of the Dalmatian miniature painting while also focussing on the history of the Adriatic area between the 11 th and 13 th centuries. 2he systematic and comparative analysis of the surviving evidence has shown two main groups of manuscripts.The first one, based on the scriptorium of St. Chrysogonus in Zadar, implies manuscripts dated to between the mid-11 th century and the first decades of the 12 th century.The second one, based on the codices of the Cathedral of Trogir, is dated to the 13 th century.Since the beginnings, these manuscripts were characterized by a strong influence coming from Apulia, mainly from the manuscripts of Bari, and with the arrival of the 13 th century, an influence also began to penetrate from the northern part of the Adriatic Sea, particularly from the area of Venice and Padua.
A completely isolated case and undoubtedly the most problematic of the whole corpus of Dalmatian manuscripts is Missal MR 166 of the Metropolitana Library, Zagreb (the library of the Zagreb Archbishopric, now located in the Croatian State Archives), a composite manuscript made up of two different codicological units: the first one, mutilated at the end, is characterized by a Beneventan script, which is absolutely different from that of the coeval Dalmatian codices, and more similar to the typology of Montecassino rather than to the Bari type.The second one, on the other hand, is characterized by a script that, even though careless and unadorned, seems like a rounded Beneventan script and therefore it is more similar to the typology of the manuscripts of Dalmatian origin.The decoration of the missal clearly shows that there is a qualitative difference between the two parts of the Missal and there is no doubt that the second one was decorated later by a less skilled hand.
The decoration of the first unit of the codex includes a page with the T deriving from the Te igitur (p.209) on a colour-striped background (fig. 1) and 179 decorated initials.There are initials with geometric and ribbon-like forms (fig.2-4), zoomorphic initials (fig.5-8), and initials with squared interlaces (so-called "a mattonella") (fig.9-10).In just two cases, there are initials of the so-called "Ottonian type", used  (Castel Sigismondo, Piazza Malatesta, Rimini, 19  agosto-29 dicembre 2002), ed.F. Flores d'Arcais, G. Gentili, Milano 2002; Arte per mare.Dalmazia, Titano e Montefeltro dal primo cristianesimo al Rinascimento (San Marino Citta, Museo di san Francesco, 22 luglio-11 novembre 2007), ed. G. Gentili, A. Marchi, Milano 2007. 2 The forthcoming corpus is based on my Phd Thesis: Sulle relazioni culturali tra le due sponde adriatiche.La decorazione dei codici in beneventana della Dalmazia tra XI e XIII secolo.I wish to express here my gratitude to Prof. Maria Stella Calo Mariani, (Universita di Bari), Head of the Doctoral Program, and Prof. Giulia Orofino (Universita di Cassino), my Tutor, for their invaluable, constant support to this research.I am grateful as well to Prof. Valentino Pace (Universita di Udine) for his editorial comments.My present essay is dedicated to the late Virginia Brown, who devoted her attention to the Missals in Beneventan script.For a preliminary overview of my research, v. E. Elba, La decorazione dei codici in beneventana della Dalmazia tra XI e XIII secolo, Segno e Testo 4 (2006) 107-147.
to mark the beginning of the masses, the evangelic pericopes and the lessons.
The initials can all be attributed to the same illuminator, apart from those corresponding to the palimpsest pages (p.160-163), where the Mass-text pro imperatore was substituted by a Mass-text pro rege.Most likely these pages were illuminated by a second artist, at the same time when the text of the scriptio inferior with the Mass-text pro imperatore was written.Those done by the first artist are written with a dark inked pen.They are coloured red, yellow, blue, dark green, light green, white and black.The geometrical initials, done by the second artist on the pages 160-163 are drawn with a dark inked pen.At some points, they are marked with red lines and coloured with red, yellow and gold.
Alongside these initials, there are some others belonging to a simpler typology.They are characterized by the development of coloured backgrounds and an irregular profile created by bowlike prominences and the extension of the external lines with plant-shaped endings (fig.11).However, close to the text with neumatic notation (p.173-224), there is a set of small red inked letters drawn on a square golden background.
The decoration of the second part of the manuscript is far poorer.It only includes 35 decorated initials: they are simple letters, with characteristic bars animated by nodular, bowlike prominences.With a pen, their contours were drawn using a sepia coloured ink and simply filled in with red.
