Few classicists nowadays read the late antique Latin ‘glossary’ attributed to Placidus,Footnote ¹ and those who would are soon confronted by the fact that Placidus cannot easily be read. The most recent edition, which appeared in W.M. Lindsay's Glossaria Latina series, patched together an original text of Placidus in a manner that few readers, if any, have found convincing.Footnote ² Most default instead to the fifth volume of the Corpus Glossariorum Latinorum, in which Georg Goetz gave essentially diplomatic editions of three versions or, better, three sets of evidence for Placidus. Those sets of evidence are printed under the titles Placidus Librorum Romanorum (= CGL 5.3–43), Placidus Libri Glossarum (= CGL 5.43–104) and Placidus Codicis Parisini (= CGL 5.104–58).

Goetz's study of six extant manuscripts of the Libri Romani text and information about a lost seventh manuscript led him to the conclusion that they all descend from a single book that had come to light somewhere and at some time in the fifteenth century,Footnote ³ and, moreover, that only three of the manuscripts merited the status of independent witnesses to this version of the text. Those three manuscripts, all preserved in the Biblioteca Apostolica Vaticana, are:

1. T Vat. lat. 1552. Contains (1) Paulus Diaconus’ epitome of Festus’ De uerborum significatu, (2) ‘glosae Placidi grammatici’ and (3) Fulgentius, Expositio Virgilianae continentiae. Written on parchment in a single fifteenth-century hand. The colophon to Paulus (fol. 93v) indicates that that text was completed on 1 July 1453 in Perugia; the colophon of Fulgentius (fol. 134v) is written in an identical script.

2. W Vat. lat. 3441. The relevant section of the manuscript contains only the ‘glosae Placidi grammatici’, although that unity has been bound together with several other texts from the library of Fulvio Orsini, who annotated the manuscript. Written on paper in a single sixteenth-century hand.

3. Y Vat. lat. 5216. The manuscript originally contained, in this order, the grammatical texts (1) Velius Longus, (2) ‘Adamantius siue Martyrius’, (3) the excerpts of Arusianus Messius transmitted under the attribution ‘Cornelii Frontonis exempla elocutionum’ (henceforth ‘Fronto’), (4) Fortunatianus, (5) Donatianus, (6) Caesius Bassus, (7) Iulius Seuerianus, (8) ‘glosae Placidi grammatici’ and (9) Fulgentius, Expositio Virgilianae continentiae. It was at some time separated into two parts and later rebound, with the first three texts now set at the end of the book. Written on paper in a single sixteenth-century hand.

In the year following the appearance of Goetz's edition, Léon Dorez described a fifteenth-century parchment manuscript at Viterbo that contains four texts written in the same hand.Footnote ⁴ These texts are: (1) Paulus Diaconus’ epitome of Festus’ De uerborum significatu, written according to its colophon die XVI mensis nouembris 1433. Basilee (fol. 70v); (2) ‘glosae Placidi grammatici’; (3) Fulgentius, Expositio Virgilianae continentiae; and (4) Pomponius Mela, De chorographia. On fol. 135r a much later hand has added various short poems.Footnote ⁵ The manuscript—now Viterbo, Centro Diocesano di Documentazione (CeDiDo), Capitolare 51 (here R)—has corrections and marginal notes in several unidentified handsFootnote ⁶ and in the hand of the Viterban humanist Latino Latini (1513–1593),Footnote ⁷ who was active in Rome and was closely connected with Fulvio Orsini and other leading figures of the Roman intellectual community in the third quarter of the sixteenth century. On Latini's death it and other items of his library were left to the Archivio Capitolare in Viterbo.

