Ducks and Diagrams: Insights Into the Medieval Readership of Priscian’s Institutiones Grammaticae Preserved in Leiden University Libraries, BPL 91 and BPL 186

Grammar was the foundational art of the medieval liberal arts curriculum. In his retrospective summary of his school career written in 1159, John of Salisbury famously surveyed the methods used for teaching grammar employed by his masters, and critiqued his contemporaries who ‘profess all the arts, liberal and mechanical, but who are ignorant of this first one [i.e. grammar], without which it is futile to attempt to go on to the others’. As John goes on to note in the same passage, playing with his words, ‘while other studies may also contribute to “letters” [litteratura], grammar alone has the unique privilege of making one “lettered” [litteratum]’, that is learned.1 One of the most influential grammar manuals studied in the medieval schools was the Institutiones grammaticae, a work composed by Priscian of Caesarea in Constantinople (now Istanbul, Turkey) in the early sixth century.2 No fewer than seventeen manuscript copies of this work are preserved in Leiden University Libraries.3 In this study, I will assess annotations to two twelfth-century copies of the Institutiones (Leiden University Libraries [henceforth UBL], BPL 91, and BPL 186). In line with the theme of ‘centuries of TeXT’, I will explore how the preservation of such medieval manuscripts brings to the fore sometimes ambiguous, sometimes innovative, interactions with texts, and can indicate how medieval readers engaged with this important grammatical work.

The Institutiones Grammaticae

Priscian’s Institutiones was complex and long; it was composed of eighteen separate books, with books I-II dealing predominantly with phonetics, the sound of speech, books III-XVI with the parts of speech (the noun, verb, adjective, etc.), and books XVII-XVIII with syntax, the structure of sentences. Early copies of the Institutiones tended to include all these books, but later copies often contained only the first sixteen books, dealing with phonetics and morphology—the so-called Priscianus Major, or ‘Greater Priscian’. Meanwhile, the Priscianus Minor, or ‘Minor Priscian’, comprising books XVII-XVIII and concerning syntax (the study of which was regarded as more specialised) often circulated separately, particularly from the thirteenth century onwards.4 The Institutiones was written for advanced students of Latin in the Greek-speaking part of the Roman empire and was organised, as Vivien Law described, in a way that is ‘for the most part implicit rather than explicit’, with Robert Black even referring to the ‘confusion, disorder, prolixity and incomprehensibility’ of the text.5

Although it was composed over five centuries prior, Priscian’s work was still extensively read in the twelfth century and beyond.6 In fact, in spite of its complexity and length, this introductory work to the grammatical and syntactical forms of the Latin language survives in over 500 manuscript copies dating from between the late eighth to the fifteenth century, and no less than ten early printed editions prior to 1500.7 Owning a copy of Priscian was standard for medieval schools, and such manuscripts were valuable possessions. Law recorded that one copy of the Institutiones (now Barcelona, Archivo de la Corona de Aragón, MS Ripoll 59) is said to have cost the cathedral chapter of Barcelona a house and a piece of land in 1044.8 Writing to Bonibertus of Fünfkirchen in mid-1023 on the occasion of his establishment of a new bishopric in Pecs in Hungary, Fulbert, bishop of Chartres, sent ‘one of our copies of Priscian’, on Bonibertus’s request.9 It is a mark of the perceived quality of the work that it was seen as a suitable gift for a newly established ecclesiastical centre.

Annotations and additions

Aside from its systematic treatment of the study of grammar, Priscian’s work also attained its foundational status on account of the examples he used throughout to illustrate various points and details of the art; this marked his work apart from that of other grammarians, such as Donatus.10 Drawn from Latin and Greek classical texts, these examples were objects of study in their own right and contributed to the popularity of the work. We can gain insights into the varied emphases and points of interest of readers of the Institutiones Grammaticae by looking at the glosses, or annotations, made to Priscian’s work; many of these were copied from manuscript to manuscript, but are individualised and innovative reactions to the text.11 Such glosses can reveal a layered evidence of interest in the text, and often paid particular attention to his selection of examples.

