Variations on Some Common Topics in Medieval Latin Letters: The Case of the Salzburg Formulae Collection (Late Ninth Century)

The Latin manuscript 4650 of the Bavarian State Library is a collection of templates for charters and letters (so-called formulae) most probably written in Salzburg in the late ninth or early tenth century. Some of these formulae have only been transmitted in that manuscript, but most have been transmitted elsewhere as collections and were probably composed a few generations earlier. It is therefore obvious that this manuscript is a patchwork (or the copy of such a heterogeneous collection), but the heterogeneity of the sources is not apparent at first sight. Only a close analysis of the sequence of the formulae, the use of red ink, and small textual changes permits an appreciation of the technique of medieval scribes when adapting previous models for the creation of new collections more suitable to their own needs. This article deals with the ways in which medieval scribes used letters or models for letters to create new templates for their own use. During the early Middle Ages (c. 500–1050), models of this kind were often copied as collections containing templates for charters as well as models for writing letters, both of which were called formulae. This study is devoted to a collection made in Carolingian times and preserved in the manuscript Munich, Bayerische Staatsbibliothek, Clm 4650 (or Clm 4650 for short). This rather small codex (135 × 107 mm) was written around the end of the ninth or the beginning of the tenth century. The quality of || 1 On collections of formulae, see Brown 2009; Rio 2009. On medieval letters, see Constable 1976; Perelman 1991; Ysebaert 2015. On formularies as a mixture of letters and charters, see Depreux (forthcoming). 2 A description can be found in Glauche 1994, 283–284; Rio 2009, 247–248. For more information on the codex, see the book on East Frankish manuscripts containing collections of formulae to be published by Till Hennings and myself. 418 | Philippe Depreux the parchment is average: the volume is not a deluxe edition, but a booklet intended for everyday use. Most of the codices containing formulae are miscellaneous manuscripts. Only a few of them just contain templates for charters and letters. Clm 4650 is one of these; a medieval scribe who lived later, probably in the eleventh century, wrote at the top of the first folio that the codex was a ‘handbook for various matters’ (fol. 1: liber breviarius uniuscuiusque rei), and a late medieval scribe wrote on the verso of the cover sheet that it was a ‘formulary for letters’ (formularius epistolarum) – in noting this, he was only focusing on one specific kind of text copied in the codex, though. Because of a mistake made by a bookbinder, the manuscript is not preserved in its original form, but the right order can easily be restored. The end of the codex has been lost, however. During the Middle Ages, this manuscript was kept at Benediktbeuern Abbey from at least the eleventh century onwards, but it was probably written in or near Salzburg. The close connection to the archbishopric church of Salzburg is attested in some of the charter models by the mention of saints who were particularly revered in Salzburg. Since formulae are generally anonymised documents, such information is excellent evidence of the collection’s place of composition. For this reason, the editors of the collection in the nineteenth century called it the ‘Salzburg Formulary’ (Salzburgisches Formelbuch – meaning the whole codex – or Formulae Salzburgenses [abbreviated as Form. Salzb. hereinafter] when referring to the formulae transmitted at the end of the manuscript). || 3 By way of example, the size of fol. 38 and fol. 65 is smaller than the others because the scribe used waste leaf. 4 Rozière 1859, 11; Bischoff 1980, 201–202. See the table in the present article indicating the right order of the quires and the precise description of their content. 5 Bischoff 1980, 201–202; Bierbrauer 1990, 78–79 (Kat. 144). 6 Rozière 1859, 13; Schröder 1892, 165–166. 7 Rockinger 1858, 45 (concerning the whole manuscript). 8 Zeumer 1886, 438 (specifically relating to the models only preserved in that manuscript). Variations on Some Common Topics in Medieval Latin Letters | 419 1 The homogeneity of a heterogeneous manuscript Clm 4650 contains formulae of diverse origin copied without any indication of their heterogeneous provenance; the layout is very uniform, in fact. It is unclear whether the scribe made a selection himself or if he copied a ready-made collection that was at his disposal. Consequently, it is hard to say if he was aware he was copying texts that belonged to various collections. Indeed, it is unclear whether these different collections ever existed in the form in which they were published in the late nineteenth century. Traditionally, Clm 4650 has been described – and perceived – as a series of three collections, but this is actually far from certain. Historians were sure for a long time