In his influential study of the rediscovery of the Latin classics in the fourteenth and fifteenth centuries Remigio Sabbadini took note of the manuscript at Viterbo and connected it generally with Goetz's text of Placidus Librorum Romanorum,Footnote ⁸ but its importance for editing the glossary has not been recognized. When preparing to edit Placidus for Lindsay's series, J.W. Pirie could write ‘we should like, if possible, to see either a Placidus MS of another family, or, failing that, an earlier MS of the same family’ to improve the text, and the evidentiary foundations of the glossary have received no augmentation in the years since.Footnote ⁹ One part, at least, of Pirie's wish can be granted, for R will prove to be an earlier manuscript of the same family. Although the relationship of R to Goetz's three witnesses means that its fresh contributions to a new edition of Placidus will be modest, the manuscript will, all the same, simplify a future editor's task and clear away some problems that the known manuscripts have left unresolved. Moreover, the Viterbo manuscript throws new light on the circulation of the text in the fifteenth and sixteenth centuries, and its history, in turn, offers important information about early modern readers of Placidus and their sources for the text. In this paper we show that R is the parent of TWY, and we discuss the kinds of gains to be had in a new edition produced from it. We then turn to the history of the manuscript in the fifteenth and sixteenth centuries; by tracing the travels of Placidus in humanistic and antiquarian circles, we aim to corroborate the claim that R provides the only independent evidence for the glossary in this period, and that Placidus therefore drew readers who had no evidence to work with other than R.

The unusual pairing of the glosses of Placidus with the Expositio Virgilianae continentiae of Fulgentius in RTY immediately suggests a close relationship among at least those manuscripts, which each have those two texts in the same order. Full collation of RTWY corroborates that impression and, moreover, confirms that R stands above TWY in the stemma, for each of those three manuscripts preserves all of R's readings or adds new errors of their own.Footnote ¹⁰ Limitations of space preclude our presenting the full collation and, since Goetz edited the Libri Romani essentially diplomatically (by reconstructing the shared ancestor, with all its faults, of TWY), only the complete collation could demonstrate that R both has all of the readings that Goetz hypothesized on the basis of TWY and has no uncorrectable readings that cannot be hypothesized from them.Footnote ¹¹

In lieu of printing the complete collation, we shall demonstrate, from peculiarities in the textual layout and in the ductus litterarum in R, that each of TWY descends from R and that they were in fact copied directly from it. R, like many other manuscripts of glossographical texts, sometimes engages in the economizing practice of writing the end of a note in blank space after a previous or subsequent note, a style sometimes called ‘head-under-wing’. This practice often generates errors of transposition in derivative manuscripts, when the trailing part of one note is wrongly added to another, and errors of omission, when a copyist overlooks either the end of a note or a line of text adjacent to it.Footnote ¹² A striking example of the latter is provided by Placidus 17.14 Decrepiti non qui a senec-] om. T. That error arose from the layout of Placidus 17.13–14 in R, reproduced here:

T passes from ponitur to -tute, and thereby conflates the two glosses into one. Similarly, the following errors in TWY are explained by the layout of the text in R (unless otherwise indicated, the error affects an entire gloss):

Particularly significant evidence comes from the two transpositions—or, more precisely, two omissions that were soon corrected by the original copyist—that were made by W at 12.18 and 41.2. In each instance, the layout of R caused the copyist of W first to omit a gloss; when that individual reached the end of a page in R, he ‘checked his work’ and, noticing the omission, added the missing gloss to the relevant page in W.Footnote ¹³ These errors confirm that TWY descend from R, for it is reasonable to assume that these peculiarities of layout belong to it alone. That fact, when taken in tandem with the absence of similar errors that cannot be accounted for from the layout of R, suggests that TWY were copied directly from R.Footnote ¹⁴

Many other readings in TWY can be explained from peculiarities in the lettering or corrections of R. Errors of this sort put the descent of TWY from R beyond doubt; their preponderance, in combination with evidence previously reviewed and with the absence of errors that cannot be explained from the appearance of R, establishes as securely as is possible that they were copied directly from R itself. We list here a sampling of these distinctive errors and explain any that are not self-evident, beginning with the evidence of T:

At Placidus 6.4 a ‘surgical’ correction through the bowl of the ‘d’ in R (meant to be read as allaterati) was understood by the copyist of T as a deletion of the entire letter. At 16.1 the correction of consatius to conlatius in R is executed in such a way that it easily reads as compatius. At 20.20 T writes twice a single line of text in R.

In the case of WY, the copyists of those manuscripts often encountered difficulties over the same features of R, such that it will be convenient to treat their errors together:

Two tendencies are worth noting. W often records an uncorrected reading of R, generally also with its correction, as at 8.8 and 19.11.Footnote ¹⁵ The copyist of Y often left lacunae when the reading was in doubt; at 25.15, lases in R is written in such a way that it could be read as sases, whence Y's uncertainty. Unusual suspensions or ligatures account for most other errors in this list, but even the perfectly ordinary superscript i to indicate -ri- (at 7.11 and 7.45) sometimes caused trouble for W. At Placidus 19.16, the first reading of R was corrected to cū garatulitate, with the first word written in such a way as to account for ái garatulitate W^ac, and in the word Concitos (13.8) the letters -ci- are written so closely together in R as to easily be read as an a.