Some additions to manuscript copies of the Institutiones are, however, harder to interpret. UBL BPL 186 is a composite volume comprising two grammatical manuscripts copied in different contexts and periods, but now bound together.12 In the first part of the manuscript (fols. 1 - 86), we find a twelfth-century Italian copy of Priscian’s Institutiones Grammaticae, Books XVII-XVIII (i.e., the Minor Priscian).13 The manuscript contains a number of (marginal and interlinear) glosses, added by various hands, demonstrating an accretion of reader interactions with the text, as well as further non-textual interventions, such as small crosses and nota signs placed beside passages of relevance, and appended paragraph markers which serve to divide up the text.14 The manuscript also preserves a number of additions which seem to veer on the nonsensical. For example, it contains a number of penwork drawings of flowers (with the drypoint tracings that inspired them still visible) along with various sketches of decorated initials in the margin.15 At a number of points, brief exclamations, such as ‘may we rejoice!’ (‘gaudeamus’), have been added, perhaps functioning as pen trials.16

Figure 1, UBL, BPL 186, fol. 47v: In addition to the alphabets in the lower margin, attempts to trace decorated initials can be observed towards the top of the page.

On fol. 47v [Figure 1], an annotator has practised writing the letters of the alphabet. Struck apparently by the similarity between the letter ‘g’ and a sketch of a bird they have drawn a line of patient-looking ducks in the lower margin. It is difficult to interpret, and even harder to date such annotations, although they attest to continued use of the manuscript by readers, working with a pen in hand. One drawing, however, seems to imply an interest in Priscian’s classical examples. On fol. 71v [Figure 2], we find a marginal sketch in red ink of a man holding what appears to be a knife and threatening a woman in a supplicatory pose. This sketch corresponds with an example found in the text; Priscian refers to a scene from the Roman comedian Terence’s Adelphoe where a character exclaims ‘Won’t you keep your hand off me, you scoundrel?’ [Ad. 780: ‘Non manum abstines, mastigia?].17 In this case, at least, the sketch brings the text to life, and with it the work of a reader who wished to draw attention to and perhaps dramatise this passage.

Figure 2: UBL, BPL 186, fols. 71v-72r. The sketch corresponding to the Terence quotation is found in the margin of fol. 71v, while the words ‘ave maria. gratia plena. dominus tecum’ are written in the margin of fol. 72r (a possible pen trial).

A Text Prepared for Reading

As the preceding section indicated, annotations and additions to copies of Priscian’s Institutiones Grammaticae present interpretative challenges for any reconstruction of medieval readers’ interactions with this important grammatical work. In the remainder of this study, I will investigate evidence for readership preserved in another twelfth-century Italian copy of the Institutiones, UBL BPL 91.18 This manuscript is a copy of Books I-XVI of the Institutiones, i.e. a copy of the ‘Major Priscian’. It contains a number of annotations in different hands, dating from different periods; these are particularly intense across Book I-II of the text which concerns the units of speech (i.e. letters and syllables), nouns, and patronymics.19 In fact, Priscian’s treatment of the units of speech is presented with curious coherence in this manuscript, with the start of Book II (which opens with a discussion of the syllable) following immediately on from the end of Book I (fol. 12v), without any break in the text.20 This scribal choice gives that part of the text a unitary quality which is absent in Priscian’s original arrangement. It also gives enhanced prominence to Priscian’s discussion of the noun. Rather than following the prescribed textual-structural break, it is instead the opening of that discussion (fol. 17r), found partway through Book II, that is accorded particular attention [Figure 3]. It is decorated with a complex initial ‘N’ for nomen [noun] and a rubric in alternating blue and red majuscules. It is also separated from the preceding section by a half-blank folio.21

Figure 3: UBL, BPL 91, fols. 16v-17r: The opening of the section concerning the noun (nomen).

This distribution of text and decoration sets the scene for the subsequent detailed discussion of the noun, which continues up to the conclusion of Book VII, and is followed by Priscian’s treatment of the verb. The discussion of the noun and that of the verb are separated in this manuscript by a further half-blank folio (fol. 85r); it is clear that the scribe of this manuscript intended its reader(s) to pay particular attention to the overarching themes of the Institutiones, in addition to its standard book divisions. However, this space was filled in (also in the twelfth century, based on the proto-gothic script used) with a diagram in branching and radial form concerning the noun and its types [Figure 4]. This diagram has been briefly treated by Vivien Law, who described it as an instance of commentators on Priscian making ‘visible the implicit structure of the [Priscian’s] doctrine’, but it has yet to receive a detailed description and analysis, particularly one which considers its material features and potential contexts of use.22

Figure 4: UBL, BPL 91, fol. 85r: Diagram at end of Book VII. The differences in ink colour and hand between different parts of the diagram are clearly visible.

