Bolonyai ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 2017 Gábor Bolonyai Benedictus and his Greek-Latin Dictionary: Escorial Σ I.12 Gábor Bolonyai HE ONLY EXTANT Greek-Latin alphabetical dictionary from antiquity, falsely attributed to Cyril, has not at- tracted much scholarly attention. Its textual history in the 15th century may seem to be an even less fascinating topic. We have a single exemplar, Harleianus 5792 (H) dated to the eighth century; the fifteen known copies of it made in the 15th/16th century are derivative, offering nothing relevant to the constitution of the original text. However, at the time of its recovery, around the third and fourth decades of the 15th century, attempts to revive Hellenic studies entered a new phase when the need of a usable Greek-Latin dictionary was felt throughout both academic and educational circles,1 and the 1 For a comprehensive and autopsy-based survey of the different types of bilingual dictionaries during the period see P. Thiermann, “I dizionari greco-latini fra medioevo e umanesimo,” in J. Hamesse (ed.), Les manuscrits des lexiques et glossaires de l’antiquité tardive à la fin du Moyen Âge (Louvain-la- Neuve 1996) 657–675, and P. Botley, Learning Greek in Western Europe, 1396– 1529 (Philadelphia 2010) 55–70, who mistakenly lists four copies of pseudo- Cyril’s monolingual dictionary (Venice, Marc.gr. 10.30; Vatican, Pal.gr.195; Milan, Ambr. B 46 sup. olim T.211; Yale, Beinecke 291) among the exemplars of its bilingual namesake. For the vocabularies compiled by humanists see P. Thiermann, Das Wörterbuch der Humanisten (diss. Hamburg 1994). For an analyis of one particular early humanist vocabulary see A. Rollo, “Alle origini della lessicografia umanistica: prime ricerche sul Vat. gr. 877,” in J. Hamesse and J. F. Meirinhos (eds.), Glossaires et lexiques médiévaux inédits (Porto 2011) 181–214; see also N. Wilson’s essay providing new insights on certain details: “ ‘Utriusque linguae peritus’: How Did One Learn Greek and Ac- quire the Texts?” in C. Caruso and A. Laird (eds.), Italy and the Classical Tradition: Language, Thought and Poetry 1300–1600 (London 2009) 62–70. For T GÁBOR BOLONYAI 793 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 freshly rediscovered ancient manuscript “written in old letters” (“libros litteris vetustis descriptos”)2 appeared to be a most precious treasure. And if one traces the history of one particu- lar apograph, Escorial Σ I.12, and tries to identify the Bene- dictus who copied it, the story has unexpected turns. The aim of this paper is, by focussing on the manuscript evidence and the scribes’ activity, not only to define the place this dictionary may occupy in the manuscript tradition but also to explore the context of its origin. Accordingly, textual-critical analyses will be combined with narrative, biographical sections. The bilingual Pseudo-Cyril became available to Italian humanists during the Council of Basel (1431–1438) via Harleianus 5792, belonging then to Nicholas of Cusa.3 Before long several copies were made, either directly or indirectly. Goetz, the last editor of the pseudo-Cyril dictionary, lists ten 15th- or early 16th-century copies;4 Dionisotti lists sixteen.5 In his sketchy overview of Greek-Latin lexica,6 Thiermann enumerates only those six whose owner or scribe can be identified by name: Laur. Acqu. e doni 92 which was possessed Francesco da Castiglione, Laur.Edil. 219 possessed by Giorgio Antonio Vespucci,7 ÖNB Suppl.gr. 45 possessed by Janus Pan- ___ the thematic bilingual word-lists of Hermeneumata see E. Dickey, The Colloquia of the Hermeneumata Pseudodositheana I (Cambridge 2012) 20–24. 2 See P. Sabbadini, Le scoperte dei codici latini e greci ne'secoli XIV e XV (Florence 1905) 112. 3 It is mentioned by Francesco Pizolpassi in a letter of 1437, see Sabba- dini, Le scoperte 112 and 118. 4 CGL II (Leipzig 1888) XXX–XXXI. Goetz erroneously included in his list Neapolitanus II D 34, presumably by confusion with II D 33. In fact the for- mer MS. contains an unfinished humanist dictionary, see M. R. Formentin, Catalogus codicum graecorum bibliothecae nationalis Neapolitanae II (Rome 1995) 37–38. 5 A. C. Dionisotti, “From Stephanus to Du Cange: Glossary Stories,” RHT 14–15 (1984–1985) 304. She adds II D 33 to Goetz’s list, but does not recognize that II D 34 is an erratum. 6 Thiermann, in Les manuscrits 659–660. 7 They were already indentified by G. Goetz, “Glossographie,” RE 7 794 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 nonius,8 ÖNB Suppl.gr. 47 copied by Cristoforo Persona, Basil. A III 17 by Giovanni da Ragusa, and Escorial Σ I.12 by an other- wise unknown scribe who calls himself Benedictus. 1. Two exemplars of Pseudo-Cyril: Escorial Σ I.12 and ÖNB Suppl.gr. 45 Of these apographs Thiermann calls Benedictus’ “the most interesting one” on the grounds that it is lavishly supplemented with marginalia.9 He even quotes three short passages from the scholia of Aristophanes’ Nubes and Plutus that Benedictus wrote in the margin of his dictionary. It appears, as pointed out by Thiermann, that these scholia derived from the second Tri- clinian recension. In fact, this copy is not the only one with marginal notes: ÖNB Suppl.gr. 45 is furnished with similar material. And if the two manuscripts are compared, it is obvious at once that they are related to each other. In a recent study of the Vienna man- uscript, Zsuzsanna Ötvös made a partial collation of the two codices.10 She focussed on entries beginning with alpha (about 2500) and came to the conclusion that, although the two ver- sions “belong to the same branch of the tradition,” they “can- not have the same direct source since differences also occur.”11 Upon closer examination, however, a slightly different pic- ture seems to emerge about their textual relation. To go into all the relevant passages is beyond the aim of this paper, so I will restrict myself to one particular case: the marginal note where the scribe introduces himself as Benedictus (TABLE 1). ___ (1910) 1433–1466. 8 J. Bick, Die Schreiber der Wiener griechischen Handschriften (Vienna 1920) 54– 55, thought Janus was also the scribe of this MS. I. Kapitánffy, “Aristopha- nes, Triklinios, Guarino und Janus Pannonius,” AAntHung 36 (1995) 351– 357, convincingly argues that Janus was only the possessor of it. For further details see Zs. Ötvös, “Janus Pannonius’s Vocabularium”: The Complex Analysis of the Ms. ÖNB Suppl. Gr. 45 (Budapest 2015) 27–39. 9 Thiermann, in Les manuscrits 659–660. 10 Ötvös, “Janus Pannonius” 96–103. 11 Ötvös, “Janus Pannonius” 101–102. GÁBOR BOLONYAI 795 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Escorialensis Σ I.12 ÖNB Suppl.gr. 45 ἐγγυῶµαι τόνδε ἐπὶ γάµου ἐγγύης· ἐγγιῶµαι [sic] σοι δὲ τόνδε. οἷον ἀντ’ αὐτοῦ καθίσταµαι ἐγγυητὴς ἐγγυῶµαι τόνδε ἐπὶ γάµου ἐγ- γύης· ἐγγυῶµαι σοι δὲ τόνδε, οἷον ἀντ’ αὐτοῦ καθίσταµαι ἐγ- γυητής. alibi ita ego benedictus legi ἐγγυῶ ἐνεργετικῶς γαµϊκοῦ συναλλάγµατος· οἷον ἐγγυᾶ ὁ δεῖνα τῶ δεῖνϊ τὴν ἑαυτοῦ θυγατέρα, ἐγγυῶµαι δὲ σοι παθητικῶς, ἀντὶ τοῦ ὑπισχνοῦµαι σοι. ὅθεν καὶ ἐγγύη ἡ ὑπόσχεσις alibi ἐγγυῶ σοι θυγατέρα despondeo ἐγγυῶµαι σοι, τὸ ὑπισχνοῦµαι σοι, ἐγγύη ὑπό- σχεσις TABLE 1 In both manuscripts two notes are added to the entry ἐγγυῶ spondeo. The first observation (which is not noted by Thier- mann) supplements the entry by pointing to a special usage of ἐγγυῶ: the middle voice ἐγγυῶµαι means “betroth/pledge oneself to someone.”12 The two versions, at this point, differ only in orthographical details. There are more substantial divergences concerning the second note. Here Benedictus’ exemplar offers a much longer text in both Greek and Latin than the Vienna manuscript and, more importantly, the intro- ductory words are phrased in revealingly different ways: alibi ita ego benedictus legi versus alibi. The question is how the two versions are related to each other. Theoretically, there are three possibilities: the Vienna is an abridgment of the Escorial, the Escorial is an extension of the Vienna, or they have a common source. A fourth option (that the two notes were made independently of each other) can be excluded given the rarity of the two quoted passages and the 12 The first part of the note seems to come from a lexicon very close to the Lexicon Vindoboniense (ÖNB Phil.gr. 169), in which a very similar note can be read: ἐγγυῶµαι τὸν δεῖνα ἐπὶ γάµου ἐγγύης. καὶ ἐγγυῶµαί σοι τὸν δεῖνα (see Ötvös, “Janus Pannonius” 290), whereas the second half of the sentence may either be Benedictus’ own words, or simply continue the text of the same source. 796 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 extensive overlap between them. First the common source hypothesis: this means that the statement “I Benedictus read something different elsewhere” was already in the supposed antecedent text. But is it con- ceivable that someone, while copying this particular tag, leaves both the name Benedictus and the first person verb untouched if he is not identical with Benedictus? It seems to be, even for a careless copyist, a most unlikely error. Normally, of course, a copyist transcribing a sentence like this should have changed the self-referential elements, either by eliminating them or by transforming ego and legi into third-person forms. Nor is it easy to imagine that the longer remark (“alibi ita ego benedictus legi”) stems from the shorter one (“alibi”). In that hypothetical case, we should assume that the copyist, who is called, accepting this logic, actually Benedictus, made the abridged lexicon entry complete again (obviously in the pos- session of the same—quite rare—lexicon) and then sold it as his own find—a most unlikely and unmotivated instance of plagiarism. Therefore, it seems fairly certain that the Vienna manuscript depends on the Escorial. The same conclusion can be drawn from the content and the position of the two annotations. The second note, which is based on a scholium to Plutus 1202,13 corrects the preceding one. Benedictus observes that he found a piece of slightly different information about the usage of the verb in another source, according to which the active form means “to betroth a daughter to a man,” while the passive means “to promise.” The scribe’s introductory words by which he reveals his name (“ego benedictus”) and calls attention to the different origin and content of his sources (“alibi ita”) are obviously intended to make the reader aware of this