Much evidence therefore demonstrates that TWY not only descend from R but also are direct copies of that manuscript. In his edition of Placidus Librorum Romanorum Goetz aimed at a diplomatic reconstruction of the common source of TWY; armed with, as it turns out, three independent apographs, he was able to restore the readings of R with a high degree of accuracy. In that respect, a future diplomatic edition of Placidus Librorum Romanorum would still look much like Goetz's text did before R was known: there is no ‘new Placidus’ here. But a future edition of the text will still draw many benefits from R, in the simplification of the apparatus criticus, in the corroboration of many of Goetz's intuitions about the text and, occasionally, in the correction of points on which the truth could not have been seen through TWY. In the following six notes we discuss occasions where R springs a surprise; for each we give Goetz's text and a limited apparatus criticus reporting important readings from R, followed by short explanations:

R also yields new information about the division of individual glosses. In perhaps the most important of these, at Placidus 5.36, R marks the end of a gloss after fugit and sets Aduerbium as the second word of the next gloss, confirming for this branch of the tradition the division hypothesized by a corrector of W and found in the other evidence for Placidus’ glosses.Footnote ¹⁸

Although Goetz already intuited much of the truth about R from the three apographs at his disposal, access to the manuscript thus provides better insight into the text of Placidus that came to light in the fifteenth century, and a better means of distinguishing archetypal readings (of the Libri Romani) from early improvements to the text. The readings of R that were effaced by early readers, and therefore are not found in the texts of TWY, themselves have value, for they may give a future editor insight into the earlier history of the Libri Romani text and provide ways of diagnosing and correcting the abundant errors in it.

While R simplifies the editor's task by providing new information, it also reveals new problems. Chief among these is the question of whether R and its descendants alone account for knowledge of Placidus in the fifteenth and sixteenth centuries, or some other source for the text—the antigraph of R that reached Basel, for example, or another independent witness—was available to the readers whose intermittent but focussed attention on the text of Placidus left its mark in the many corrections and annotations in R. The history of R, from its production in Basel through the death of Latino Latini in 1593, therefore takes on particular importance not only for the history of scholarship on the text but also for the editor's assessments of the tradition and its readings. We believe that full consideration of the history of R confirms that that manuscript alone provided independent evidence for the text, and that humanist scholars interested in Placidus had only it and their ingenium to work with. Establishing the history of R, then, both corroborates the stemmatic argument advanced here and provides a framework for further study of the text of Placidus. The evidence for Placidus, including any evidence that may yet come to light, needs to be assessed against the history and readings of R, which must be presumed to account for knowledge of Placidus in the fifteenth and sixteenth centuries until evidence that does not fit this framework is found.

That history begins at Basel on 16 November 1433, when, according to the colophon of Paulus in R, that text was completed. Since all four texts in R are in the same hand, it has long been regarded as certain that they were written on the same occasion. Discussing anonymous discoveries of Latin texts, Sabbadini indicated without much fanfare that the Glossae of Placidus had come to the attention of the humanists by that year.Footnote ¹⁹ Notes in some fifteenth-century manuscripts of Paulus that are closely related to R, but independent of it, connect that text more precisely with the Council of Basel.Footnote ²⁰ Those notes, which suggest a manuscript written in 1434, together with a letter of Francesco Pizolpasso to Nikolaus von Kues written in Basel on 17 December 1432 that pleads for help in obtaining a copy of Festus Pomponius (sic),Footnote ²¹ corroborate the impression that R must have been put together from texts available in diverse manuscripts in Basel during the Council. Moreover, a peculiar feature of this manuscript seems explicable only in such circumstances: R contains four texts attributed to Paulus, Placidus grammaticus, Fabius Planciades Fulgentius (or something similar)Footnote ²² and Pomponius Mela, four authors with names beginning with ‘P’ arranged in strict alphabetical order.Footnote ²³ That such an arrangement was possible must be the result of a remarkable confluence of books.Footnote ²⁴

