Notes on the Cassiodorus Tabulation


The total number of 37 stemmata from the Institutiones in the tabulation was first noted by Michael Gorman.[*]Gorman, Michael: The diagrams in the oldest manuscript of Cassiodor's Institutiones. In: Revue bénédictine 110 (2000), 27-41.. The tabulation by Professor Giulia Orofino.[*]Orofino, Giulia. Da Montecassino a Nonantola: La tradizione illustrativa delle Institutiones di Cassiodoro, in Il monachesimo italiano dall?età longobarda all?età ottoniana (secc. VIII-X), Convegno, Nonantola, 9-13 September 2003. <>. embraces only the 34 stemmata in Book II plus one additional motif which is not a stemma, but rather an introductory drawing to stemma No. 27. (This drawing is found only in the Bamberg manuscript and depicts two kids, one page before the leopard stemma).

The layout of the stemmata has been flipped in the X axis and turned 90 degrees to the left for this HTML transcription. In other words, the manuscripts lay out the stemmata with top roots, downward divisions and successive generations left to right. Here they are laid out with left roots, rightward division and successive generations top to bottom for improved legibility. This transposition has been strict: if the St Gall manuscript shows a childless branch to the left of a subdivided branch, then it will appear here above that branch. The source code can easily be converted into nested-list presentation.

A majority of the stemmata comprise only two generations. Number 4, which comprises seven generations, is the deepest of the collection.

Numbers 18, 19 and 20 are all sub-stemmata of number 17. This suggests that layout practices by late antiquity allowed stemmata to be distributed over several pages for greater clarity. Easy disassembly is one of the benefits of stemma notation. In the "Group III" manuscripts (described below), stemma No. 7 is extensively disassembled into separate sections.

There can be little doubt that many of the stemmata have been abbreviated in the course of copying. The lack of any third-generation divisions in No. 3 can only be a mistake. It is a sure sign of division in the Institutiones when a tally of components is given and followed in our manuscripts by the words, id est. In all three stemmata describing the composition of the Christian bible (Nos. 1, 2, 3), one supposes that Cassiodorus, true to his teaching method, must have originally drawn separate nodes for each of the four Gospels for his students, but that a hasty scribe has saved time by simply writing "4 Gospels, i.e. Matthew, Mark, Luke, John".

In the transcript I have punctuated the most likely further ramifications in these three stemmata with semi-colons and parentheses. No. 2 (Augustine's overview of the bible) must have formerly contained a fourth generation differentiating among the 22 historical books (Historia libri XXII). That now-lost generation would surely have contained nodes for "Moysi libri V" (the Pentateuch), "Iesu Nave liber I" (the single book of Joshua) and so on. The 22 prophetic books in No. 2 would have been similarly differentiated. Indeed, one suspects there must have been a fifth generation of nodes in No. 2, further differentiating the component elements within, for example, the category of the four "Prophetas maiores" (major prophets) or the twelve minor prophets.

Close examination of No. 25 suggests that it too may have lost a further differentiation into three parts in the node a comparatione ("argument by comparison"). The parts of this term, which are now merged into the text, must originally have been presented as further sub-branches: a maiore ad minus ("bigger with smaller"); a minore ad maius ("smaller with bigger"); a pari ad parem ("like with like"). It is also arguable that the third branch of Number 8 may have originally been two sub-branches from the iuncta (complex) node: ex pluribus quaestionibus and ex aliqua comparatione.

Comparison of the manuscripts suggests transmission errors were common when copying the layout. No. 29 for example is shown in St Gall as having only two generations: a root and five terms. Based perhaps on the later corrected manuscripts, Mynors ignores this and converts it to a stemma of three generations.[*] Cassiodori Senatoris Institutiones, Mynors, Roger A.B. ed, Oxford. Clarendon, 1937, repr. 1961.. The susceptibility to corruption of stemmata during reproduction by hand was only effectively cured with the introduction of printing.

