Jan Herlinger

Biblioteca Nazionale Marciana, MS Latini, Cl. VIII.85: A Preliminary Report

The manuscript Venezia, Biblioteca Nazionale Marciana, Latini, Cl. VIII.85 is a collection principally of music theory treatises copied in and near Mantova in 1463-64.[1] It has served as the source for critical editions of five texts (see Appendix A), three of them of great importance. The Lucidarium of Marchetto of Padua was the seminal Italian treatise on mode of the later Middle Ages and laid the groundwork for the Renaissance theory of modes in plainchant and polyphony;[2] the Ars practica mensurabilis cantus secundum Iohannem de Muris (formerly known as the Libellus cantus mensurabilis) provided the first complete codification of the system of rhythmic notation that was to prevail till the end of the Renaissance. The Liber de proportionibus of Johannes Ciconia is one of very few medieval treatises known to have been written by a major composer.

Though the importance of the “big” treatises in the manuscript is unquestionable, several shorter texts are of considerable interest: six works on counterpoint (items 5a, 5b, 10, 14, and 18) and two on the theory of accidentals (5c and 18). Counterpoint and accidentals converge in item 18, the intriguingly titled Tercius liber musice, which expounds the teoria del grado, a type of counterpoint that considers how the notes of one hexachord harmonize with those of the same or another hexachord.[3] Traditional theory used seven hexachords built on Cs, Gs, and Fs, which involve only the natural notes plus B flats. The Tercius liber provides an additional ten hexachords, coniunctae, built on B flat, D, A, E, B natural, and F sharp; they add to this scale E flats plus sharps on F, C, G, D, and A. Treatises on coniunctae are rare; very few of them go as far in the sharp direction as this one (the treatment of coniunctae in the Berkeley Treatise, by contrast, includes hexachords built on B flat, E flat, D, and A, and thus uses, in addition to flats on B, E, and A, sharps only on F and C).[4] But what really sets the Tercius liber apart from any other coniunctae treatise I know is that it shows how to make counterpoint in those remote hexachords.[5]

Not surprisingly, there are almost twenty works of scholarship, aside from the editions, that deal with Marciana VIII.85; see Appendix B. Yet despite so much attention, little is known about the manuscript itself, primarily because we who studied it each focused on one treatise or another, and considered the whole only as a container for that text. Scholars disagree on the number of scribes who worked on it; the author of one of the texts it transmits has consistently been misidentified; nor has the identity been established of the person in whose house at least a part of the manuscript is known, and has been known, to have been copied since the manuscript was first catalogued by Valentinelli in 1872. During a sabbatical in Venezia I had the chance to examine the manuscript in detail, and I’m now able to propose answers to these questions.

Among aspects of the manuscript ignored in every published description—including, regretfully, that in my own edition of Marchetto’s Lucidarium—is its construction, a matter considered de rigueur in most branches of codicology. The manuscript consists at present of 91 paper leaves gathered in nine fascicles comprising five libelli (as I use the terms here, a fascicle is a collection of bifolios folded together—thus one of the basic physical units of the manuscript—and a libellus one or several fascicles treated as a unit for copying). A catalogue of its contents, showing their distribution among the fascicles and libelli, appears as Appendix C.

The first libellus begins on fol. 1v with the opening chapters of Marchetto’s Lucidarium (item 1), preceded by a caption (in red):

In festo virginis katherine et matris sanctissime principiaui Mantue hunc tractatum. In nouembrio 25 currente die. Anno domini 1464.

On the feast of Catherine the Virgin and the Most Holy Mother I began this treatise at Mantova on the 25th day of November in the year of the Lord 1464.

This is the only date in the first libellus of the manuscript; others appear in the second libellus, at the end of items 2 and 3, dated respectively May and June 1464; and in the fourth libellus, at the end of items 11 and 13, September and October 1463—if Michaelmas was celebrated, as now, on 29 September. Thus of the libelli bearing dates, the fourth was copied first, then the second, and finally the first.

The first libellus of the manuscript is copied on thick paper of high quality, so that there is very little bleedthrough; the writing block is uniformly marked off and ruled; the scribe wrote a careful scholastic book hand; chapter captions appear in red, as do staff lines in musical examples and labels in diagrams; large red initials alternate with empty spaces that were probably meant to have been supplied with initials in blue; names of authorities cited in the text are written in the margins, these marginal annotations underlined in red. Diagrams are drawn with straightedge and compass; the paper still shows pinholes left by the compass point.