There is no evidence to suggest that any of the four texts travelled to Basel in the same manuscript. Only the similar subject matter of Paulus and Placidus suggests a reason for any of them to have stood together before 1433, and even this pairing is fragile.Footnote ²⁵ Without further evidence about the antigraphs of R the question must be left open, but there seems to be little reason to believe that the other three texts will necessarily reveal information about an earlier copy of the Libri Romani version of Placidus, nor that it, in turn, will throw light on the descent of those texts. Results may yet emerge, however, from more information about the possessors of the antigraphs of R, and the following modest indications are offered in the hope that they may point a future editor toward further information about this book and its texts.

Little information, whether because apparently none exists or because none has yet been uncovered, is available to situate R's texts in the transmissions of Paulus, Fulgentius and Pomponius Mela. The Fulgentius certainly gave rise to the copies of the text in T and Y. The text in R is marked by many distinctive readings, most of which were reported by Helm from the descendants of R he labelled deteriores.Footnote ²⁶ It is not possible to connect R with any of the known medieval witnesses of the Expositio Virgilianae continentiae, and Fulgentius’ route to Basel remains a mystery.Footnote ²⁷

Paulus and Pomponius Mela offer moderately better prospects for individuating the sources of those texts, but for the present they must remain merely prospects. The fifteenth-century tradition of Paulus’ epitome is rather different than has yet been recognized,Footnote ²⁸ and a full account of the manuscripts will alter the narrative such that only preliminary statements are possible here. The text of R cannot be connected with any of the known medieval witnesses and it seems not to align consistently with either the α or the β manuscripts of that text. The question must therefore be left open for the moment.

The medieval circulation of Pomponius Mela has been so well explained as to require no recapitulation of the history and travels of Vat. lat. 4929 and of a copy of the text annotated by Petrarch around 1335.Footnote ²⁹ Some aspects of the subsequent transmission of the De chorographia have been partially described, others not at all.Footnote ³⁰ Although we cannot yet relate R to any extant fourteenth- or fifteenth-century manuscript, some eliminations are possible, for it is not related to the manuscripts connected to Coluccio Salutati, Niccolò Niccoli, Simon de Plumetot, Jean de Montreuil, Guillaume Fillastre, or Giordano Orsini.Footnote ³¹ If the antigraph of R's Pomponius Mela does survive, two omissions seemingly of a single line in length—Pomponius Mela 1.21 uasta est magis quam frequens mare quo cingitur and 1.50 Nilus est et cum diu simplex saeuusque descendit circa—and an interpolation (1.30 Numidia] Numidia a graecis appellata METIOSOPITHG [sic]) that shows up occasionallyFootnote ³² in the tradition will perhaps help to identify it.

At present, then, we can say only that the Libri Romani version of Placidus was noticed during the Council of Basel, where someone plucked it from among the available books and copied it along with three other texts whose authors’ names began with ‘P’. Important evidence for demarcating phases in the history of R after Basel comes from the accumulated strata of annotations in the manuscript. Although Dorez observed only two annotators at work in R, in our view at least three and perhaps as many as six different hands annotated the Placidus in the fifteenth and sixteenth centuries. Nearly all of these annotations are modest, consisting of individual words and conjectures. They are not known equally to TWY, which instead reflect different stages in the history of R and in the scholarly improvement of its text. Sorting these annotations and their presence in the descendants of R, then, will establish a preliminary chronological framework for work on the text, which we will subsequently fill out with information drawn from the history of those descendants. Named and arranged from their first annotation in the text of Placidus, these groups of annotations are:

1. [a] ‘Accipenser’ (fol. 71v): written in slightly rough lettering, often using the symbol ‘+’ to tie annotations to the main text. The same hand added the following notes, which we report in their entirety: antes (fol. 72v), cohercere (fol. 75v), Silones Non(ius) Marcel(lus) (fol. 76r), .c. explicuit (fol. 79v), hera (fol. 79v), Ianus autem in correction of Ianua (fol. 82v), non sic (fol. 82v) and proprie pedum (fol. 87v); perhaps also ał praecipitabantur (fol. 74r) and epythia (twice, fols. 79v and 83r).Footnote ³³ These annotations are known to all of TWY.