discrepancy. Moreover, his two observations differ from each other in their arrangement and appearance as well ( fig. 1). While the first remark occupies the space between the Greek and Latin columns, the second is 13 As pointed out already by Thiermann, in Les manuscrits 660. GÁBOR BOLONYAI 797 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 positioned at the foot of the page and written in noticeably smaller letters. All these features suggest that in the Escorial codex the two notes were not entered at once, in one move, but in two distinct phases. Figure 1: Escorialensis Σ I.12, 141v In contrast, the Vienna manuscript not only offers a shortened version of the Plutus-scholion and without Benedictus’ personal references, but it also has both annotations written in letters of the same size ( fig. 2). Figure 2: ÖNB Suppl.gr. 45, 76v Unlike Benedictus, the scribe of the Vienna codex seems to have written both notes at once, without interruption. The ex- planation is obvious: in his master copy the two notes had already been entered and he treated them as a unit. Even more telling is that they are inserted in the wrong place in the Vienna copy, beside the lemma ἐγγυητής, which is seven items earlier on the previous page, instead of ἐγγυῶ. The scribe obvi- ously missed the point of the notes, thinking that they provided information about the special meaning of the noun ἐγγυητής. Presumably, that is why he left out also all the references to the active and medio-passive usage of the verb; he failed to under- stand their importance to the explanation. Thus, seen from any viewpoint, Suppl.gr. 45 must be considered a direct apograph of Benedictus’ Esc. Σ I.12. 798 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 2. Who is Benedictus? The evidence for Suppl.gr. 45 being a direct copy of Esc. Σ I.12 could easily be multiplied, but in this paper I wish to explore Benedictus’ copy and identify his person. As a starting point, a terminus ante quem may be provided precisely by the Vienna codex, to which a flyleaf was attached (still extant in the 18th century but now lost), saying that “it was written by Janus Pannonius when he began to learn Greek.”14 Since he started his studies at Guarino’s school in spring 1447 at the age 13, his source must have been available to him by 1447 at the earliest (but more probably he took up Greek one or two years later); consequently, Benedictus’ exemplar, as the master copy of Suppl.gr. 47, dates before 1447 or not much later. One feature of the marginal notes and observations deserves special emphasis: an almost equal stress is placed on the Latin equivalents as on the Greek lemmas. Benedictus makes correc- tions and emendations, adds observations about the correct spelling and proper usage of words in both languages, and supplies more Latin equivalents, when needed, in the same way as he does Greek synonyms. While commenting on the Latin text, at times he turns to contemporary authors as well. His notes concerning spelling are quite often accompanied by a reference to a certain “Gasp.” This no doubt stands for Gaspa- rino da Barzizza and his popular handbook on Orthographia, the first version of which appeared around 1417.15 The same applies to his illustrative quotations. Although the majority are from Greek authors, a great many come from classical Latin writers, which are again quite independent of the Greek entries. Not surprisingly, his main sources are first Cicero, then Quintilian, Pliny, Terence, Plautus, and Vergil; but occasionally other authors such as Suetonius, Livy, Lac- 14 Traditionally the statement was understood as referring to the whole dictionary, while Kapitánffy, AAntHung 36 (1995) 351–357, has seen it as a reference only to the sentence itself. 15 W. K. Percival, “The Orthographia of Gasparino Barzizza,” AION(filol) 14 (1992) 264. GÁBOR BOLONYAI 799 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 tantius, and Lucan are also quoted. He often uses the Digesta in order to give precise definitions of certain legal terms or po- litical notions. Among the classical authors a modern one, too, turns up, altogether eight times,16 whose name is abbreviated as “Leon.” and who can be identified via the quoted texts with Leonardo Bruni. These references give us a terminus post quem for dating Benedictus’ marginalia: the latest work he cites from the Florentine humanist is his letter essay An vulgus et literati eodem modo per Terentii Tullique tempora Romae locuti sint, published 7 May 1435. Figure 3: Note on urbs, Escorialensis Σ I.12, 240r In a similar way, Benedictus often makes bilingual observa- tions while adding a synonym, an equivalent, or a brief ex- planation to the entries. For example, concerning the word cives first he cites Nonius Marcellus’ definition from his Compendium ( fig. 3): proletarii cives dicebantur qui in plebe tenuissima erant et non amplius quam mille et quingentos aeris in censum deferebant.17 The note is followed by a suggestion about how to say it in Greek: οὓς 16 Added as a note to the item “dolabrum” (164v): Ep. 6.5; to “delirus” (202r): “quin etiam si ineptire placet” (Ep. 5.1); to “villam” (241r): Ep. 6.10 (=An vulgus); to “urbs” (240r): Ep. 3.9; to “signum” (254v): “tamquam sagitta- rius optimus signum, virtutem ipsam colendam sibi proposuit”; to “stamen” (259v): Ep. 2.1; to “rutuba” (267v) Ep. 6.10; to “otium” (268v): Ep. 4.16. 17 Nonius Marc. De compendiosa doctrina II De honestis et nove veterum dictis (I 228.19–21 Lindsay). 800 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 λέγοιµεν ἂν καὶ ἴσως οὐκ ἀπεικότως ἑλληνϊστὶ, πολίτας εὐτε- λεῖς. ἅτε δὴ µικρόν τϊ τελοῦντας ἐν ταῖς τῆς πόλεως εἰσφοραῖς etc. What is noteworthy about this suggestion is that it is made by Benedictus himself, who, without referring to any source of information, relies exclusively on his own competence and ac- tive vocabulary. From all these notes a picture emerges of Benedictus: he possesses an excellent command of both Latin and Greek, in both their written and their spoken forms. He expresses himself easily and fluently in both languages, freely switching between them. He has also read fairly widely in both literatures and has a particular interest in questions of grammar and correct usage of words; he is thoroughly bilingual and bicultural. Further, he is keen to solve problems of translation and interpretation, especially finding right equivalents in both directions. And as Janus Pannonius’ copy is a direct apograph of his exemplar, we may also assume that he had some personal contact with Gua- rino and Ferrara in the late thirties or early forties. Admittedly, this description does not yield an unmistakably characteristic image, and it was by chance that I could take the next step. I got a clue from John Monfasani’s monograph on George of Trebizond. In surveying George’s teaching activity in the twenties and thirties, Monfasani suggests that a certain Benedictus Bursa may possibly be reckoned among his students, who probably learned Greek from him in Venice sometime after 1427.18 The idea is based on Benedetto Bursa’s dedication of his translation of Libanius’ Alcidamas decla- mation to George of Trebizond. In his prefatory letter of dedication, which was later edited by Monfasani, Benedetto calls George “an excellent professor in all branches of science and a famous expert of both languages” (“Hanc igitur a me traductam tibi mitto, summo omnium doctrinarum praeceptori ac utriusque lingue claro lumini et ornamento”).19 Monfasani, 18 J. Monfasani, George of Trebizond. A Biography and a Study of his Rhetoric and Logic (Leiden 1976) 22–23. 19 J. Monfasani, Collectanea Trapezuntiana: Texts, Documents, and Bibliographies GÁBOR BOLONYAI 801 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 as it seems, understands this tag as a reference to Bursa’s own relationship to George as that of a pupil to his teacher. (The inference seems quite reasonable but, as we shall see, it will prove to be false; nevertheless, that for the moment does not affect our present question.) Benedetto Bursa is an almost completely unknown figure, whose name appears in none of the current handbooks and histories of literature. For his activity, Monfasani can refer only to a passage in Borsetti’s history of the University of Ferrara published in 1735.20 As appears from the Annals of the Stu- dium of Ferrara quoted by Borsetti, Benedetto Bursa obtained a position in Ferrara and gave lectures on various Latin authors some time before 1442. According to the records, he was com- missioned (along with two other “well-known colleagues”) by the city magistrates when the Studio of Ferrara after a grave period of moral and intellectual decline was, on Prince Leo- nello’s initiative, reorganized, incompetent teachers were sacked, and, on Guarino’s suggestion, new capable professors of grammar were employed.21 The official documents con- cerning the dates of his employment do not give us unequivocal information. Two registers reporting Bursa’s employment are ___ of George of Trebizond (Binghamton 1984) 146–147. 20 Monfasani, George of Trebizond 23, raises doubts about Benedetto’s origin from Mothone, as was assumed by S. Prete, Two Humanistic Anthologies (Vatican City 1964) 40, relying on the evidence of a Padua manuscript where he is called Mothonensis, and perhaps also on another manuscript which names him as Benedictus Bursa de Modono. Monfasani considers much more probable that he originated not from the small town on the southwest coast of the Peloponnese but from Forli, as it is attested by a Palatine man- uscript that calls him Foriiuliensis. 21 F. Borsetti, Historia almi Ferrariensis gymnasii I (Ferrara 1735) 50: “a Duodecimvirali Magistratu Franciscus de Campanea, Franciscus de Roma ac Benedictus Bursa de Modono [sic], qui ob eximiam eius virtutem Leo- nelli Principi charus admodum fuit, grammatici praeclarissimi conducti fuere, ut pueros Latinam linguam docerent, onere Benedicto iniuncto, Rhe- toricam diebus festis publice legendi, ac Ciceronem Plautumque inter- pretandi.” 