Lacking the source manuscript, we have of course no conclusive way of proving that it was the copyist who suppressed these nodes. One must, in fairness, entertain the possibility that it might have even been the elderly Cassiodorus who pruned the stemmata and compressed his own classroom presentation in his haste to complete the textbook before death overtook him.

At the end, No. 30 has been shown redrafted according to the recommendation published in 2000 by Michael Gorman.[*]Gorman, Michael: The diagrams in the oldest manuscript of Cassiodor's Institutiones. In: Revue bénédictine 110 (2000), 27-41., who suggests that the version going back to the Bamberg manuscript is logically incorrect. His argument is based on Isidore's surviving abstract from the lost archetype.

The six-generation "mystery" stemma employs text from Quintilian's Institutio Oratoria liber III and deals with jurisprudential and businesslike rhetoric. It includes only one branch of Cassiodorus's stemma No. 7, but adds a further generation to it. This stemma appears in the Group III manuscripts, or "second interpolated form", of the Institutiones. That in the manuscript at St Gall code-named θ by Mynors (online) is muddled and words like δικαιολογικο seem to be erroneously copied. That in the manuscript at Valenciennes which Mynors codes ι (online: frame 11) is better executed. It is peculiar that two of the "children" in the stemma descend from relatio and comparatio without any division. The "second interpolated" manuscripts are a kind of college textbook with readings not just from Cassiodorus, but also from other authors.[*]Mynors xxxix suggests they derive from an archetype compiled some time in the 8th century.. This new work also includes other stemmata dealing with grammar, which have not, to my knowledge, been comprehensively studied.[*]Transcribed at Mynors xxxvi. Mynors wrongly describes these three stemmata as "lists.".

The redactor places the Qualitas Generalis stemma about a dozen pages into Book 2 proper, directly after a passage which is now credited to Quintilian and which ends with the prayer, "Incipit in nomine patris et filii et spiritus sancti." Quintilian is clearly also the source of the first and second generations of text in this "mystery" stemma.[*]Iuridicialis est qui Graece dicitur dikaiologikos. 3.6.33 Negotialem, quam πραγματικα vocat (Hermagoras), in qua de rebus ipsis quaeritur, remoto personarum complexu, ut "sitne liber qui est in adsertione", "an divitiae superbiam pariant", "an iustum quid, an bonum sit"; iuridicialem, in qua fere eadem, sed certis destinatisque personis quaerantur. 3.6.57-58. . I have not yet been able to establish the origin of the sentence of text found in the last generation, or even to transcribe it with certainty. This stemma serves as introductory matter to the Rhetoric chapter, and its placement there leads to some modifications in the subsequent Cassiodorean passage, which has been re-edited to offer stemma No. 4 in snippets only, as noted above.[*]Mynors xxxvii offers only a cursory discussion of this stemma, describing it as "the table of the parts of 'qualitas generalis' ... with a few further explanations ... Some further definitions are added in the margin, among them 'concessio', 'imprudentia', 'scriptum et voluntas' (from Quint. VII 6, 2)..." Mynors excludes the passage altogether from his edition.

There are three possible explanations for the presence of these additional stemmata dealing with the Qualitas Generalis and grammar. One is that an 8th-century redactor devised these stemmata, using Cassiodorus as his model and Quintilian's text for example as a source. A second is that the adaptation of the grammar text and Quintilian to stemma form may have been the work of Cassiodorus himself, written up in the 6th century in a now lost early draft of the Institutiones. The third and most tantalizing possibility is that the redactor might have had access to, and copied material from, other diagrammatic texts dealing with grammar and rhetoric which were older still than the work of Cassiodorus. If 4th- or 5th-century books other than the Cassiodorean Institutiones could be demonstrated to have employed stemmata, this would of course be of huge importance to the history of stemma notation and diagrams.

Back to History Table of Contents

Back to Macro-Typography on the Web