The diagram as presented in BPL 91 was seemingly executed in two stages and by two separate hands (as apparent from the colours of ink and slight variants in the execution of the script). Hand 1 (possibly identifiable with the scribe of the main text, albeit writing with a finer nib) was responsible for the initial part of the diagram, which is presented in branching form. This distinguishes the two species of noun [‘nomen’], primitive [‘primitivum’] and derivative [‘derivativum’]: that is, nouns which do not originate in other words, and nouns that are derived from other words. It then notes that nouns within each of these species can be either proper [‘proprium’] or appellative [‘appellativum’]. A proper noun is one that refers to a specific person, place or thing, whereas an appellative noun (or common noun) is a generic name for a class of people, places or things.23 Hand 2 (who writes in a darker ink, and with more rounded letter forms), is responsible for the radially arranged section, which identifies various subtypes of nouns. This appended section of the diagram is executed with substantial care; the blind-ruled lines drawn to accommodate the text of the subtypes are still visible (extending out beyond the scheme to the edge of the ruled text block) and further lines have been ruled in ink from each subtype to connect with the existing nodes of text of the branching diagram. These inked lines have been traced without smudging (suggesting that time was taken to allow the ink to dry before ruling the next line) and text is oriented towards the heart of the diagram, with some terms written at an angle or even upside-down (suggesting that this scribe rotated the page as they worked).

Reading the radially constructed section of the diagram in a clockwise direction, it opens with the nine subtypes described by Priscian as specific to the category of derivative common nouns: the adverbial, participial, verbal, denominative, diminutive, superlative, comparative, possessive, and patronymic nouns.24 These are followed by twenty-seven further subtypes of derivative common nouns (such as those that signify place, time, order, number, etc.). We then find four subtypes which are shared by both proper and common derivative nouns (synonyms, homonyms, incorporeal, and corporeal nouns), before the diagram presents the four subtypes of proper derivative nouns (surname, family name, name, and forename). The radial subdivisions then treat the subtypes of primitive common nouns (identical to the list of twenty-seven subtypes already enumerated for common derivative nouns), followed by the subtypes shared by proper and common primitive nouns (again, identical to those previously identified for derivative nouns) and, finally, the four subtypes of proper primitive nouns (surname, family name, etc.). The diagram is, in effect, a visual summary of Priscian’s introductory discussion on the subtypes of the noun.25

Reading the Diagram

While Law associated this diagram with the burgeoning interest in speculative, or theoretical, grammar typical of the mid-twelfth century onwards, the oldest copy of it that I have located thus far—in Munich, BSB Clm 14272, an early eleventh-century collection of texts on the liberal arts owned by the monk Hartwic of St Emmeran—in fact pre-dates the dominance of this trend. Hartwic’s collection contains an excerpt from Priscian’s discussion of syntax (Institutiones Book XVII, 1-71), with the diagram presented on a full page, adjacent to the extract’s opening, serving, perhaps, as a reminder to the reader of Priscian’s account of the noun.26 Two other twelfth-century Priscian manuscripts also contain versions of the diagram; in these it is found towards the manuscripts’ conclusion, nestled among other grammatical notes, suggesting that it played the role of an appendix.27 The fact that the diagram is present in multiple manuscripts means that the scribes of BPL 91 were not novel in compiling its content, but instead copied it from an existing (as yet unidentified) exemplar. However, the variety of contexts in which the diagram is found within these four manuscripts implies that scribes could exercise some agency governing its placement and in so doing influence its impact on the reader. For example, by integrating the diagram into the body of the text of BPL 91, the diagram assumes a navigational and emphatic function for the reader that is absent in the other examples. Placed at the conclusion of the section on the noun, it not only provides a skeleton outline of the preceding content, but also emphasises the separation between Priscian’s discussions of the noun and verb.