Libellus 2 is copied on the same paper as libellus 1 (they have the same watermark, a Greek cross); and it is similarly prepared and executed, though with not quite so much care. It opens with the inscription “Michi Resera”; both Johannes Wolf and Suzanne Clercx-Lejeune took this as the name of the scribe, and, as it recurs five more times in various parts of the manuscript, declared that he had copied it in its entirety. But as Pieter Fischer pointed out in his RISM inventory of Italian manuscripts containing treatises on music theory (1968), “Michi resera” is a prayer formula (“Open unto me”) directed in this case to “B[eatae] I[mmaculatae] V[irgini]”.[6] Fischer also noted two sections of the manuscript that had been copied by other scribes, the last page of libellus 1 (f. 10v, which presents an excerpt from Marchetto’s Lucidarium, item 1b-c),[7] and the final three folios of the manuscript (ff. 89r-91v, items 19, 20: a treatment of the prolations, followed by a diagram of the hexachords, along with miscellaneous notes). But Fischer overlooked the hand of item 6 (f. 61r), which opens libellus 3; as the words on this page have nothing to do with music, he dismissed them (as was RISM’s policy) with the words “Non agunt de musica”—“They don’t deal with music theory.”[8] Ellsworth, accordingly, in the introduction to his edition of Ciconia’s Liber de proportionibus, stated that this scribe “is of little importance for us,” as he copied only one item, “which does not concern music and is not an integral part of the manuscript.”[9] As we’ll see, the text on this page is not without its significance.

Libelli 3 and 4 continue in a hand that Ellsworth called a “scholastic cursive hand but without the clarity and precision of the first scribe.” These libelli certainly lack the refinement of the first two. The paper is thin and poor in quality; bleedthrough often makes it difficult to read the text, even when working with the manuscript itself. The writing block is unruled for the most part, and lines of text tend to stray across the page. Initials are few, smaller than those of libelli 1 and 2, and in the same dark brown ink as the text; the scribe seems to have written them in as he went along. And when there are diagrams they are done freehand.

Libelli 3 and 4 are so different in appearance from libelli 1 and 2 that I worked with the manuscript for two months before reaching the conclusion that, with the exception of eight pages (ff. 10v, 61r, and 89r-91v; Items 1b-c, 6, 19, and 20—those enclosed in shaded fields in the contents list of Appendix C), the entire manuscript had, indeed, been copied by one scribe. I began to suspect this was possible when I noticed that, though the scribe of libelli 3 and 4 seldom wrote a beautiful hand, he was capable of doing so: in the middle of the clumsily written folio 62v, for instance, appear three words (Huc videlicet graves) much more elegantly written than the rest. The appearance of the unusual prayer formula “Michi resera” in both libelli 2 and 4 suggests a single mind at work in both; a number of letter forms that recur throughout the manuscript (e.g., a distinctive capital N with angular corners, a capital I decorated with two dots vertically aligned, a g with the lower loop either left open or jutting down sharply to the left), along with similar marginal flourishes, suggest a single hand as well—as does the use in every libellus of the same sign where words are broken between one line and the next. Usually scribes use a pair of short lines something like an equal sign in such cases; this scribe, idiosyncratically, did not lift his pen between the two strokes, so that the lines are connected, in a manner resembling the numeral 2. This last bit of evidence I regard as particularly telling, as signs like these are something a scribe would write without giving them conscious attention.

If the scribe was capable of writing beautifully, why did he not do so in libelli 3 and 4? Perhaps he lacked the time or the materials to do a proper job, or perhaps even the models that would have provided a worthy example for him to follow. If the scribe’s circumstances had improved during the six months intervening between the copying of libellus 4 and libellus 2, that change might have given him access to better models and materials and the time to use them properly.