Missal MR 166 was discovered by the palaeographer Viktor Novak in December 1916.He described it in detail, attributing it to a southern Italian Scriptorium of the late 11 th century for codicological and palaeographical reasons. 3hile the Italian origin of the manuscript had been confirmed by the liturgist Dragutin Kniewald, 4 Novak himself revised his initial hypothesis after about thirty years.He claimed that the manuscript should have been produced in Dalmatia, basing his new hypothesis on two different considerations. 5The first one was that the "semi-angular" script was not unusual in the context of 11 th century Dalmatian library production, in which it was characteristic to use both types of script, as shown by the Gospel Book of the Cathedral of Trogir, mistakenly attributed by him to the 11 th and 12 th centuries.The second consideration was that on the palimpsest pages of the Missal (pp.160-161), the mention pro imperatore of the scriptio inferior, replaced with a mention in favour of a king -in whose initials "Em" Novak identified the abbreviation of the Hungarian king's name Emericus (1196-1204) -coincides with the same mention which exists in the Exultet of the Evangeliary of Zadar (Bodl.Canon.Lat.61).This manuscript is considered without any doubt to be of Dalmatian origin and datable, according to Novak, to the last decade of the 11 th century, on the basis of the allusion to the emperor.In fact, after the death of King Zvonimir (1076-1089) there was a long period of political uncertainty, during which the power of Byzantium strengthened.According to these circumstances, Novak claimed that the dating to the last decade of the 11 th century excludes the possibility that MR 166 was produced in Italy, since the allusion to the emperor could be explained only in the 'unlikely' case of the completion of the manuscript before the fall of Bari to the Normans in 1071.6 Fig. 1.Metropolitana Library, Zagreb.MR 166, p. 209   Fig. 2. Metropolitana Library, Zagreb.MR 166, p. 103  Branka Pecarski, a student of Novak, supported his ideas and developed a monographic study based entirely on the decorative set of the manuscript.7She tried to demonstrate that the codex, while characterized by clear references to the miniature painting of southern Italy, shows evident connections with other Dalmatian manuscripts, in particular with the Evangeliary of Zadar.In addition, following Novak's hypothesis about the existence of a scriptorium that in the 11 th century already used the semi-angular script in Trogir -when the rounded script was used in Zadar -she concluded that MR 166 was probably produced right in Trogir.
The subsequent studies disagree with the thesis claiming the Dalmatian origin of the manuscript and its dating to the 11 th century, unanimously maintaining that MR 166 was done in the area of Montecassino during the 12 th century and that only later was it exported to Dalmatia, where it was completed with a final part, presumably at the beginning of the 13 th century. 8is position, based on the musicological considerations by Boe, 9 was further supported by the studies of Virginia Brown on the "votive Missal", 10 a type of monastic book that was particularly widespread in the area of Benevento and Montecassino from the 11 th century.The "votive Missal" differs from the Sacramentary because of the presence of Masses-texts ascribed to the domain of private worship and a heterogeneous set of prayer intentions that belong to public and private worship. 11The preliminary comparative analysis by Virginia Brown on the votive Missals of southern Italy, produced between the 11 th and 13 th century, 12 pointed out that MR 166, considered as one of those examples, is very similar to the codices of Montecassino.This means that they were probably produced in the same scriptorium or in a monastery under its control. 13 .4. Metropolitana Library, Zagreb.MR 166, p. 149  Fig. 3. Metropolitana Library, Zagreb.MR 166, p. 63  The analysis of the manuscript decoration contained in this paper on the one hand calls into question the hypothesis of the Dalmatian origin; on the other hand, it aims to confirm the thesis of a southern Italian origin currently supported by the scholars.
After closer observation, it is possible to identify some specific elements belonging to the ornamental repertoire, such as the vegetal endings made up of lanceolate leaves, and rarely of curled leaves, or buds.There are zoomorphic appendices, characterized by protomes of a dog or bird of prey, or by full or half-length bodies of dogs with the coat marked with small red signs on the back.Finally, there are anthropomorphic heads that in two cases are located at the end of the letter.
Pecarski's analysis aimed to underline the relationship between the Dalmatian miniature painting tradition and that of southern Italy through the systematic classification of the Initialornamentik of the manuscript and a direct comparison (according to typology and single ornamental elements) with some of the most important Beneventan illuminated manuscripts. 14However, the methodological validity of her analysis fails when she tries to demonstrate, with the same emphasis, the affinity of some decorative elements with codices that were definitely created in Dalmatia, the most important testimony of which is the Bodl.Canon.Lat.61.