802 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 dated 18 April 1439 and 5 November 1440, while a Memoriale in which he is mentioned belongs to 1442. Anyhow, on the next register, made in 1450, his name is now missing. Ac- cordingly, Monfasani assigns Bursa’s lectures to 1442 and dates both his student years with George and his translation of Libanius to the 1430s, before his stay in Ferrara. This dating tallies nicely well with the possible date of the apograph of which Janus Pannonius got possession in 1447 or soon after. Since Monfasani’s book a few relevant contributions have appeared, but Benedetto Bursa’s profile is still rather scanty (henceforth I use his Italian name).22 The identity of the two Benedettos is, at this point, no more than speculation lacking direct evidence. How can we verify it? If we turn to the differ- ent volumes of Iter Italicum with the intention of finding Benedetto’s hand, there are eleven manuscripts which contain one piece or another from Benedetto Bursa,23 but none of the 22 His laudatory speech on Lodovico Foscarini is briefly analysed by M. O’Connell, Men of Empire: Power and Negotiation in Venice’s Maritime State (Balti- more 2009) 58, and touched upon by M. King, Venetian Humanism in an Age of Patrician Dominance (Princeton 1986) 41 and 377, who also gives an (incom- plete) list of his writings. His Libanius translation finds mention in L. van Hoof, Libanius: A Critical Introduction (Cambridge 2014) 173, and is dated, erroneously, to 1439 by Q. Breen, “Francesco Zambeccari: His Transla- tions and Fabricated Translations of Libanian Letters,” Studies in the Renais- sance 11 (1964) 50, who also surmises him to be Guarino’s or Aurispa’s student, probably following R. Förster’s conjecture (Libanii opera VI [Leipzig 1911] 440), who suspected the dedicatee to be Guarino. The guess is taken over also by L.-A. Sanchi, “Diffusion et reception de Libanios à la Renais- sance,” in O. Lagacherie and P.-L. Malosse (eds.), Libanios, le premier huma- niste (Alessandria 2011) 21–22. E. Barile, Littera antiqua e scritture alla greca: notai e cancellieri a Venezia nei primi decenni del Quattrocento (Venice 1994) 27, quotes Filippo da Rimini’s funerary epigram written in 1448 on the news of Benedetto Bursa’s death. His literary essay comparing Plautus and Terence was published by A. Tontini, “Per la storia del testo di Plauto nell’Umane- simo. Benedetto Borsa, ‘De legendo Plauto potius quam Terentio’,” RCCM 56 (2014) 447–543. 23 Vat.Chig. J VI 21, Vat.gr. 1392, Padua 87, Ven.cod. 63 (4221), Ven.cod. XIV 250 (4717), Marc.lat. XI 141 (3942), Ven.cod. 245 (4682), Flor.Magl. VIII 1439, Ven.cod. 253 (4575), Ferrara II 305, Vat.Barb.lat. 42 (VIII 42). GÁBOR BOLONYAI 803 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 codicological descriptions suggests an autograph copy that would make a comparison with the hand of the Escorial man- uscript possible. I continued my search for possible evidence by browsing through catalogues of major manuscript collections. Finally, I came upon a Vienna manuscript, ÖNB Lat. 3370, containing, according to its index, “the works of Benedetto Bursa of Mothone written in his own handwriting.”24 We are also informed by that index that his writings were acquired, collected, and bound together by a certain Petrus Delphinus after he had sent several requests to Benedetto’s brother, Se- bastiano, chancellor of Mothone to obtain them. In examining the manuscript and comparing it to the Escorial codex, a glance is enough to recognize immediately that both stem from one and the same hand. A few words shown together will illustrate ( fig. 4.a-d): Escor. Σ I.12 213r ÖNB Lat. 3370 269r Escor. Σ I.12, 124v ÖNB Lat. 3370, 277v Figure 4.a-d What should be emphasized is not simply the striking sim- ilarity of the shape of each letter and the whole ductus of the two handwritings: the presence of the same, rather uncommon, 24 A note written by Petrus Delphinus (Pietro Dolfin) on a parchment fly- leaf at the beginning leaves no doubt about it: 1457, “OPERA BENEDICTI Burse Mothonei, que ipse propria manu scripsit atque composuit, in hoc uolumine continentur. Que ego Petrus Delphinus Geor. f. assiduis meis litteris impetraui a Sebastiano Bursa, eius fratre, cancellario Methoneo ad me mittenda, ipsaque compaginaui et in uolumen redegi. Sunt et alia opera minora minore uolumine composite ab eo.” On Pietro Dolfin see R. Zac- caria, “Dolfin, Pietro,” Dizionario Biografico degli Italiani 40 (1991) 565–571. 804 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 words in both codices is even more significant. How did these Greek words turn up in Lat. 3370 containing Benedetto’s own writings? They are marginal glosses to certain words of his own Latin poems: βρυχηθµός to fremitus, µωκίζω τὸ ἐµπαίζω to irridet. The reason was clearly didactic; Benedetto was consoli- dating his knowledge of Greek vocabulary by doing a kind of retranslation exercise. If we cross-check his achievement, by looking up these inserted Greek lemmas in his Pseudo-Cyril, he deserves an excellent mark. In the dictionary βρυχηθµός is ren- dered with the Latin equivalent fremitus and µωκίζω τὸ ἐµπαίζω with illudo, irrideo—he appears to have memorized each word quite accurately. And the latter entry presents exceptionally strong internal evidence for the close relationship between the two manuscripts on another ground as well: being a later ad- dition, it does not belong to the original vocabulary of pseudo- Cyril’s dictionary; this verb along with its Latin meaning appears in this very form exclusively in this particular exemplar (and its apograph Suppl.gr. 45). Their correspondence, there- fore, can hardly be interpreted otherwise than by the fact that both the Vienna manuscript (Lat. 3370) and the Escorial were possessed and used by the same person. The marginal glosses show that Benedetto’s back translation exercises involved several hundred entries and testify to an ex- ceptionally intimate familiarity with the thesaurus of Pseudo- Cyril. An example will illustrate this vividly ( fig. 5): GÁBOR BOLONYAI 805 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Figure 5: ÖNB Lat. 3370, 266r Benedetto uses the word crepusculum ‘dusk’ in one of his hexameters (“Me tunc expulsae liquere crepuscula noctis,” 266r), and then—presumably in a later phase of writing—he adds four Greek words to which crepusculum is given in his Pseudo-Cyril as a Latin equivalent: ὀρθρινός is written above the line, λυκόφως, δείλη, and ὀψινὴ ὧρα in the right margin ( fig. 6.a-b): Escor. Σ I. 12, 222v ÖNB Lat. 3370, 266r Figure 6.a 806 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Figure 6.b: ÖNB Lat. 3370, 266r; and Escor. Σ I.12, 203v, 130r, and 226r If we look them up in the dictionary, we can conclude again that Benedetto’s notes are completely identical with them (TABLE 2): the adjective ὀρθρινός is explained in his copy by matutinus. crepusculus. antelucanus, the noun λυκόφως τὸ περὶ τὴν αὐγὴν φῶς25 by crepusculum. diluculum, δείλη πρὸς ἑσπέ- ραν by crepusculum, and the collocation ὀψινὴ ὧρα ἡ µετὰ ἡλίου δύσιν by crepusculum.26 25 Bold letters indicate additions to the original text of the Harleianus in the Escorial copy. 26 The last two items seem to be inversions of entries from a Latin-Greek dictionary, i.e. the Greek lemma (ὀψινὴ ὧρα ἡ µετὰ ἡλίου δύσιν) seems to have been originally a definition for the Latin equivalent (crepusculum), which was a lemma originally. GÁBOR BOLONYAI 807 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Words from Benedetto’s own poem (Lat. 3370) Benedetto’s annotations to his poem (= Greek equivalents taken from Pseudo-Cyril) Entries in Benedetto’s copy of Pseudo-Cyril (Escor. Σ I.12) crepuscula ὀρθρινός λυκόφως δείλη πρὸς ἑσπέραν ὀψινὴ ὧρα ἡ µετὰ ἡλίου δύσιν ὀρθρινός matutinus. cre- pusculus. antelucanus λυκόφως τὸ περὶ τὴν αὐγὴν φῶς crepusculum. diluculum δείλη πρὸς ἑσπέραν cre- pusculum ὀψινὴ ὧρα ἡ µετὰ ἡλίου δύσιν TABLE 2 It should be stressed that each Greek word, beginning with a different letter, is found in a different place of the dictionary, so it was Benedetto who collected and ‘united’ them. He evidently learned the Greek lemmas along with their Latin equivalents so thoroughly that, encountering a particular Latin equivalent, he was able to recall all the Greek lemmas to which it belonged in his dictionary. On the one hand, these ‘reverse search’ exer- cises reveal his admirably active knowledge of Greek vocabu- lary; on the other, as they do not aim to find synonyms or phrases fitting into a given context, their usefulness is limited to translating a given word or tag (e.g. it is not certain that each of the four Greek expressions for crepusculum would be suitable for a translation of the sentence in Benedetto’s poem). 3. Benedetto Bursa’s early career The Vienna manuscript (Lat. 3370), at the same time, offers much more than a possibility to identify Benedetto Bursa’s hand. From the writings it contains—almost one hundred let- ters, several dozen poems, various kinds of speeches, personal notes and sketches—an almost completely unknown humanist’s career emerges, a remarkable life full of twists and turns. He was born around 1414 in Modone (henceforth I use the Italian name of the town instead of other versions: Mothone in 808 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Latin, Methone in ancient Greek, Methoni in modern Greek), an important port at the southwestern corner of the Pelo- ponnese, on the route between Venice and the eastern Medi- terranean. The chancellor of the town was his father, who was succeeded in his office by Benedetto’s elder brother Sebastiano. After his years of study in Venice, Benedetto began his teaching career with a flying start: in 1435, before the age of twenty-one, he became professor of poetry and rhetoric in the Florentine Studio as a colleague of Carlo Marsuppini.27 His correspondence does not reveal how and on what grounds he was appointed to this highly prestigious position; but his name in the form “Benedetto di Niccolò Borsi da Venezia” with the salary of 50 florins duly appears in the Communal Fiscal Records among the names of other readers employed at the Studio.28 He left Florence in his second school year when the term was interrupted and cancelled because of the plague in 1437.29 The next year he moved to Ferrara, where he joined the entourage of Pope Eugene IV arriving at the council. Not much later he was invited by the “twelve wise men,” i.e. town councillors of Ferrara.30 No matter how successful he might have been in his 27 “Vigesimum et unum annum nondum agebam cum adolescentulus paene imberbis rhetor publice ad tradendam rem oratoriam sum delectus. At in qua urbe? Nempe florentiae. In ea ciuitate nondum paene genis pri- mam lanuginem ducebam cum florentinorum ciuium magistratus quin- queuiralis mihi et nomen praeceptoris detulit et mercedem decreuit ut in pubico gymnasio hominum doctorum una cum carolo Aretino rem poeti- cam et oratoriam traderem” (ÖNB Lat. 3370, 9v [29r]; the Vienna codex was later rebound and renumbered: I give the earlier page numbers, originating presumably from Pietro Dolfin, then the later ones in brackets). 