Was libellus 3, which lacks any indications of date or place of writing, copied at about the same time as libellus 4? Their general similarity of appearance would suggest that it was. And there is yet one other bit of evidence that ties the contents of libelli 3 and 4 together: a number of the texts in both show clear connections of one sort or another to the city of Mantova and its environs. In libellus 4, item 12, the short text De vi et laudibus musice bears a colophon stating, “Hec ex libro francisci de robertis cuntis mantuani”—“these things from a book of the Mantuan Franciscus de Robertis Cuntis.” Item 13a, the discussion of music from Isidore’s Etymologies, bears the colophon “Hec Ysydorus sanctissimus ubi supra,” where ubi supra may indicate that the scribe copied this text from the same book as the preceding (it could be, on the other hand, that the words refer to the place of copying given in the colophon of item 11). Item 11, Ciconia’s De proportionibus, is transmitted in only two other manuscripts, into one of which it was copied, also in Mantova, just ten years after Marciana VIII.85.[10] Concerning libellus 3, item 7, the Introductiones artis musice of Nicolaus “AurItus” de Buccellanito, Bonnie Blackburn has suggested that the “de Buccellanito” element may refer to the Bizzolano quarter of Canneto sull’Oglio, just west of Mantova.[11] I see no such specifically local connections for the contents of libelli 1 or 2; the wider perspective in the contents of these parts of the manuscript (which begin respectively with what are arguably the two most important music theory treatises of the early fourteenth century) would be consistent with the improvement in the scribe’s circumstances I’ve hypothesized.

The contents of libelli 3 and 4 betray another trait in common: an interest in matters pertaining to classical antiquity, presumably on the part of the person for whom the manuscript was copied. Item 12, De vi et laudibus musice, is a florilegium drawn mainly from Boethius and Macrobius (though also from Isidore, Marchetto, and Johannes de Muris). Item 13b, a note written in the upper right margin of f. 78r that gives the names of the muses and their meaning, is a gloss on the text of item 13a, an excerpt from the Etymologies of Isidore, who here derives musica from musae. (Parenthetically, I’ll note that on this same page the scribe uses a marginal “No[ta] b[e]n[e]” to call the reader’s attention to a particular clause in Isidore’s text:

[…] eratque tam turpe Musicam nescire quam litteras.
… and not to know music theory was as disgraceful as not to know grammar.

Surely this scribe was aware that the person for whom he was copying had an interest in learning and the arts.) Item 15 is a “Guidonian” hand—one example of what is surely the most ubiquitous image in medieval music theory manuscripts—but here with Greek note names replacing the usual Latin ones, and with the names of the fingers given on the palm of the hand in both Latin and Greek. The Greek note names recall the practice of Nicolaus de Buccellanito in item 7, who liked to use Greek names for the notes of the scale—in a manner similar to that of Johannes Ciconia, from whom he seems to have cribbed a good deal (Blackburn, “Buccellanito, Nicolaus de”). Finally, item 16, on the last verso of libellus 4, is a collection of rather elementary notes on various classical subjects—like these:

Laodomia vxor fuit prothegilay qui interfectus fuit primo ab hectore In bello troyano […]
Callimachus et philetus diu fuerunt poete qui elegos conscripserunt […]
Cos co uel coos coo declinatur et est insula in mari […]

Laodamia was the wife of Protesilaus, and he was slain by Hector in the Trojan War.
Callimachus and Philetus were poets a long time ago, and they wrote elegiac verses.
Cos is declined Cos, Co or Coos, Coo, and it is an island in the sea.

This makes five items with classical allusions in just two fascicles of the manuscript.

To them we can now add item 6, which has long been regarded as the letter of a certain “Syrus” to his friend Candidianus (this is the text RISM B III² disregarded as not concerning music theory). Valentinelli, the librarian who first catalogued the manuscript in 1872, called the author “Syrus lugdunensis”—Syrus of Lyon—as if he knew perfectly well who this person had been; others of us who inventoried the manuscript followed his bad example. Considerable searching, however, yielded no “Syrus” likely to have been the author. He writes that he prefers “my fogs of Lyon” to the sunshine of Rome, which he now suffers; he berates his correspondent’s birthplace, Cesena, as “an oven,” and his residence, Ravenna, as ridden with mosquitoes and frogs, a place where the natural order has been so perverted that “walls fall down and waters stand up, towers float and ships are grounded, the sick walk and physicians lie in bed, … clerics practise usury and Arabs sing the psalms.”[12] While this last pair of contraries reflects a Christian background, there is nothing to indicate when the author might have lived. At any rate, the reference to “my fogs of Lyon” makes it clear that Valentinelli need not have known who the author was to characterize him as “lugdunensis”; he could have deduced as much from the letter itself. But even the name “Syrus” was a confection of Valentinelli’s. What actually stands for the writer’s name in the headline are the letters “S” and “y” plus the hook that indicates a -us ending; given the conventions of the time, the writer could be anyone whose name begins with “Sy-“ (or “Si-“) and ends with “-us.”