2. [b] ‘Aporria’ (fol. 71v): written in a small and graceful sixteenth-century hand that is at least close to, and possibly identical with, the hand that wrote W. The same hand added: Ad incitam, iter, trutinae and modum (all on fol. 72v), as well as tenebantur (fol. 78v), exerto humero (fol. 79v), f. fracibus (fol. 80r), f. Q. Fabius Eburnus (fol. 88r), f. fulguratus (fol. 88r) and Scythia (fol. 90r). These annotations are known to W alone; a few of them have been added to Y in obviously later hands, and there is no reason to suspect that they were present in R when Y was first copied.

3. [c] ‘Echini’ (fol. 79r): written in brown ink in the hand of Latino Latini. In Placidus, Latini annotated just two pages, adding expopulariter iactas id est, uulgoque and Equitium, deleting ost (all on fol. 79r) and also adding rubor and erogantem (fol. 90v). Only W knows these annotations.

4. [d] ‘De mensium appellatione’ (fol. 82v): marginal annotations and additions, often preceded by a paragraphus, that are written in a hand close to, but not necessarily identical with, the ‘Accipenser’ group. The hand has written the annotations Lancinare per lances diuidere, magnalia uile uerbum (both fol. 85r), and Nuptiae quamquam a nubendo dicantur scribuntur tamen per .p. quia .p. littera mitior est quam b (fol. 86v). As these annotations are known to all of TWY, they may well belong with the ‘Accipenser’ group.

5. [e] ‘Ius praetorium’ (fol. 82v): large letters written in dark ink, not precisely connected to the main text. The same hand is responsible for forte illibus sicut hibus (fol. 83v), and is perhaps a corrector contemporaneous with the first hand of R. These annotations are known to all of TWY.

6. [f] ‘Iuuentus’ (fol. 83r): large, perhaps slightly inexpert lettering written directly adjacent to the text block to indicate insertions. The same hand wrote sicut (fol. 87r) and perhaps Senectus est a senectute appellatus a Romulo (fol. 89v), unless this belongs instead with [d]. These annotations are known to all of TWY, although their response to individual annotations varies.Footnote ³⁴

The annotators thus fall into two larger groups divided by their presence in the apographs of R. Groups a, d, e and f are known to all three apographs, and so must be presumed to have been added to R in or before 1453. Conversely, the annotations of Latino Latini and those of the ‘Aporria’ group (that is, groups b and c) are known at first hand only to the copyist who wrote W, who is perhaps identical with the ‘Aporria’ annotator. Two important points emerge from this evaluation of the annotations. First, scholarly work on R falls into two broad phases, one early and represented by the state of the text as copied by T in the fifteenth century and by Y in the sixteenth, and the other later and represented by the state of the text as copied by W. Second, nearly all of the annotations are modest: they offer conjectures of the sort that would be within the reach of readers working from the corrupted text of R alone. Only a few of the annotations of the earliest phase have even the possibility of offering independent evidence for the text of Placidus.Footnote ³⁵ But their very paucity is a mark against that possibility, and it seems safer to assume that these early and limited corrections represent conjectures or, at best, access to the antigraph of R, and not evidence from a different part of the transmission, from which more extensive correction might be expected.

From the annotations of R and from its apographs, it becomes possible to establish stages in the history of the manuscript, which will corroborate the impression that R was the only source of information about Placidus available to Italian humanists. When and with whom R left Basel is not known. The manuscript was next certainly at Perugia in the summer of 1453, when its texts of Paulus, Placidus and Fulgentius were copied in the same hand in T, and when an annotator of R may have written epythia in the margin of T, if those hands are indeed identical.Footnote ³⁶ T reached the Vatican before 1475, and its descendants show no signs of acquaintance with R.Footnote ³⁷

At around the time T was written, the glossary of Placidus came to the attention of Biondo Flavio (1392–1463). Placidus, whom Biondo calls a ‘grammaticus non incelebris’,Footnote ³⁸ is named as an authority eight times in the Roma Triumphans (c.1453–1459); little evidence suggests that Biondo knew Placidus much before his work on that text.Footnote ³⁹ Four of those eight citations have nothing to do with our text of Placidus, and probably nothing to do with Placidus at all.Footnote ⁴⁰ The four citations left standing, however, lead back to Placidus, and often specifically to the Libri Romani text of the glossary. These are the citations at 13, on amphitrite (= Placidus 4.10, admittedly not distinguishable from CGL 5.47.5); at 50–1, on scaena (= Placidus 41.9 scena—exclamationibus tragicis, with the distinctive reading criminis [R] for carminis in evidence at CGL 5.98.3 and 5.148.5);Footnote ⁴¹ at 94, on ergastulum (= Placidus 19.9, with the reading damnantur);Footnote ⁴² and at 196, on murex (evidently a simplification of Placidus 32.5).