28 K. Park, “The Readers at the Florentine Studio according to Comunal Fiscal Records (1357–1380, 1413–1446),” Rinascimento 20 (1981) 294. 29 “docendi in ea ciuitate annum munus confeci: et altero deinceps anno ni ita fortuna tribuisset vt pestis in vrbem serperet ciuesque fugaret: maiore fructu et laude non caruissem” (127r [149r]). 30 “a senatu duodecim sapientium ferrariae … essem delectus magister ludi” (64r [86r]). GÁBOR BOLONYAI 809 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 task as a severe grammar teacher in enhancing the students’ language skills, he did not find much pleasure in this tiresome job.31 And for several months he did not receive his full salary as agreed to in his contract,32 so he had to turn to Prince Leonello for supplementary support to bridge the gap.33 On top of that, he was not on very good terms with his colleagues, so he was eager to leave Ferrara at the end of his contract. He had several offers from different places; eventually his choice fell on Rome over Padua and Siena. He was employed by the city government as professor of rhetoric and oratory in 1441 at the university called Studium Generale,34 reorganized by Eugene IV in 1431.35 In obtaining his appointment he was supported by a secret patron, a cardinal he usually calls reveren- dissime pater but never by his name, who is perhaps to be iden- tified with Lodovico Trevisan. The next three years in Rome proved successful in both pedagogical and financial terms. He managed to save enough money to fulfil the dream of his life: a several-year journey in Greece where he could learn Greek properly from a competent native-speaker. 31 A couple of years later he described his activity in Ferrara in bitter words: “cum laboris et fastidii plenam rem tum infructuosam et paene sterilem” (1r [21r]). 32 “Dabantur mihi ferrariae quotannis de publico aurei quinquaginta” (1r [21r]). As appears from his letter to Leonello d’Este, he was not satisfied at all with this salary: “constitutum est mihi stipendium in singulos menses librae quinque: laborum meorum, illustrissime princeps, exigua merces” (64r [86r]). Moreover, in the end he was given only 20 florins: “legi iam annum unum cumulatissime plenum, et eo tempore vniuerso exigere a com- muni viro mercedem non potui praeter uiginti libras” (21v [92v]). 33 As we learn from a later letter, Leonello gave him an additional 25 florins from his own fortune: “Illustris princeps Leonellus qua est eruditione et liberalitate ad publicam mercedem meam annuum munus adiecerat de ratione sua aureos uiginti et quinque” (1r [21r]). 34 “Delegerunt me rhetorem principes urbis huius, et ad docendum in re oratoria publice praeceptorem me vobis constituerunt” (6r [26r]). 35 See P. F. Grendler, The Universities of the Italian Renaissance (Baltimore 2002) 121. 810 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 4. Benedetto Bursa and George of Trebizond Before exploring the details of his journey, we should return to the question of his relationship with George of Trebizond. Among his letters written in Rome are two that show in a special light the possible motives of Benedetto’s dedicating his Libanius to George. Both letters were written to his younger brother Santius (Sante?), who attended George’s classes in Florence between 1440 and 1442.36 In one, Benedetto ex- presses his respect for George as a scholar but does not hide his very low opinion about him as a teacher and a man. He considers him haughty and arrogant (“sibi multa attribuen- tem,” 194v [207v]), whose knowledge is awesome, but whose Latin is seriously deficient.37 In the other, he calls into doubt George’s conscientiousness and effectiveness as a teacher.38 Considering these pointed remarks, we can be fairly certain that it was not the gratitude of a pupil to his teacher that moti- vated Benedetto to send George his translation of Libanius: it must have been rather a mixture of inferiority complex, jealousy, and self-esteem of a rival who desires to prove his own abilities to his more successful and more respected elder colleague, whose success and respect he considers as not fully deserved. In all likelihood, Benedetto during his stay in Ferrara must have been ashamed of his ignorance of Greek before George, and now in the possession of a workable knowledge of Greek he must have wanted to show him the high level he had reached since their last meeting.39 36 See Park, Rinascimento 20 (1981) 298–299. 37 In one of his letters addressed to his brother he ‘accidentally’ uses the words subucula and subligaculum, then rebukes himself for using these rare words which his brother, being a pupil of George, is not supposed to know: “Sed usitatius et aptius tecum loquar. Nam subucularum et subligaculorum significatum a Trapezuntio te non didicisse certo scio” (192r [206r]). 38 “eius praeceptoris diligentia atque traditione proficere neminem posse adolescentem nisi sublevetur et erigatur suopte ingenio atque industria” (192r [206r]). 39 Thanks to the identification of Benedetto Bursa’s hand, we can recog- GÁBOR BOLONYAI 811 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 5. Benedetto’s study trip to Greece The affair should be dated to his stay in Greece. He set off for Greece in autumn 1444. He was accompanied by one of his students, a certain Cristoforo, whom he regarded as his most talented pupil. After a stormy voyage, they cast anchor in his native town, Modone. Benedetto became so sick from the rough sea that he suffered from fever and vomiting, and swayed between life and death for three months (at least, as he perceived the causes of his own illness). When he recovered, his first trip led to Mistra to meet Gemistus Pletho, the passionate adherent of Platonic philosophy, Bessarion’s master, who also took part in the Council in Florence, where he exerted such a strong influence on Cosimo de’ Medici that Ficino, among many other Platonists, saw the starting point of Plato’s later popularity and cult in their meeting and discussions.40 Bene- detto, however, was frustrated in his hopes. In Gemistus’ per- son he found a tired and morose old man, who responded to questions with several minutes long silence, for a start.41 ___ nize him as the scribe of one of the copies that preserve his own Libanius translation. This autograph manuscript is Vat.gr. 1392, which contains the Greek text on the verso of each leaf and the Latin translation on the recto (93v–107r). Both the Greek and the Latin text were written by Benedetto himself, as well as the rest of the codex containing five more declamations of Libanius, one of Choricius, three orations of Isocrates, and one of Aeschi- nes. For a detailed description of the manuscript see M. Fassino, L’Encomio di Elena e il Plataico di Isocrate: studi sulla tradizione manoscritta e testo critico (diss. Milan 2010) 30–31, who dates the manuscript on the basis of watermarks to the middle of the 15th century and describes the Greek and Latin scribes as two unknown coeval persons. 40 J. Monfasani, “George Gemistus Pletho and the West. Greek Émigrés, Latin Scholasticism, and Renaissance Humanism,” in M. S. Brownlee and D. Gondicas (eds.), Renaissance Encounters: Greek East and Latin West (Princeton 2013) 19–34, convincingly argues that Pletho’s influence was rather a back- projection by later Neoplatonists than actual fact; and already J. Hankins, “Cosimo de’ Medici and the ‘Platonic Academy’,” JWarb 53 (1990) 156– 157, is sceptical about the reliabiity of Ficino’s much later account. 41 “ille in respondendo primum duxit horam silentio” (76r [97v]). 812 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Finally, he told him that he was not master of his own person. As he was slowly stammering his explanation, Benedetto soon realized that there was no chance that he could learn Homer’s and Demosthenes’ language from the old philosopher even if he had had time for it. He spent the next couple of months trying to find a capable teacher, but despite several attempts he failed. Finally, a gentle- man from Negroponte he had met in Venice as a young student came to his mind. In the meantime, this acquaintance had gained general recognition as the official interpreter of the Greek delegation at the Council of Florence. His achievement was so highly respected even by the Latins that after the Coun- cil he was commissioned by the Pope to perform several diplo- matic tasks as his personal emissary. The man, who was still in the service of the Venetian State as well, accepted Benedetto’s request despite his many and various obligations, and so it hap- pened that Benedetto Bursa became Niccolò Sagundino’s pupil in late August 1445.42 He spent the following months along with Cristoforo in his new teacher’s house in Negroponte, learning with full intensity. To deepen his knowledge, he also made a trip to Constantinople and Crete the next spring, but on his return trip his boat got into stormy weather again. After landing in Negroponte he suffered from recurrent bouts of fever again, and he died perhaps as a consequence of this ill- 42 See his letter sent to Sebastiano on 4 November 1445: “ igitur ut profecerim euripi: in qua tandem urbe praestantissimum praecep- torem inveni nicolaum sagudineum quem iam supra duos menses habeo” (46r-v [66v–67r]). In a letter dated 13 June 1445 and sent by Benedetto to Sagundino with the request to be taken on as a student, he reminds him of their previous meeting in Venice: “Teneo enim memoria vir praestantissime verba illa quae mihi uenetiis dixeras cum in patriam tuam esses rediturus, cum in ea urbe veneta tecum collocutus ostenderem tibi auiditatem meam graecae eruditionis. hortabaris ut Euripum venirem tecum: futurumque pollicebaris ut in ea tua patria non modo graece perdiscerem te praeceptore sed etiam quaestum consequerer adiutore eodem te mihique etiam nescio quem nummum aureorum certum expressisti: quos affirmares me capturum esse quotannis in ista urbe vestra si latine docerem” (201v [216r]). GÁBOR BOLONYAI 813 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 ness. His last letter, which relates his sea voyage, is dated May 1446; his death occurred presumably not much later, though the news of it reached Florence only in 1448.43 As to the vocabularies, it seems, purely on external grounds, quite certain that Benedetto’s copy was made directly from his teacher’s exemplar. As appears from his letters, although he tried to learn Greek on his own already in Italy, his knowledge was only rudimentary when he arrived in Greece. One aspect of his intensive language learning has already been touched upon. His memorizing exercises reflect an especially deep familiarity with his copy of Pseudo-Cyril: he appears not only to have learned by heart all the meanings of each Greek word, but also to be able, given a Latin equivalent, to recall all the Greek lemmas it stands for in the dictionary. These exercises can be found in the manuscript beside letters that are dated to around 1445; fig. 7 shows a page full of annotations: Figure 7, ÖNB Lat. 3370, 277r 43 See n.22 above (Filippo da Rimini was instructor at the San Marco chancery school from 1446 to 1450). 