The anthropomorphic heads and the dark backgrounds filled with small blank discs (in Italian so-called "motivo a occhi", i.e. eye-like) are the elements that the two manuscripts MR 166 and Bodl.Canon.Lat.61 share, both directly derived from the manuscripts in Bari type. 15In the case of MR 166, they do not characterize the decoration of the manuscript at all, while in the Evangeliary of Zadar Bodl.Canon.Lat.61 they systematically appear, showing the reiteration of some typical elements of the illuminated manuscripts from Apulia, probably used for years in the scriptorium of St. Chrysogonus (figs.8, 12).In addition, the heads, in profile with a beard under the chin or with peaky ears (the features of a satyr rather than a human being), differ remarkably from those which are characteristic of the production of Zadar and Dalmatia.These heads are in profile (or frontal), with thick hair defined by a rounded mass often surmounted by a head covering or a crown -so related to the model of Bari that, in some cases, they can be considered almost identical. 16Fig.6.Metropolitana Library, Zagreb.MR 166, p. 25   Fig. 5. Metropolitana Library, Zagreb.MR 166, p. 128   14 Pecarski subdivided the initials into four categories (Telebakovi}-Pecarski, Iluminacija, 150-152).The first category (composed of two groups), included the bow-shaped and geometrical letters ("In the first group the letters consist of stems, of ribbons intertwined in various ways, and zoomorphic motives… The initials of the second group consist exclusively of interlaced multi-coloured ribbons with spaces in between, most often filled with pearls"): for those she referred to the initials of the Exultet rolls of Avezzano.In this scroll she found many direct similarities, such as the initials O (probably initials with squared interlaces).The second category was composed of simpler initials characterized by the "the omission of the interlace".Drawing these initials "the illuminator rounds them off, splits or tapers them, and adds fine strokes; sometimes even a highly stylized leaf".To the third category belong initials characterized by a "ornamentally floral character", coloured with only blue and red, some traces of green and mainly gold (among these is also the monogram Vere Dignum).For these initials, identifiable with those of the Ottonian type, the proposed comparisons, based on the features of their general structure, were: the Vat.Lat.1202, the Vat.Barb.Lat.529 and the Exultet roll of the British Library (Add.MS 30337).The Exultet rolls of Pisa and Fondi were used for a comparison of the similar shape of the interwoven ribbons. 15Ibid., 158.The recent contributions to the study of the Apulian miniature painting coincide with this remark of Pecarski.They claim that the head motif can be considered as the "trademark" of the manuscripts manufactured in the scriptorium of Bari, such as the Exultet scrolls of Bari.See: G. Orofino, Miniatura in Puglia agli inizi dell'XI secolo: l'Omiliario VI B 2 della Biblioteca Nazionale di Napoli, Miniatura 3-4 (1990-1991)  21-32; eadem, L'illustrazione delle Metamorfosi di Ovidio nel ms.IV F 3 della Biblioteca Nazionale di Napoli, Ricerche di Storia dell'Arte 49 (1993) 5-18.For the Exultet scrolls of Bari cf.Exultet.Rotoli liturgici del medioevo meridionale (Catalogo della mostra, Montecassino, 1994), ed.G. Cavallo, G. Orofino, O. Pecere, Roma 1994. 16 In the Dalmatian illuminated manuscript tradition, the head motif appears not only in the Evangeliary Bodl.Canon.Lat.61, but also in the slightly earlier Monastic Book of Hours Ms. K. 394 of the Majar Tudomanios Akademia of Budapest and in the Missal Lat.fol.920 of the Staatsbibliotek of Berlin, dating back to the 12 th century.As already mentioned, the close comparison, particularly in the case of the heads in the The comparison with Bodl.Canon.Lat.61 fails completely when analyzing the Byzantine elements that characterize -according to Pecarski -the fauna of the zoomorphic initials, in particular the letter A on page 249 and the letter G on page 128 (fig.5, 7).Rather than the Byzantine models, the animals of the Missal -dogs, birds of prey and even a dragon -belong to the decorative model of the area of Benevento and Montecassino -revitalized by the contribution of Norman art. 17Furthermore, there are no peacocks, birds that preponderantly differentiate the decoration of the Oxford manuscript and, "typical of Byzantine art, not so freely or independently applied in southern Italy," they can be considered as distinctive features of the codices of Zadar, dating back to the second half of the 11 th century (fig.13). 18n contrast with what Pecarski claimed, the fact that the decoration of the Missal of Zagreb is completely different from that of the Oxford Evangeliary shows the "uniqueness" of the manuscript compared to the other more or less coeval Dalmatian examples.These Dalmatian manuscripts, even though they were not created by the same scriptorium -like in the case of the Vat.Borg.Lat.339 -clearly show the diffusion of a homogeneous decorative language on the whole of Dalmatian territory, based on the illuminated manuscript production of St. Chrysogonus.As a result, because of the lack of comparable manuscripts, it is impossible to affirm that MR 166 was produced in Dalmatia, even less by a scriptorium such as the one of Trogir, whose very existence, due to the lack of documentary evidence, should be called into question even for the 13 th century codices. 19oing back to the illumination of the manuscript, the morphological and decorative features of the initials point to some elements characterizing a scriptorium operating in the area of Montecassino, according to the text analysis of Brown, or rather to the scriptorium of Benevento, 20 17 This consideration is mainly due to the presence of the initial shaped by the figure of the dragon.It is a typical element of the zoomorphic repertoire of Norman art that spread in Dalmatia thanks to the influences of the models coming from southern Italy and particularly attested by the codices in Beneventan script of the 12 th and 13 th centuries, as the already mentioned Missal of Berlin, the Lectionary and the Evangeliary of Trogir.