The diagram offered its potential reader several functions, including and going beyond the primary one identified by Law (‘making the invisible visible’, that is making the implicit explicit). Indeed, it clarifies the wording of Priscian’s text. For example, in his text Priscian refers to the twenty-seven subtypes of common nouns, and only at the conclusion of his list adds that these are subtypes of both primitive and derivative common nouns. In repeating this list of twenty-seven subtypes twice, the diagram makes Priscian’s words abundantly clear. Drawing on texts such as Priscian’s Partitiones (an analysis of the first lines of Virgil’s Aeneid), parsing grammars—where a reader was asked to asked to analyse individual words (and thereby parts of speech)—were popular in the Middle Ages; a diagram like this could have been a handy heuristic for illustrating and recalling the types of categorisations which parsing required.28

All four examples of the diagram, a convenient overview of the subtypes of noun, would have satisfied a reader’s need for a memory aid. On account of its unusual radiating form, any reader engaging with this diagram would have also been inclined towards a physical interaction with its host manuscript, angling their head to read the supplied text. This could have encouraged processes of ‘embodied cognition’, potentially enhancing the understanding of abstract concepts or memory performance.29 The presentation of the diagram in BPL 91 may have amplified such interactions, however. In all other examples of the diagram discussed, the radial section is always oriented towards the page’s long edge, meaning that this portion of text is also oriented to the reader. However, on account of the limits of the space it occupies, the diagram in BPL 91 is vertically rather than horizontally oriented, with the radial section of text angled away from the reader, its words sometimes even inverted. Thus, the physical interactions required to read the diagram in BPL 91 (and with that the processes of embodied cognition involved) were potentially intensified, with the reader required not only to angle their head, but perhaps even rotate the book.

Finally, the presence of the diagram across multiple manuscripts increases the likelihood that it served as a teaching aid. It is striking that one of the manuscripts containing the diagram (Reims, Bibliothèque municipale, MS 1096) also contains a number of references to the Priscian commentary, Summa super Priscianum, written in the 1140s by Petrus Helias—a teacher of John of Salisbury—suggesting that it comes from a school context.30 The diagram could even have been a point of departure for more complex classroom discussions. In BPL 91, a thickening and overlapping of the web of lines within the diagram draws attention to the subtypes shared by common and proper nouns, both primitive and derivative (corporeal things, incorporeal things, synonyms and homonyms, as discussed in Book II, 59, 9-19). For Priscian, a noun signified ‘substance and quality’, or as Karin Margareta Fredborg put it, ‘nouns signify substance (a what) and quality (an of what kind)’.31 Grammarians from the twelfth century on would debate intensely what precisely it meant for a noun ‘to signify’. Ambiguities, such as cases where a noun designates more than one thing (through homonymy) or where multiple nouns designate the one thing (synonymy), were fertile material for such discussions.

Conclusion

This study has primarily treated annotations to Priscian’s work dating from the twelfth century, a period after which his influence would begin to wane. In the thirteenth century, grammarians would seek other means to capture the attention of readers, with Alexander of Villedieu and Evard of Béthune composing popular grammars written in an easy-to-remember verse form. Thus, the additions made to the Institutiones in BPL 186 and BPL 91 capture moments in a period of transition in learning grammar. The presence of the diagram in a number of twelfth-century copies of the Institutiones seemingly illustrates attempts to impose structure and render memorable Priscian’s text in an environment where new texts and pedagogical approaches were competing for recognition, while the occasionally nonsensical additions to BPL 186 may suggest the drifting attentions of readers who struggled with Priscian’s prolixity. Although we cannot capture with certainty the intentions of those that made these additions to BPL 186 and BPL 91, their preservation on the manuscript page illustrates the flexibility of that space to serve as a conduit of interactions with Priscian’s text. These examples also show how readers individuated their copies, rationalising research approaches which address copy-specific interventions—research that is only feasible in environments where the careful preservation and making accessible of the manuscript carrier is prioritised.