If equal measures of accuracy and inaccuracy in Biondo's citations prompt doubts about his acquaintance with this version of Placidus or with R, a mistaken citation of ‘Placiades’ (that is, Fulgentius) in the Roma Triumphans shows that, whatever degree of confusion reigns in the attributions, it is beyond doubt that Biondo had his knowledge of Placidus from R or from a manuscript very similar to it. The citation illustrating the word culleus follows immediately on that of ergastulum (94):

Fulgentius’ discussion of (neuter) culleum in the Expositio sermonum antiquorum (page 125.3 Helm) overlaps only trivially with this information. Biondo's information derives instead from the text of Placidus 10.3; we give here the text of R:

Biondo therefore certainly knew this version of the glossary, and although proiciuntur is not the sturdiest of reeds on which to lean, it does suggest that Biondo knew Placidus not from T but from R or from a manuscript very similar to it.

R seems to have made little impact elsewhere at this time. No other fifteenth-century copies of its Placidus or its Fulgentius are known. The same may turn out to be true of its Pomponius Mela.Footnote ⁴³ The impression that the manuscript was little read in a lengthy period after 1453 is not altered by the fact that just one other descendant of its Paulus is extant.Footnote ⁴⁴ From the third quarter of the fifteenth century until the middle of the sixteenth century there is a gap in the history of R. The manuscript is next glimpsed in a list of unpublished Latin texts compiled probably, but not indisputably, in the 1550s. However, since the connection between that list and R is, at present, obscured by two small but significant errors, and since that connection can only be established retrospectively, we defer discussion of that evidence and return, instead, to Latino Latini. The Viterban humanist, as we have indicated, annotated R very sparsely, making only occasional notes in the texts of Paulus, Placidus and Pomponius Mela.Footnote ⁴⁵ There is no external evidence known to us to suggest when and how the manuscript came into his possession, and the manuscript itself gives no evidence about when he annotated its texts.

Although R itself therefore reveals relatively little about its whereabouts before it came to Latino Latini, its two sixteenth-century apographs shine some light on the history of the manuscript and permit us to be a bit more precise about Latini's annotations. The argument is straightforward in the case of W, which will set a firm terminus ante quem for Latini's work on R. The individual who wrote the Placidus of W transcribed Latini's notes in the margins on fols. 77r and 95v, writing them in the same hand used for the main text and for its other inherited marginalia. Accordingly, there is no doubt that Latini annotated R before W was written. This Placidus, as is well known, was owned by Fulvio Orsini and annotated in his own hand.Footnote ⁴⁶ It is less frequently remarked that Placidus had come to his attention before 1581. In several notes in the edition of Festus printed in Rome by Giorgio Ferrario in that year Orsini cites ‘Placidus in glossis’, quoting words that correspond to the text of W, sometimes with Orsini's own conjectures and corrections silently incorporated.Footnote ⁴⁷ That sequence of events puts Latini's annotations in R a comfortable interval before the first printing of Orsini's edition of Festus.Footnote ⁴⁸ The known descendants of the Placidus of W are consistent with that claim but do not permit further precision.Footnote ⁴⁹

Rather more information emerges from the other sixteenth-century descendant of R. As we have observed, the copyist of Y seems not to know the annotations in our groups b and c, and therefore that manuscript was presumably copied before Latini annotated R. Secure information about when Y was produced therefore would provide both a terminus ante quem for the resurfacing of R in the sixteenth century and a terminus post quem for Latini's annotations. Although an exact date proves elusive, greater precision is possible on the basis of some unrecognized evidence for the history of Y. An argument that necessarily follows a meandering path through Roman libraries and Roman intellectual circles of the 1550s and 1560s will establish that Y was written before 1563, that R must have resurfaced before that date and that therefore we will not miss the mark by much, if at all, if we put Latini's work on the manuscript in the period c.1563–1581.