18 Telebakovi}-Pecarski, Iluminacija, 158.This opinion is widely supported by the fact that, as opposed to the models of the Bari type, in the codices of Zadar and, particularly in the Bodl.Canon.Lat.61 and in the Bodl.Canon.Lat.277, the peacock, instead of being a simple decorative motif of the zoomorphic repertoire of the manuscript, systematically substituted the eagle, symbolizing the evangelist John.Cf.Elba, La decorazione, 126 (and tab.8 and 11d); eadem, Lungo le rotte, 48 (and figs.8-9). 19The quality of the decoration and the peculiar iconographical affinities that in particular relate the Evangeliary of Trogir both to the miniature painting of southern Italy and to the library production of the 13 th century in the area of Padua and Venice, support the hypothesis that the manuscript, even if done for the cathedral of Trogir, could have been produced in Zadar.Zadar was the full-fledged, leading centre of the book tradition of Dalmatia and it remained so even when the Franciscan scriptorium replaced that of the Benedictines of St. Chrysogonus.Cf.E. according to the research by J. Boe and A. E. Planchart. 21learly the manuscript was influenced by both traditions. 22his is proved, from a compositional point of view, both by the letter V on page 25 (fig.6), shaped by the figure of a dog, lying on its back and licking its paw -reusing a solution very similar to the one used in the codices of Montecassino and Benevento -and by the small letter D on page 154, where two dogs compose the shape of the letter, like in many other examples belonging to the repertoire of Montecassino, in which animals substitute the contour of the letter in a symmetrical position. 23On the other hand, at the ornamental level, the comparison with the miniature painting tradition of Montecassino and Benevento is essentially focused on the presence of the eye-like motif, on the use of vegetal endings with lanceolate leaves and, particularly, on the use of a systematic and differentiated use of squared interlaces that shape the letter O on ff.103-157.There are many similarities with the manuscripts of Montecassino also in these small details. 24t is also likely (but still needs to be checked) that in the case of the manuscripts from Benevento, the major similarities of the decorative solutions (fig.14 The relation with the book models of Montecassino and Benevento also clarifies the technical quality that Fig. 10.Metropolitana Library, Zagreb.MR 166, p. 103   Fig. 9. Metropolitana Library, Zagreb.MR 166, p. 157   Fig. 11.Metropolitana Library, Zagreb.MR 166, p. 152 21 J. Boe (Beneventanum troporum corpus II: Ordinary Chants and Tropes for the Mass from Southern Italy, A. D. 1000-1250.Part 2: Gloria in excelsis, ed.J. Boe, Madison 1990) believes that the analysis of the order of the Gloria hymns and the Tropes follows that of the tradition of Benevento, as the 'prosula' Puer ascendentem, in the f.16v of the Ms. Benev.35.
22 I exclude the connection of the decorative parts in MR 166 with the illuminated manuscripts from Apulia, particularly in Bari type, since there are too many differences in style.Therefore, I do not agree with Golob's claims, when she compares, with reference to the second part of the manuscript in the Beneventan script of the Bari type, the Initialornamentik of the manuscript of Zagreb to the manuscript of Virgil Vat.Reg.Lat.2090;  cf.N. Golob, Ungarn, Slowenien und Kroatien, in: Romanik, II, ed. A.  Fingernagel, Graz 2007, 92 e n. 38.  2V., for example, the small letter D on f. 26r bis of the Ms. 36 of the Biblioteca Capitolare of Benevento or the letter O on page 54 of the Casin.51, reproduced in: Orofino, La miniatura nel ducato di Benevento, tab.XI, pict.22; eadem, I codici decorati dell'Archivio di Montecassino, II/1: I codici preteobaldiani e teobaldiani, Roma 1996, tab.CV. 24 Among the many possible comparisons, this one refers in particular to the capital geometrical letters of Grimoaldo in the Casin.104, such as the letter N on page 155, reproduced by Orofino, I codici cit., II, 2, tab.CXX-d. 25For example, in the case of MR 166, the following elements are normally considered meaningful: the hooked knots like those characterizing the initials of the Ms. Benev.29; the way in which the geometrical letters are structured, characterized by empty sections with a golden background and crossed by ribbons ending with long lanceolate leaves, as in the case of the letter P on f. 149r of the Ms. Benev.20; or the presence of Ottonian plant shoots of the letter V on page 202, contoured by a thin red line and coloured with gold on a blue background, as those that fill the white polka-dot field of the letter T on f. 108v. of the Ms. 19 and on f. 149v of the Ms. 20.