I am grateful to Leofranc Holford-Strevens of Oxford for identifying the author: Sidonius Apollinaris. Sidonius, a Gaul, married the daughter of Avitus, who became emperor in January 456. Sidonius not only survived the coup that undid his father-in-law eight months later, but delivered the panegyric for his successor, Flavius Maiorianus. He was appointed prefect of the city of Rome in 468 and bishop of Clermont in 469; as such he saw the Auvergne sacrificed in 474, by treaty, to the invading Goths in an attempt to save Italy from them. He died around 480, and is an important witness to the end of the Roman Empire.[13] The placement of a letter of his at the front of the libellus that opens the oldest portion of the manuscript attests yet again to an interest in classical antiquity on the part of the person for whom it was destined.[14]

For whom was the manuscript destined? The colophon of item 11, Ciconia’s De proportionibus, f. 77r, states that this part of the manuscript was “scriptum mantue in domo strennui militis illustrisque uiri domini baltasaris de castello leonis”—“written at Mantova, in the house of the vigorous miles and illustrious gentleman, Lord Balthasar de Castello Leonis)” (I leave miles untranslated, as it can carry a range of meanings); the colophon was recorded by Valentinelli and—again—reproduced by those of us who have inventoried the manuscript, but without determining who this person might have been. Actually, it is not difficult to render the Latin name Balthasar de Castello Leonis into Italian: Baldassarre Castiglione. The great humanist, author of Il libro del cortegiano, was born in Mantova, but only in 1478—fifteen years, inconveniently, after this portion of the manuscript had been copied.

Two exceedingly rare histories of the Castiglione family, however, both held by the Marciana,[15] report that the Baldassarre Castiglione known to all students of Renaissance culture was the namesake of his grandfather, “Baldassarre Primo”; and they call Baldassarre Primo a miles, the same honorific used for him in the colophon of item 11. It was he in whose house at least libellus 4 was copied.

This Baldassarre Primo (1414-77) was the son of Cristoforo (d. 1425), an eminent jurist of Milano, who had studied law at Parma and later taught there. Giovanni Maria Visconti, Duke of Milano, named Cristoforo Consigliere some time after 1402, and the Emperor Sigismondo named him Conte and Cavaliere in 1414, titles that were to pass to all his male descendants. Baldassarre Primo, Cristoforo’s third son, was a favorite of Filippo Maria Visconti, alongside whom he fought in battles against the Aragonese and the Venetians; some time after 1444 Baldassarre Primo accepted a call from Ludovico Gonzaga, second marquess of Mantova, to become his vassal; he built a house at Mantova and married Polisena Lisca, the daughter of a Veronese knight. Though Baldassarre Primo had grown up in a learned household, he seems to have become more a soldier than a scholar, and if the manuscript was intended for him, that circumstance makes the presence of the rather elementary notes on classical subjects at the end of libellus 4 perhaps understandable.

I suggested earlier that the improvement in quality of execution for libelli 1 and 2, copied in 1464, over that of libelli 3 and 4, copied in 1463, might be explained by an improvement in the scribe’s circumstances, and another bit of information from one of the family histories suggests what this change might have been. Clearly, the fourth libellus of the manuscript (and with it probably the third) was copied in Mantova, “in domo Baltasaris.” But the second libellus, according to its colophons, was copied in Bozzolo, a village some 20 km from Mantova. Bozzolo, in 1435, had fallen to the ownership of Ludovico Gonzaga by forfeit from Carlo degli Albertini, Count of Prato (Beffa Negrini, 280). One might wonder whether in the six months intervening between the copying of the two sections the scribe left the patronage of Baldassarre for that of Ludovico. This at present can be only conjecture; but under Ludovico’s patronage the scribe would certainly have found better materials and more splendid models than he had had access to under Baldassarre’s. Ludovico was a product of the humanistic school of Vittorino da Feltre; his learning, his wealth, and the prodigality of his patronage are well attested. It was he who called Andrea Mantegna to Mantova to paint the walls of his palaces with the frescos that remain famous today.

Whatever the precise circumstances surrounding the completion of Marciana VIII.85, it is clear that it originated in the early 1460s in and around Mantova, and very close to the corridors of power. By the end of the fifteenth century Mantova was to become a leading musical center, especially after the marriage of Ludovico’s grandson Francesco to Isabella d’Este in 1490. The musical life of Mantova in the 1460s, however, is poorly documented,[16] so the contents of Marciana VIII.85—including, of course, the treatise on how to make counterpoint using the remote hexachords with all their accidentals—provide a tantalizing glimpse into the musical interests, tastes, and practices of the city during that decade.