814 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Further, Sagundino’s person can explain several features of the Escorial manuscript and details concerning its origin. First, given what can be known about the lives of the persons con- cerned, he can serve as the perfect missing link between Nicholas of Cusa, the owner of the only extant copy of Pseudo- Cyril, and Benedetto Bursa. The German cardinal also took part in the Council of Florence, and it seems quite natural that he showed and lent his dictionary “written with old letters” to the main interpreter of the assembly.44 Sagundino was not only fluent in both languages, but also well educated in both cultures, and even familiar with theological and philosophical questions. It is mentioned by several eye-witnesses that he translated complicated argumentations with admirable com- petence and speed in both directions. On one occasion, he is recorded to have corrected one speaker’s inaccuracies, Andreas of Rhodes, who quoted one of Maximus’ letters imprecisely.45 His later translations of Greek historical and tactical writings into Latin, as well as his own literary works written in Latin, clearly manifest a broad familiarity with classical Latin litera- ture. It can be reasonably assumed, therefore, that Sagundino not only put in missing diacritical signs while copying the original text but also emended corrupt readings, added Greek synonyms and more fitting Latin equivalents when needed, and supplied in the margins several hundred quotations (ranging from one word to whole sentences), producing all the details 44 J. Monfasani, “Nicholas of Cusa, the Byzantines, and the Greek Language,” in M. Thurner (ed.), Nicolaus Cusanus zwischen Deutschland und Italien (Berlin 2002) 215–252, makes a good case for thinking that Nicholas was not completely ignorant of Greek as is often assumed, but had a tolerable reading knowledge. 45 For Syropoulos’ report on Sagundino’s activity as an interpreter in the Council see L. Laurent, Les ‘Mémoires’ de Sylvestre Syropoulos sur le concile de Florence (Paris 1971) 335. Testimonies and evaluations of his performance are collected by Π. Λ. Μαστροδηµήτρη, Νικόλαος Σεκουνδινός (1402–1464) (Athens 1970) 39–42; see also J. Gill, The Council of Florence (Cambridge 1959) 165; Laurent 262, 326, 493; and C. Caselli, Ad serenissimum principem et invictissimum regem Alphonsum Nicolai Sagundini oratio (Rome 2012) xiii–xiv. GÁBOR BOLONYAI 815 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 which make the Escorial copy special and unique. A beginner like Benedetto would not have been able to do that. In con- trast, Sagundino is the perfectly fitting person, who can be imagined to have had both the intellectual capacity to carry out this uncommon proof-reading process and the actual chance of having access to the only extant copy of pseudo-Cyril’s dic- tionary. In addition, certain types of passages cited in the margins can be also more easily accounted for if we attribute them to Sa- gundino. The group of legal definitions taken from the Digesta or its Greek translation is a case in point. Most of the Greek passages appear in Janus Pannonius’ copy as well, and the pos- sible source of these notes is to be identified, as pointed out by Zsuzsanna Ötvös, as the Basilica, the Greek version of Corpus Iuris Civilis, more particularly with an epitome of it containing abridged extracts of the entire original text.46 Now, Sagundino was a practising lawyer, a legal representative of the Venetian State serving as advocatus curiae in Negroponte.47 His later diplo- matic career was also partly based on his legal expertise and experiences (apart from his exceptional language skills, which allowed him to acquire fluency in Turkish as well while he spent several months in Turkish captivity as a prisoner of war). It is easily imaginable therefore that he knew by heart the pre- cise definitions of certain legal terms and, when he encountered a given expression while copying, added the relevant definition currente calamo in the blank space of his manuscript. Seen from the opposite side, one might say the Pseudo-Cyril went into the best hands possible, since in its vocabulary is a considerable layer formed of basic expressions of Roman law.48 These terms 46 Ötvös, “Janus Pannonius” 119–135. 47 F. Babinger, Johannes Darius (1414–1494), Sachwalter Venedigs im Morgen- land, und sein griechischer Umkreis (Munich 1961) 11–12; Μαστροδηµήτρη, Νικόλαος Σεκουνδινός 30–31. 48 See W. Bannier, “Die römischen Rechtsquellen und die sogenannten Cyrillglossen,” Philologus 71 (1912) 238–266. 816 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 might have easily invited a reader with legal training to make comments on them. Nevertheless, it would be premature to conclude that all the annotations stem from Sagundino. Just the passage cited above, where Benedetto reveals his name, shows clearly that some marginalia may have two strata. As we saw, the very reason why Benedetto introduced himself was that he had found a piece of information (in an Aristophanes scholion) slightly different from what he found in the note he had just copied. Also, the two annotations stand apart and differ sharply from each other even in their outlook and arrangement ( fig. 1): the second does not continue the first between the Greek and Latin columns, but is placed in the lower part of the page, and is written in decidedly smaller letters and much longer lines. The most obvious explanation of their separate position and differ- ing appearance is that whereas the first was transcribed from Sagundino’s exemplar, the second was, in all likelihood, added in a later phase by Benedetto himself using, as we saw, a differ- ent source. We can dispense with further examples; in general, the greater part of the notes seem to have come from Sagundino, the smaller part from Benedetto, though in several cases it is impossible to determine the authorship for certain. The same applies to the process of emendation and supplementation of the entries: one has the impression that in this respect, too, Sagundino was more active than Benedetto.49 But rather than going into the details of this process, I will focus on one im- portant event in Benedetto Bursa’ s life. 49 The question is to remain undecided, presumably, forever, because Sagundino’s copy was, in all likelihood, lost in the sea with all of his other books and possessions, when he suffered a shipwreck in 1460. See his friend Pietro Parleo’s words in his letter of consolation (Padova Lat. 87, 16r): “aurumque argentum, libros, supellectilem, res fortunasque omnes longo tempore ac magno labore partas uno momento fracta navi fortuna eri- puisset”: Miscellanea di varie operette all’illustriss. sig. Abate d. Giuseppe Luca Pasini pubblico professore nella Regia Università di Torino II (Venice 1740) 48. GÁBOR BOLONYAI 817 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 6. Benedetto and his companion Cristoforo In one of his last letters he is overcome with rage. Bitter disappointment is mingled with fury and a desire for revenge, anger with shame and frustration. The cause is that his com- panion, Cristoforo, who swore an oath of fidelity to him on Christ’s body that he would never abandon him during the whole journey, left him alone when he was lying sick in Sagun- dino’s empty house in Negroponte and returned to Sebastiano in Modone. Naturally enough, Benedetto is angry with his brother as well for receiving his treacherous pupil in his own house. In this situation Benedetto asks the addressee, his brother-in-law Jacobus (Giacomo), living also in Modone, to read aloud a curse poem, attached to the letter, to the “rascal” ( furcifer) Cristoforo, directly into his ears, warning Giacopo not to let anyone in Modone obtain knowledge of the poem, be- cause he would feel ashamed before his fellow-citizens. At the same time, he has another request: he asks him to forward a copy of the same curse to Rome in order to inform his friends and acquaintances there about Cristoforo’s perfidy. Unfor- tunately, the poem, more than 500 hexameters,50 has not sur- vived, but the text of the violated oath is cited word-for-word in the letter, so we learn that “this treacherous villain,” “this Roman wolf” (“lupus iste Romanus”), this “wretched dwarf” (“nannus aerumnosus”) is none other than the future prior of the Monastery of Santa Balbina, the would-be chief librarian of the Vatican Library, the acclaimed translator of historical and theological works: Cristoforo Persona ( fig. 8). Ego christophorus persona Romanus polliceor preceptori meo benedicto bursę me apud eum esse commoraturum! obligoque me et iureiurando diuincere fidemque do futurum ut eum tem- pore nullo deseram, non modo in his grecis finibus morantem, sed etiam si uolet in Italia. Et quoniam cupit ipse preceptor meus 50 “Hisce meis litteris subiunxi atque connexui … heroicos versus Latine per me compositos supra quadringentos”: 221v. 818 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Figure 8: ÖNB Lat. 3370, 221v Euripi et deinde in insula creta et tandem constantinapoli com- morari: promitto me secuturum preceptorem meum et in eiusmodi locis una secum habitaturum et tandem in Italiam eodem duce meo rediturum. Hec me seruaturum omnino con- firmo! Iuroque per sacratissimum corpus Iesu