Part of the history of Y is revealed by Vat. lat. 3402,Footnote ⁵⁰ written in Rome around 1515 by the Friulian humanist Niccolò Liburnio (†1557). That manuscript contains texts of Velius Longus, ‘Adamantius siue Martyrius’, excerpts of Arusianus Messius (‘Fronto’), Fortunatianus, Donatianus and Caesius Bassus, all copied from Naples, BNN, IV. A. 11, followed by the text of Iulius Seuerianus, copied from a manuscript related to Petrarch's text of that author.Footnote ⁵¹ At an unknown time and by a route not traced, Liburnio's manuscript passed to Fulvio Orsini, who annotated it in his own hand and supplied the titles ATILII FORTVNATIANI ARS (fol. lxix r), DONATIANI FRAGMENTVM (fol. lxxxi v) and perhaps also ARS FORTVNATIANI (fol. lxxxiii r).

Much evidence indicates that Vat. lat. 3402 was in Orsini's possession at least by 1559, when copies of its ‘Fronto’ had begun to enter wider circulation both at Rome and elsewhere. Basilius Zanch(i)us (= Basilio Zanchi, 1501–1558)Footnote ⁵² had a manuscript of ‘Fronto’ that was copied from Vat. lat. 3402.Footnote ⁵³ That manuscript is lost, but excerpts keyed to it survive in Vat. lat. 7179, fols. 180r–181v, written in the hand of Petrus Ciacconius (= Pedro Chacón, 1526–1581) presumably after about 1570, when the Spanish scholar arrived in Rome.Footnote ⁵⁴ Another copy of ‘Fronto’ travelled north. In the autumn of 1559 Orsini sent to Carlo Sigonio (c.1520–1584) a copy of that text in aid of the latter's second edition of the Ciceronis fragmenta.Footnote ⁵⁵ Before 1565 Sigonio's manuscript had passed by way of Iohannes Zamoscius (= Jan Zamoyski, 1542–1605) to Andreas Patricius (= Andrzej Patrycy Nidecki, 1522–1587); its whereabouts are now unknown.Footnote ⁵⁶ Another copy of ‘Fronto’ deriving evidently from the same period, Vat. lat. 5170, has yet to reveal its history.Footnote ⁵⁷

Y likewise derives in part from Vat. lat. 3402 as corrected by Orsini. When Y was first written it contained, in sequence, copies of the texts of Velius Longus, Adamantius, ‘Fronto’, Fortunatianus, Donatianus, Caesius Bassus and Iulius Seuerianus (all deriving from Vat. lat. 3402), to which were joined the texts of Placidus and Fulgentius copied from R.Footnote ⁵⁸ Three recent and substantial editions of texts contained in Y have established three facts relevant to the date of the manuscript.Footnote ⁵⁹ (1) These texts are written on paper with two different watermarks, one in common use throughout the period at least 1502–1587, the other in use in Rome in the period at least 1534–1546.Footnote ⁶⁰ (2) From the Velius Longus in Y was copied Rome, Biblioteca Vallicelliana, B. 104, fols. 205r–219v, written in the hand of Achilles Statius (= Aquiles Estaço, 1524–1581). Although the date of that Velius is, strictly speaking, limited only by the death of the Portuguese scholar, the text is joined to a collection of epigraphic material that was transcribed in preparation for a work on Latin orthography and gathered probably in the 1560s and 1570s.Footnote ⁶¹ (3) Y was apparently separated into two parts before 1597, about which more in a moment. Before that separation occurred, or at least at a time when the original arrangement of the manuscript was still perceptible, the great bibliophile Iohannes Vincentius Pinellus (= Gian Vincenzo Pinelli, 1535–1601) had a copy made of substantial portions of it, now Milan, Biblioteca Ambrosiana, D 498 inf.Footnote ⁶²

On a fourth point there is more to say. At the death of Aldus Manutius the Younger (1547–1597), many of his manuscripts and printed books were seized on the authority of Clement VIII, first becoming part of the papal collection and subsequently being added to the Vatican Library in the gift of Paul V.Footnote ⁶³ It is now well established that Y is one of those manuscripts, because the fair copy of the inventory of books seized lists an Altilii Fortunatiani Ars (Vat. lat. 7121, fol. 1v), reflecting a striking error in the title of that text, which now stands at the beginning of Y.Footnote ⁶⁴