characterizes the decoration of the first part of the manuscript.Both the drawing, well defined and marked by a thin black inked stroke, and the mise-en-page, are typical features of an important scriptorium or, at the same time, of an area implying a wide circulation of manuscripts of top quality.In fact its mise-en-page shows a well balanced attention to the relation of the text and the initials, which may be a clue for its attribution to an important scriptorium or to an area where there was a wide circulation of top quality manuscripts.Waiting for an accurate evaluation of the palaeographical aspects that could deny the possibility that MR 166 could have been copied in Dalmatia,26 and a more appropriate comparative research of the decoration regarding all the missals from southern Italy dating back to the 11 th and 12 th centuries that could properly clarify the place in which the manuscript was created,27 I can only state that the area between Benevento and northern Apulia played an important role in the network of relations between southern Italy and Dalmatia. 28is is probably due not only to the close links Benevento had with Siponto, whose port, located at the foot of the Gargano mountains, was the chief point of departure for those who were travelling to the eastern coast of the Adriatic sea from the Apulian hinterland, but also to the close ties the Benedictines of northern Apulia and those of the abbey of Santa Maria delle Tremiti in particular, maintained with the populations of the eastern coast during the Middle Ages. 29In addition, both the Abbey of Montecassino 30 and the Church of Rome were very interested in this area, as shown by the appointment of  .13. Bodleian Library, Oxford.Canon.lat.61, c. 64v  Fig. 12. Bodleian Library, Oxford.Canon.lat.61, c. 45v  Gerardo, Bishop of Siponto, as the papal legate to Dalmatia in the crucial years after the schism of Michele Cerulario. 31eturning now to the manuscript, it is possible at this point to advance the following hypothesis: just as it may have happened with Gerardo, who brought with him gifts and liturgical manuscripts when he came to Dalmatia in 1074,32 another distinguished clergyman did the same, and brought with him our Missal.He may have joined a papal delegation that left from Benevento or from one of the monasteries of northern Apulia.The sizes of the manuscript, unusual and different from those of the other Missals,33 and the quality of the decorative set, embellished by gold, in fact show that this manuscript was meant to be used as a "travel book". 34he pocket-size characteristic of the Missal and the fact that it was used by the priest in representative contextstherefore characterized by a certain "solemnity" -could help to further understand the necessity of adding, together with the masses for the bishop or for the pope, another one for the emperor.In contrast to Novak's statement, this characteristic is not at all anachronistic for a southern Italian manuscript, dating back to the 11 th and 12 th centuries, as shown by the presence of the same mass both in the formulary of Casin.127 and in Vat.Lat.6082.In conclusion, I would like to draw attention to an element of this codex that has been overlooked: the texts of two votive Masses -the one Pro Christiani qui ad sepulcrum Domini perrexerunt (p.242-244) and another one In Sancti Angeli (p.347), celebrating St. Michael35 were, as far as I know, unusual in the formularies of the other missals of Montecassino and Benevento. 36 This may be the clue for the identification of its scriptorium or, at least, for the destination of the manuscript.The presence of both Masses-text can be perfectly associated with the context of Fig. 14.Metropolitana Library, Zagreb.MR 166, p. 148   Fig. 15.Biblioteca Capitolare, Benevento.Ms. 26, iniziali S e O, c. 89r  Izme|u Ju`ne Italije i Dalmacije: Misal MR 166 iz Metropolitanske kwi`nice u Zagrebu
-15) can be seen in the missals of the Biblioteca Capitolare of Benevento, cod.19, cod.20 and cod.29 -all probably created in the local scriptorium of Santa Sofia in the 12 th century 25 -or the Missal Ms. Egerton 3511 (ex Benev.29).