Christi. Iuro ad aram stans apposita manu dextera ad ipsum corpus Iesu Christi modo sacratum a praesenti sacerdote et suo numine suisque sacris uiribus integratum. I, Cristoforo Persona, promise that I will stay with my teacher Benedetto Bursa, obliging and binding myself by oath, and I pledge that I will never abandon him, neither in these parts of Greece, nor in Italy if he wishes. And since my teacher intends to stay in Negroponte, then in Crete, and finally in Constan- tinople, I promise that I will follow him, dwell with him in all these places, and finally return to Italy under his guidance. I confirm that I will keep my oath in any circumstances. I am swearing on Jesus Christ’s sacred body, and I am swearing by standing close to the altar and touching with my right hand Jesus Christ’s very body that has just been consecrated by this priest and recreated in his name and through his sacred power.51 51 Both the sacred oath of fidelity sworn in front of an altar and the in- tensity of anger caused by its breach suggest an uncommonly strong—and asymmetrical—emotional bond between the two persons. Their teacher- pupil relationship seems even more unconventional, if we take into con- sideration the difference in social and financial status between them. Al- GÁBOR BOLONYAI 819 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Up to now little has been known about Cristoforo Persona’s Greek studies.52 The only relevant evidence is provided by a letter written to him by Theodore of Gaza.53 The famous Greek scholar calls upon Cristoforo to take over the task of translating Origenes’s Contra Celsum that Theodore had prom- ised a few years earlier to Pope Nicholas V, because, with his other obligations, he had no time to accomplish it. Praising Cristoforo’s excellent knowledge (and even his native-like pronunciation) of Greek, Theodore refers to his study “with Greek teachers in the heart of Greece”: “quem unum novi ab ineunte adolescentia sic graecas litteras imbibisse et, quod plu- rimum iuvit, in Graecia ipsa et Graecis ex praeceptoribus ut, nisi te civem Romanum scirem et propinquos tuos primarios urbis viros sat nossem, dicturus facile sim e Graecia te ori- ___ though Cristoforo, born to an aristocratic family, was presumably much better-off than his master, it was Benedetto, as becomes evident from his letters, who alone paid for their accommodation and covered perhaps the whole cost of their shared journey. In Negroponte Benedetto seems to have run through his money and they were forced to make copies of official doc- uments to get some income. The business did not fare well, so financial difficulties also might have easily contributed to Cristoforo’s decision to abandon his teacher. 52 For his career see A. Zeno, Dissertazioni Vossiane II (Venice 1753) 134– 149; P. Paschini, “Un ellenista romano del Quattrocento e la sua famiglia,” Atti del Accademia degli Arcadi 19–20 (1939–1940) 45–56. 53 The letter was published by Cristoforo Persona himself as a preface to his Origen translation, which was dedicated to Sixtus IV and published in 1481, Origenis Contra Celsum et in fidei Christianae defensionem libri (Rome 1481) 20. On the translation see A. Villani, “Cristoforo Persona et la première traduction en latin du Contre Celse d’Origène,” in A. Villani (ed.), Lire les Pères de l’Église entre la Renaissance et la Réforme (Paris 2013) 21–54; A. Rita, “La versione latina di Cristoforo Persona del Contra Celsum di Origene nell’ esemplare della Vaticana di Sisto IV,” Miscellanea Bibliothecae Apostolicae Vati- canae 20 (2014) 679–694; L. M. Santini, “Le traduzioni dal Greco,” in M. Miglio et al. (eds.), Un pontificato ed una città. Sisto IV (1471–1484) (Vatican City 1986) 95–96; C. Stringer, “Italian Renaissance Learning and the Church Fathers,” in I. Backus (ed.), The Reception of the Church Fathers in the West (Leiden 1997) 493. 820 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 undum; nam ipsa et tua graeca pronunciatio te graecum praefert.” As to the identity of his teachers, only guesses have been made so far;54 his former rhetoric teacher Benedetto’s corres- pondence now throws new light upon this aspect of his life as well. Nonetheless, in Theodore’s appreciation there are two minor details that may raise further questions, especially in light of the new evidence provided by Benedetto Bursa’s cor- respondence. First, Theodore uses the word praeceptor in the plural, and it is not easy to decide whether it is only an exag- gerative plural, a sort of flourish, or should be taken literally. Perhaps Theodore knew, or simply assumed, that Cristoforo during his stay in Greece got in contact with other scholars in addition to Sagundino; he might have referred to such personal meetings and communications, from which he obviously must have profited greatly. But it is not to be excluded either that what Theodore meant is that Cristoforo actually spent further time in Greece after his journey with Bendetto Bursa and studied with other teachers as well (after the first steps made with Benedetto). The question should remain undecided. The second point concerns his age. Theodore Gaza describes 54 Hankins, JWarb 53 (1990) 158, referring to a private communication with N. Wilson, tentatively identified the scribe of Laur. LXXXV.9 (the famous exemplar of Plato’s works that Cosimo gave Ficino to translate) as Cristoforo Persona. Their suggestion was taken up as a fact by P. B. Rossi, “ ‘Diligenter notare’, ‘pie intelligere’, ‘reverenter exponere’: i teologi medie- vali lettori e fruitori dei Padri,” in M. Cortesi (ed.), Leggere i padri tra passato e presente (Florence 2010) 62. The hypothesis is contradicted, as it seems, even by chronology, since there is a growing consensus that the manuscript should be dated to the 14th century: S. Azzarà, “Note su alcuni codici di Platone e Diogene Laerzio: la datazione del Laur. 85.9 e il Marc. Gr. 189,” Res Publica Litterarum 25 (2002) 164–171; T. Dorandi, “Estratti della recen- sione laerziana delle Divisiones aristoteleae nell’Ambrosianus C47,” CodMan 41 (2002) 31–32; D. J. Murphy, “The Basis of the Text of Plato’s Charmides,” Mnemosyne 55 (2002) 150–151. Hankins also raised the possibility that Per- sona was a student of Gemistus Pletho, just like his later mentor, Isidore of Kiev. Persona did come very close to this opportunity, as we saw, but Pletho decided otherwise. GÁBOR BOLONYAI 821 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Cristoforo as a very young person (ab ineunte adolescentia) at the time of his trip to Greece. Though at first glance this detail may seem only a plausible guess by Theodore, it is confirmed by what we can know about him from Benedetto’s letters. Benedetto started to teach him oratory around 1442, when he was obviously a teenage student attending one of Benedetto’s classes. As appears from their first communication, though Cristoforo had already achieved a certain level in Latin com- position, he was a naive inexperienced novice at that time.55 He was certainly not a twenty-six-year old pupil, a kind of late- beginner, let alone a young scholar, as we ought to think if we accept the traditional date of his birth, 1416. This date would also imply that he was only two years younger than his pro- fessor, who was born in 1414, but there is nothing in the text of the letters that might suggest that there was only a small age difference between them; and Gaza’s expression, too, sharply contradicts such an assumption. So we have good reasons to date his birth date about ten years later. 56 7. A copy of Pseudo-Cyril by Cristoforo Persona Cristoforo Persona, however, did not simply accompany his teacher on his journey to Greece. As already mentioned, he also played a part in the textual history of pseudo-Cyril’s dictionary, by transcribing a copy, which is also now in Vienna (Suppl.gr. 47). If we collate it with Benedetto’s exemplar we can conclude that Cristoforo’s copy was also made—either directly or indirectly—from that of Sagundino. In order to show their close genealogical connection, I cite, for the moment, only one passage containing a major conjunctive error. In Benedetto’s exemplar at the letter χ a scribal note warns the reader that the 55 At Cristoforo’s request Benedetto sent him a kind of letter of assess- ment, which is basically enthusiastic praise of his pupil’s exceptional writing skills as well as his character (202v [217r]). 56 Hankins, JWarb 53 (1990) 158, also voices reservations regarding the traditional birth date, assuming that it “is probably a deduction of the three- score-ten variety” (for Persona died in 1486). 822 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 order of entries has been confused: πρόσχες ὁ ἀναγινώσκων τὸ ἀπὸ λήθης συµβὰν. ἐντεῦθεν οὐκ ἐφεξῆς ἕπεται τὸ συνεχὲς καταβατὸν ἀλλὰ τὸ µετ’ αὐτὸ ἀρχό- µενον ἀπὸ τοῦ χρίω· µετ’ ἐκεῖνον δὲ, εὐθεὺς τὸ συνεχόµενον τῷ παρόντι ἀρχόµενον ἀπὸ τοῦ χυδαῖος, κἀντεῦθεν σύναψον τὸ µετὰ τὸ δεύτερον ἀρχόµενον ἀπὸ τοῦ χωροφύλαξ. Watch out, reader, a mistake has been made. Here comes not the page which should follow in order and begins with χρίω but the next one after that. And immediately after that the page should come which now follows this one and begins with χυδαῖος, and then you should continue with the next one, be- ginning with χωροφύλαξ. What might have happened, it appears, is that two passages were transposed, resulting in a sequence of ACBD instead of ABCD (B beginning with χρίω, C with χυδαῖος, D with χωρο- φύλαξ). Since both B and C consist of 31 lines, it is reasonable to infer that each passage originally filled one page in its master-copy and it was two pages that changed places—for example, by one leaf simply being turned around. It is im- portant to stress that in Benedetto’s copy the boundaries of the misplaced passages do not fall on page-openings and page- endings. The error, therefore, must have occurred in its pre- decessor (by the same token, this discrepancy may also serve as evidence that it does not stem directly from the Harleianus, since there the relevant passages are not dislocated ). As to Cristoforo’s copy, the entries in question follow each other in the same wrong order as in the Escorial manuscript. The lemma χρίστης (i.e. the last item of A) is not followed by χρίω (first item of B), but by χυδαῖος (first item of C). From χυδαῖος up until χωροµετρῶ (last item of C) all the entries are transcribed in alphabetic