Aldus Manutius the Younger was, or was at least represented as, a prodigy. Already in 1556 a collection entitled Eleganze della lingua toscana e latina appeared under his name. To his teenage years belong, among others, the small version of his Orthographiae ratio (1561), which was followed in 1566 by a much expanded version under the same title.Footnote ⁶⁵ In the pages of the latter the authority of Velius Longus is cited twice.Footnote ⁶⁶ And at Venice, in 1563, there appeared under the name of Aldus the Younger an edition of Sallust that included the fragments of that author's Historiae.Footnote ⁶⁷ These include a fragment of the first book that is printed nihil ob tamtam [sic] mercedem sibi abnuituros (fr. I 50 Maurenbrecher = I 47 La Penna–Funari) together with the report that Aldus took those six words ex Cornelii Frontonis de exemplis elocutionum libello, nondum edito, quem legi commodatum a Scipione Pettio, uiro optimo, & singulari quadam in peruestigandis ueterum libris diligentia praedito (fol. [130v]).

If this Scipio Pettius ever existed, he has done a remarkable job of covering his tracks; no independent reports of his name are known to us. Scipio Tettius, on the other hand, is somewhat better known. That individual compiled, perhaps principally but not exclusively between the years 1550 and 1558,Footnote ⁶⁸ an Index librorum nondum editorum that was based upon his inspection of libraries and inventories, including, inter alias, those of Fulvio Orsini and Achilles Statius.Footnote ⁶⁹ In this document a lengthy list of Greek titles follows a short, but potent, list of twenty-two Latin anecdota. Six texts in that alphabetically arranged list are plainly relevant to this argument; we report them from Paris, BNF, Dupuy 651, fol. 236r, since the published version of the list makes a significant omission:

Each of these six texts appears in Y,Footnote ⁷¹ and at least two of them were cited by Aldus in the 1560s. We can bid farewell to Scipio Pettius. Instead, it was Scipio Tettius—diligent, as reported, in locating manuscripts—who provided Aldus with the unpublished ‘Fronto’ before 1563; Y, which was copied directly from R and Vat. lat. 3402 and which was in Aldus's possession at his death, must be the manuscript in which the text travelled.Footnote ⁷²

Knocking the dust off of Y and its connection with Scipio Tettius therefore allows some light to shine on Placidus and Latino Latini. R had doubtless resurfaced before 1563 and perhaps even by the previous decade, if we could be sure that the relevant items in Tettius's Index were compiled only in the period 1550–1558. Who owned R, and where it was, when Tettius saw it are questions still without answers. In subsequent years, and certainly before 1581, R was annotated both by Latino Latini and by an individual who perhaps also wrote the Placidus that belonged to Fulvio Orsini. That period cannot yet be narrowed.Footnote ⁷³ At the risk of undue speculation, a moment worth considering is the summer of 1568, when Latini enjoyed the pleasant company of Orsini, Girolamo Mercuriale and Lorenzo Gambara ‘con le Muse al fresco’ at Caprarola.Footnote ⁷⁴ Little else can yet be said about the history of R in the last decades of Latini's life, before it passed to the Archivio Capitolare in 1593.

This account of the travels of R and the gaps that remain in it provide a framework both for the early modern history of Placidus and for the editorial assessment of any evidence that has not yet come to the attention of scholars. At the beginning of this paper we recalled Pirie's wish to see a manuscript of another family, or at least an earlier witness to the text of the Libri Romani. Our argument about R has confirmed the status of that manuscript as an earlier source—and, in fact, as the only independent source yet known—for the Libri Romani text. The three manuscripts on which Goetz relied to edit this version of Placidus were copied from R at different stages of its history, and their modest improvements to the archetypal text seem to be purely conjectural. Moreover, our reconstruction of the history of R corroborates the stemmatic argument about its descendants and provides a framework for assessing any further evidence for the text of Placidus. Since R and its apographs seem entirely to account for the circulation of Placidus in the fifteenth and sixteenth centuries, it is a reasonable hypothesis that they will also account for any traces of Placidus not yet known. Should evidence come to light that cannot be explained from R and its history, it may well grant the other part of Pirie's wish.