Bibliography

Manuscripts:

Leiden University Libraries, BPL 91

Leiden University Libraries, BPL 186

Munich, Bayerische Staatsbibliothek, Clm 14272

Paris, Bibliothèque nationale de France, MS lat. 7544

Reims, Bibliothèque municipale, MS 1096

Sources:

Bauer, Bernhard, and Victoria Krivoshchekova, ‘Definitions, Dialectic and Irish Grammatical Theory in Carolingian Glosses on Priscian: A Case Study Using a Close and Distant Reading Approach’, Language & History, 65 (2022), 85–112

Black, Robert, Humanism and Education in Medieval and Renaissance Italy: Tradition and Innovation in Latin Schools from the Twelfth to the Fifteenth Century (Cambridge: Cambridge University Press, 2001)

Cinato, Frank, Priscién Glosé. L’Ars Grammatica de Priscien Vue à Travers Les Gloses Carolingiennes (Turnhout: Brepols, 2015)

Copeland, Rita, and Ineke Sluiter, eds., ‘“Priscian, Institutiones Grammaticae and Institutio de Nomine Pronomine Verbo, ca. 520”’, in Medieval Grammar and Rhetoric: Language Arts and Literary Theory, AD 300-1475 (Oxford: Oxford University Press, 2012), pp. 167–89

Fredborg, Karin Margareta, ‘Speculative Grammar’, in A History of Twelfth-Century Philosophy, ed. by Peter Dronke (Cambridge: Cambridge University Press, 1988)

Fulbert of Chartres, The Letters and Poems of Fulbert of Chartres, ed. by Frederick Behrends, trans. by Frederick Behrends, Oxford Medieval Texts (Oxford: Clarendon Press, 1976)

Gibson, Margaret, ‘Priscian, “Institutiones Grammaticae”: A Handlist of Manuscripts’, Scriptorium, 26 (1972), 105–24

Gumbert, J. P., Illustrated Inventory of Medieval Manuscripts 2: Leiden, Universiteitsbibliotheek BPL (Hilversum: Verloren, 2009)

Holtz, Louis, ‘L’émergence de l’oeuvre Grammaticale de Priscien et La Chronologie de Sa Diffusion’, in Priscien: Transmission et Refondation de La Grammaire, de l’antiquité Aux Modernes, ed. by Marc Baratin, Bernard Colombat, and Louis Holtz (Turnhout: Brepols, 2009), pp. 37–55

John of Salisbury, The Metalogicon of John of Salisbury, trans. by Daniel D. McGarry (Berkeley and Los Angeles: University of California Press, 1955)

Law, Vivien, The History of Linguistics in Europe: From Plato to 1600 (Cambridge: Cambridge University Press, 2003)

———, Wisdom, Authority and Grammar in the Seventh Century: Decoding Virgilius Maro Grammaticus (Cambridge: Cambridge University Press, 1995)

Luhtala, Anneli, Grammar and Philosophy in Late Antiquity (Amsterdam: John Benjamins, 2005)

Matthews, P. H., What Graeco-Roman Grammar Was About (Oxford: Oxford University Press, 2019)

Passalacqua, Marina, I Codici Di Prisciano (Rome: Edizioni di Storia e Letteratura, 1978)

Priscian, Prisciani Institutionum Grammaticarum Libri I-XII, ed. by Martin Hertz, Grammatici Latini, 2 (Leipzig: Teubner, 1855)

———, Prisciani Institutionum Grammaticarum Libri XII-XVIII, ed. by Martin Hertz, Grammatici Latini, 3 (Leipzig: Teubner, 1859)

Sirrigde, Mary, and Karin Margareta Fredborg, ‘Demonstratio Ad Oculum and Demonstratio Ad Intellectum: Pronouns in Ps.-Jordan and Robert Kilwardby’, in Logic and Language: A Volume in Honour of Sten Ebbesen, ed. by Jakob Leth Fink, Heine Hansen, and Ana María Mora-Márquez (Leiden: Brill, 2013), pp. 199–220

Teeuwen, Mariken, ‘Practices of Appropriation: Writing in the Margins’, in The European Book in the Twelfth Century, ed. by Erik Kwakkel and Rodney Thomson (Cambridge: Cambridge University Press, 2018), pp. 139–56

Terence, The Comedies, trans. by Peter Brown (Oxford: Oxford University Press, 2006)

Williamson, Kellie, and John Sutton, ‘Embodied Remembering’, in The Routledge Handbook of Embodied Cognition, ed. by Lawrence Shapiro (Abingdon, Oxon.: Routledge, 2014), pp. 315–25

Zetzel, James, Critics, Compilers and Commentators: An Introduction to Roman Philology, 200 BCE-800 BCE (Oxford: Oxford University Press, 2018)

Irene O’Daly is a lecturer in Book History at Leiden University. She is a specialist in medieval manuscript studies, with research interests in the fields of classical reception and the history of knowledge.