order. After χωροµετρῶ comes the passage starting with χρίω and ending with χρώς (i.e. page B). Thus, the sequence is again ACBD, just as in Benedetto’s copy. Since it is hard to imagine that this particular error of trans- position was committed in the two manuscripts independently of each other, there is an especially strong case that they derive from one and the same ancestor. With a close stemmatic relation established between Escor. Σ GÁBOR BOLONYAI 823 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 I.12 and Suppl.gr. 47, another disputed issue can be revisited and judged from a new angle: the attribution of the hand of the Vienna manuscript. In fact it is not unanimously agreed that it was Cristoforo Persona who copied the dictionary section in Suppl.gr. 47.57 Two points may raise doubts about the attri- bution. First, Suppl.gr. 47 consists of two parts: the Greek-Latin dictionary (3r–94r) and Guarino’s Erotemata (94v–101r), each ending with a colophon; but whereas the first indicates on f. 94r simply the end of the “book” (τέλος τῆς βίβλου ταύτης. θεῷ δὲ διὰ παντὸς δόξα), the second on f. 101r is more detailed, giving the title of the copied work, the scribe’s name (τέλος σὺν θεῷ{ς} τῶν ἐρωτιµάτων[!] παρ’ ἐµοῦ γεγραµµένων τοῦ Χριστο- φόρου Περσόνας Ῥωµαίου), and also the commissioner (πρὸς χάριν τοῦ … ἐπισκόπου τῆς Καφὰς). Since the end colophon mentions explicitly only the Erotemata as copied by Cristoforo, one can argue that his signature applies only to the second part of the codex.58 Second, there are undeniable differences be- tween the two parts in the style of handwriting. Gamillscheg, who considers the scribe of the dictionary as not identical with Cristoforo Persona (and even takes the Latin copyist as a third person), lists seven characteristic letters and ten ligatures as differentiating traits.59 Without entering into the details (I intend to deal with the question elsewhere), I would make only three comments, each of a palaeographical nature. First, it has not been pointed out in previous discussions that there are also similar traits that mark both hands of the Vienna manuscript and these sim- ilarities are much more striking and numerous than the differences. Second, of the characteristic features listed by 57 See Bick, Die Schreiber 56; H. Hunger, Katalog der griechischen Handschriften der Österreichischen Nationalbibliothek IV (Vienna 1994) 89–90; E. Gamillscheg, “Übersetzer und Kopist – Beobachtungen zum Werk des Christophoros Persona,” BBGG 51 (1997) 233–242; Ötvös, “Janus Pannonius” 80–81, takes a more cautious stance. 58 Ötvös, “Janus Pannonius” 80–81. 59 Gamillscheg, BBGG 51 (1997) 235. 824 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Gamillscheg, each can be observed in the first copyist’s script as well. The difference lies only in quantity and scale. The scribe of the dictionary uses ligatures far more rarely, and he is also more sparing with the different versions of a given letter. This difference, however, can be easily explained by text-types. A dictionary is expected to provide an easily readable and un- ambiguously identifiable text with separate, distinct, and more uniform letters, with fewer ligatures and abbreviations. It is primarily meant for a beginner or advanced language learner who would want unfamiliar words to be clearly spelled and easily recognizable. In contrast, the potential reader of a gram- mar book can be assumed to have a greater facility in reading, and the text itself consists mainly of continuous grammatical explanations or illustrative paradigms and examples, with sev- eral recurring grammatical terms and notions—all these features allow, or even prompt, a more cursive style with frequent ligatures, contractions, and abbreviations. Finally, the lack of reference to the dictionary in the end colophon, ad- mittedly, does speak against Cristoforo’s authorship, but this argumentum ex silentio is far from being a conclusive proof. There may be several other reasons why he failed to mention the dictionary in the end colophon (he may have simply forgotten it, or, alternatively, he might have assumed that the reader would automatically recognize his hand in the first part of the codex as well).60 In the light, however, of the new evidence provided by Benedetto Bursa’s correspondence and the close stemmatic relationship between the two dictionaries, the question of attribution seems to be clearly decided in favour of Cristoforo Persona. In other words, both the biographical data and the results of textual critical analysis support the attribution and 60 Thus e.g. H. J. Hermann took the signature as holding good for the whole manuscript, “Die Handschriften und Inkunabeln der italienischen Renaissance,” in J. Schlosser and H. J. Hermann, Beschreibendes Verzeichnis der illuminierten Handschriften in Österreich VIII.6.3 Mittelitalien: Toskana, Umbrien, Rom (Leipzig 1932) 135–136. GÁBOR BOLONYAI 825 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 considerably weaken any doubt based exclusively on palaeo- graphic observations. 8. Another copy of Pseudo-Cyril by Cristoforo Persona Cristoforo’s role in the transmission history of pseudo-Cyril’s dictionary, however, does not end here: he made another copy of it. This exemplar is now in the Biblioteca Laurenziana in Florence (Laur.Edili 219). The scribe of the manuscript has been considered unknown until now, but if we juxtapose it to the Vienna copy, a single look is enough to recognize that they were written by the same hand ( figs. 9 and 10). This impression can be strengthened again by a clarification of the position of the manuscripts in the textual tradition. If we collate the Florentine manuscript (henceforth Ed) with Vienna Suppl.gr. 47 (W2) and Benedetto’s exemplar (Es), Ed is often found to be in agreement with both Es and W2, by sharing significant textual corruptions and secondary readings; at the same time it also shares conjunctive errors and innovations only with W2. To show the close kinship of all three codices, the passage in letter χ dealt with above can be cited as a most decisive piece of evidence. The same error of transposition is also present in Ed. In it as well, ‘section C’ from χυδαῖος to χωροµετρῶ comes directly after χρίστης (A) instead of χρίω (B), though B (χρίω to χρώς) has been further corrupted, and in fact it is completely missing from the manuscript. In other words, in Ed we have a sequence of ACD, as opposed to ACBD in Es and W2. From the complete omission of B another inference concern- ing Sagundino’s exemplar (henceforth σ ) can be made. It is an obvious question how section B (and exactly B, not more and not less) could have disappeared if B and C were the opposite sides of the same page. (We should bear in mind that B and C begin and end in mid pages in Ed, too, so the disappearance of B was not the result of some physical damage to Ed, but must have been brought about by an inherited error.) A possible ex- planation might be that originally, i.e. in σ , both B and C had text on their rectos but were blank on their versos. 826 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Figure 9: ÖNB Suppl.Gr. 47, 26r Figure 10: Laur.Edil. 219, 70v GÁBOR BOLONYAI 827 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 In that case, it is conceivable that either of two types of textual corruption took place: they either changed places (as in Es and W2), or a single ‘side’ (actually, a page) fell out (as in Ed). It seems likely therefore that in their common ancestor σ it was not two sides of a page that changed places, but two separate one-sided pages. Such a spacious and open arrange- ment is not without parallels among contemporary work-in- progress dictionaries.61 Sagundino must have realized right at the beginning when he started to transcribe the ancient vocabulary that it needed massive improving and enlarging. Accordingly, during copying, he himself supplied several hundred new entries, explicative synonyms, Latin equivalents, and illustrative quotations to the original text of Pseudo-Cyril. It is easily imaginable that considering the conspicuous de- ficiencies of the dictionary he left blank each back side for possible later additions. We can draw further inferences concerning Sagundino’s exemplar. If it was two pages that changed places in it, then first, they were either bifolia (more probably), or single sheets (less likely); second, their misplacement was most probably the bookbinder’s fault. We have good reasons to assume this. Since in Benedetto’s copy the warning note refers to the misplaced passages as pages (καταβατόν) and yet they do not correspond to pages, we can be fairly sure that the misplacement originally occurred in its master copy, i.e. in Sagundino’s exemplar and the warning note itself was made by Sagundino. This implies a situation in which he was able to realize the mistake, but unable to correct it—a fait accompli that was created, most probably, by the pages having already been bound together. Anyway, Sagundino, instead of replacing the pages in their correct order, only added a warning comment on the margin. The dependence of the three manuscripts on a common hyparchetype can be demonstrated with several other linking 61 E.g. Napoli BN II D 34, or BL Harley 6313, for which see Thiermann, Das Wörterbuch 22–24. 828 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 errors and secondary readings, two of which deserve special attention. They show a similar error of textual transposition and a similar handling of the mistake: the replacement of two passages consisting of ca. 32 lines, and with both a scribal note added in Benedetto’s copy to warn the reader of the confusion. The first occurs after the lemma διόρθωσις, which in Benedetto’s copy (138r) is followed by δίυγρος, instead of διορθοῦµαι (in fact διόρθωσις should precede διορθοῦµαι, but this is their original order in H). From δίυγρος on, 32 entries follow in alphabetical order (this passage must have covered one page in σ ) up to δογµατίζω. After δογµατίζω we jump back to διορθοῦµαι, after which a passage containing 32 en- tries again starts, until we reach δίσωµος. After δίσωµος comes δογµατικός, the word which follows δογµατίζω in H. As a result, we have an ACBD sequence again, and since the boundaries of the misplaced passages do not correspond to those of the pages in Es, we can draw the same inference that two pages, each of 32 lines, must have changed places in σ . As to Cristoforo’s two copies, both show a similar error of misplacement, but their texts have also further corruptions, more particularly omissions of entries. Although the passages, because of the omissions, are shorter, they follow each other in the order ACBD. In W2 5 items are missing from ‘section C’ and 4 from B, in Ed 4 entries from B and 3 from C, reducing the length of the original pages from 32 lines to 27 + 28 and to 28 + 29, respectively. It is significant that most of the omissions are shared by the two manuscripts, pointing to a common sub- ancestor where the items in questions had already been left out.62 Considering that both W2 and Ed were commissed by others (a bishop of Caffa63 and Bartolomeo Lapacci, who was bishop of Argos between 1434 and 1439 and bishop of Corone 62 E.g. the entry διούγκιον sextans is left out in both, or the items δίχρους bicolor and δίχρωµος bicolor contract into one, δίχρους δίχρωµος bicolor. 