that these formulae belonged to different collections, but we now know that these ‘certitudes’ need to be treated with great caution, as Karl Zeumer’s edition is an ‘editorial fiction’ in many cases. Along with the Formulae Salzburgenses and a small collection of letters written by Alcuin framing the end of the codex, we find texts edited by Zeumer as parts of collections arbitrarily called Formulae Salicae Lindenbrogianae, Additamenta to the Formulae Salicae Lindenbrogianae and Formulae Marculfinae aevi Karolini. None of these collections have been copied en bloc: Clm 4650 is a patchwork – albeit a nicely arranged one (or a copy of one) since no transition from one quire to another coincides with a rupture within a text (or – in the case of the first and second quire – within a coherent group of texts). The following overview should make this clear. || 9 Sonnlechner 2007, 215: ‘Ebenso sticht die Regelmäßigkeit des Buchblocks ins Auge, wie die gesamte Handschrift auch generell einen homogenen Eindruck macht und eine einheitliche Konzeption erkennen läßt’. 10 Schröder 1892, 165: ‘Nach ihrem Inhalte zerfällt diese Sammlung in drei Teile, von denen die beiden ersten auch als selbständige Sammlungen vorkommen [...]. Der Salzburger Kompilator hat [...] diese im Lande [i.e. Bavaria] schon bekannte Sammlung nur durch Hinzufügung des die eigentlichen Salzburger Formeln enthaltenden dritten Teils erweitert’; Sonnlechner 2007, 214: ‘Dieser Codex enthält drei Sammlungen, zuallererst die sogenannten Formulae Salicae Lindenbrogianae, dann die sogenannten Formulae Marculfinae aevi Karolini und schließlich die Formulae Salzburgenses’. 11 Brown 2013, 129 (referring to Rio 2009). 12 See Rio 2009, 101–110 on these collections. 420 | Philippe Depreux Table 1: Formulae collections in Clm 4650 Quire Folios Text Edition 1 1–7 Form. Sal. Lindenbrog., nos 1–7 Zeumer 1886, 266–271 2 8–15 Form. Sal. Lindenbrog., nos 8–14 Zeumer 1886, 271–277 15 Form. Sal. Lindenbrog., Additamenta, no. 1 (beginning) Zeumer 1886, 282 3 16–18 Form. Sal. Lindenbrog., Additamenta, nos 1 (end)–3 Zeumer 1886, 282–283 18–23 Form. Sal. Lindenbrog., nos 15–20 (beginning) Zeumer 1886, 277–281 5 32–33 Form. Sal. Lindenbrog., nos 20 (end)–21 Zeumer 1886, 281–282 33–39 Form. Marculfinae aevi Karol., nos 1–12 (beginning) Zeumer 1886, 115–119 4 24–25 Form. Marculfinae aevi Karol., nos 12 (end)–14 Zeumer 1886, 119–120 25–28 Form. Marculfinae aevi Karol., nos 17–21 Zeumer 1886, 120–122 28–29 Form. Sal. Lindenbrog., Additamenta, no. 4 Zeumer 1886, 283–284 29–31 Form. Marculfinae aevi Karol., nos 22–25 (beginning) Zeumer 1886, 122–124 8 56–63 Form. Marculfinae aevi Karol., nos 25 (end)– 31 [finit] Zeumer 1886, 124–127 63 Form. Salzb., nos 1–2 (beginning) Zeumer 1886, 439–440 9 64–66 Form. Salzb., nos 2 (end)–6 Zeumer 1886, 440–441 66–68 Moral and spiritual considerations Rockinger 1858, 133–134 68–71 Form. Salzb., nos 7–16 (beginning) Zeumer 1886, 441–444 6 40–41 Form. Salzb., nos 16 (end)–20 (beginning) Zeumer 1886, 444–445 41–42 Moral and spiritual considerations Rockinger 1858, 141–142 42–47 Form. Salzb., nos 20 (end)–39 (beginning) Zeumer 1886, 445–448 || 13 Jam quondam fidelis mentem .... in anima coram Deo proficiendo solet esse. There is no physical border between Form. Salzb. 6 and these moral and spiritual considerations (only the first letter of the latter is marked in red). This text was also edited by Rozière (1859, 38–39). Karl Zeumer did not edit it, as he thought that these spiritual considerations had nothing to do with formulae (Zeumer 1886, 441, footnote 6d: ‘quae in c. sequuntur omisi, cum ad formulam pertinere non viderentur’). 14 Erat quidam iudex in civitate .... ut non desinat esse quod antea fuit. There is no physical separation between Form. Salzb. 20 and these moral and spiritual considerations either (only the first letter of the latter is marked in red). This text was also edited by Rozière (1859, 44–45). Karl Zeumer chose not to edit it for a similar reason to the one just mentioned (Zeumer 1886, 445, n. 20c). Variations on Some Common Topics in Medieval Latin Letters | 421 Quire Folios Text Edition 7 48 Form. Salzb., nos 39 (end)–43 Zeumer 1886, 448–449 48–49 Questions and answers on God and creation Rockinger 1858, 151–152 49–55 Form. Salzb., nos 44–60 (beginning) Zeumer 1886, 449–453 11 80–84 Form. Salzb., nos 60 (end)–66 Zeumer 1886, 453–455 84–85 Alcuin, letter no. 294 (to an English pupil) (beginning) Dümmler 1895, 451–452 10 72 Alcuin, letter no. 294 (end) Dümmler 1895, 452 72–74 Alcuin, letter no. 107 (to Arn of Salzburg) Dümmler 1895, 153–154 74–75 Alcuin, letter no. 167 (to Arn of Salzburg) Dümmler 1895, 275 75–76 Alcuin, letter no. 146 (to Arn of Salzburg) Dümmler 1895, 235–236 76–77 Alcuin, letter no. 165 (to Arn of Salzburg) Dümmler 1895, 267–268 77–78 Alcuin, letter no. 150 (to Arn of Salzburg) Dümmler 1895, 245–246 78–79 Alcuin, letter no. 153 (possibly to Arn of Salzburg) Dümmler 1895, 248 79 Alcuin, letter no. 173 (to Arn of Salzburg) (beginning) Dümmler 1895, 286 79 Alcuin, letter no