63 See the colophon on 101r; Bick, Die Schreiber 56, identifies him with Jacopo Campora. GÁBOR BOLONYAI 829 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 between 1449 and 1460),64 one may think of a copy which Cristoforo made for himself as a base text for producing further exemplars. This assumption involves naturally that Cristoforo’s extant copies were not made directly from σ or Es. The second displacement occurs in the section letter ο. In Benedetto’s copy on the lower part of f. 221v a warning again draws the attention of the prospective user to it. This time the lemma ὁπλίτης is followed by ὀπτός, instead of ὁπλοδιδά- σκαλος, then comes a stretch of 32 entries (counted inclusively) ending with ὀρθότατος. Then comes ὁπλοδιδάσκαλος (a jump back in the alphabetical order) and 33 other entries (inclusively) until ὀπτόν. It is followed by ὀρθοτοµῶ, the word which should follow ὀρθότατος if alphabetical order were kept. Thus, the sequence is again ACBD, only the affected lines are slightly longer: while C has 32 lines (entries), B has 33. The mistake appears, again, in both of Cristoforo’s copies, in each in an even more corrupted form. In W2 page C is reduced to 7 entries, page B to 14. The shrinkage is even greater in Ed, where only 5 entries remain from C and 13 from B. The greatest part of the omitted entries are missing in both codices, so may serve again as conjunctive errors. There are dozens more agreements that can prove the com- mon origin of the three manuscripts, but a further discussion of such examples seems unnecessary. Instead, I turn to cases that may help define the genealogical relationship of the three man- uscripts to one another more exactly. In three characteristic passages W2 and Ed agree in error against Es. 1) ΒΟΤΑΝΗ HERBA GRAMEN MEDICAMENTUM H βοτάνη herba bae. hoc gramen minis. medicamentum Es | βοτάνυ herba W2 Ed Both W2 and Ed omit the last two Latin equivalents and com- 64 For his knowledge of Greek see T. Käppel, “Bartolomeo Lapacci de’ Rimbertini (1402–1466): vescovo, legato pontificio, scrittore,” Archivum fratrum Praedicatorum 9 (1939) 9. Lapacci, who belonged to the inner circle of Eugene IV, served several times as the pope’s emissary in Greece. 830 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 mit an orthographical mistake, υ in place of η, the pronun- ciation being identical. While the first omission is clearly to be seen as a deliberate reduction (there is a general tendency in Cristoforo’s copies to keep the number of equivalents at min- imum), the second fault seems unintentional. It happens quite often that whole entries are left out in both manuscripts: 2) ΒΟΤΑΝΗΣ ΓΕΝΟΣ CALTA H βοτάνης γένος calta Es | om. W2 Ed 3) ΒΟΤΑΝΗΗΤΟΙΣΤΟΙΧΟΙΣΚΑΙΦΛΟΙΟΙΣΣΥΝΗΜΜΕΝΗ MUSCUS H βοτάνη ἡ τοῖς τοίχοις καὶ φλοιοῖς συνηµµένη hic muscus. singu- lariter tantum declinabitur Es | om. W2 Ed The examples cited so far show only the agreement of W2 and Ed in error against Es, but do not help verify the as- sumption, touched upon briefly above, that they inherit their corrupted readings from a common hyparchetype; or we should assume, alternatively, that only one of them had access to σ and the other is a direct apograph of it. Other evidence unambiguously supports the first alternative. It is a recurring case that, given two or more Latin equivalents, W2 and Ed decide in a different and opposite way, by choosing the one(s) that the other manuscript omits. A typical example: ΚΕΝΤΡΙΖΩ STIMULO PUNGO H κεντρίζω stimulo las. pungo, pungis, pupugi Es | κεντρίζω pungo W2 | κεντρίζω stimulo Ed Of the two Latin equivalents W2 chooses only pungo, whereas Ed only stimulo. It is quite clear that neither could W2 have had access to pungo from Ed, nor Ed to stimulo from W2. Both de- pend on a copy containing both stimulo and pungo (henceforth this copy, Cristoforo’s postulated personal exemplar, will be designated χ ). Another entry provides a similar, only slightly different case: ΕΙΛΩ PLICO UOLUO H εἰλῶ τὸ συστρέφω plico uoluo Es | εἰλῶ uoluo plico W2 | εἰλῶ τὸ συστρέφω uoluo Ed In Es a Greek synonym (τὸ συστρέφω) is added, obviously meant as an explanation of the rather uncommon verb εἰλῶ. GÁBOR BOLONYAI 831 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 The explanatory verb turns up in Ed but not in W2. The op- posite happens to plico, one of the two Latin equivalents: taken up by W2, but omitted by Ed. The mutually opposite choices point again to a common ancestor and exclude the possibility that either of the two manuscripts can be derived from the other. Here, however, we also have corroboration that Ed de- pends on either Es or σ , since τὸ συστρέφω is a secondary element that cannot be found in H. So far we have seen the following picture emerging: Es, W2, and Ed all depend on σ , Es directly, W2 and Ed indirectly through their common hyparchetype χ . One question has re- mained open: was χ copied from σ or Es? In other words, did Cristoforo transcribe his own (now lost) copy from Benedetto’s exemplar or from the one belonging to their common Greek teacher, Niccolò Sagundino? Because of the significant difference between the ways Bene- detto and Cristoforo carried out their work, it is not easy to find conclusive evidence. Benedetto seems to have transcribed all the material he found in Sagundino’s exemplar: the original entries in their entirety, Sagundino’s supplementary explana- tions, corrections, enlargements, new items, marginal notes, and illustrative quotations, and he even added his own amend- ments and comments as well. Moreover, his text reveals extra- ordinary care in observing grammatical and orthographical rules, especially concerning diacritical marks. He was clearly keen on making the dictionary—for his own personal use—as informative and reliable as possible. Cristoforo’s attitude was sharply different. He does not seem to have been motivated by an ambition to produce copies as precise, faithful, and rich as possible—not to mention, to improve them. He made tran- scriptions for clients, and his concern about the usefulness and philological value of the copies was quite limited. He left out not only all the marginal notes and quotations but also most of the Greek synonyms, which were added most probably by Sagundino. He systematically reduced the number of Latin equivalents to one or two—a revealing sign that his focus was on fulfilling his task as economically as possible. Even his at- 832 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 tention was not always concentrated on his job and he omitted —sometimes, as it seems, accidentally, sometimes deliberately —several dozen entries (overall, several hundreds). He proved to be especially careless while copying Ed, where he occa- sionally left out several items in a row, even up to five or six.65 Thus, it is extremely difficult to find a correct reading pre- served by either W2 or Ed against Es, or an innovation of σ taken over only by any one of them, or any reading that could prove their independent access to σ . Furthermore, even the few examples that can be cited to this effect are relatively weak. Take for example the entry ἐγκατασκήπτω irruo. It is inserted after ἐγκαινίζω innovo in both W2 and Ed, but cannot be found in either H or Es. Since there is no trace of any other inno- vation made by W2 or Ed on its own initiative (at least, I am aware of none), it seems probable, or at least imaginable, that the interpolated entry is inherited from σ . If the assumption holds, then we have an exceptional case in which an item escaped Benedetto. The same applies to scribal errors. Naturally enough, even the careful Benedetto dormitat now and then and commits mis- takes, mostly slips of the pen, but they are usually insignificant and obvious errors that are easy to correct. For example, he miswrites ἐκβολῶ exclamo instead of ἐκβοῶ exclamo, but evi- dently it would not be a safe inference from this error that Cristoforo took his correct reading from σ , as it seems equally possible that seeing the non-existent Greek verbal form (ἐκβολῶ) and the Latin translation (exclamo) of a very similar genuine Greek verb, he was able to recognize the slip of the pen and correct it himself. 65 Profit-orientation manifests itself quite noticeably in Cristoforo Per- sona’s scribal activity in general. For instance, he dedicated six deluxe copies of his Agathon translation, each to a wealthy and illustrious pur- chaser such as King Ferdinand II of Aragon, King Matthias Corvinus, his wife Beatrice of Aragon, Pope Sixtus IV, Lorenzo de’ Medici. Gamillscheg, BBGG 51 (1997) 241, suggests that the profit gained thereby might have been intended to finance a printed edition of the text. GÁBOR BOLONYAI 833 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 Of slightly more significance is the following disagreement between χ (= W2 and Ed) and Es: ἐναβρύνοµαι καὶ µεγαλοφρονῶ καὶ καυχῶµαι magnum sapio. glorior. amplus fio Es ἐναβρύνοµαι τὸ µεγαλοφρονοῦµαι magnum sapio. glorior Ed ἐναβρύνοµαι τὸ µεγαλοφρονοῦµαι magnum sapio W2 The whole entry is a later addition, in all likelihood, again by Sagundino. Both W2 and Ed give, as usual, a shortened version of it, as compared with Es, yet there is one point where they may preserve a better reading. This is the first word after the lemma ἐναβρύνοµαι, where one expects the article τό (as in W2 and Ed) rather than the conjunction καί (as in Es). The function of this word obviously should be to introduce the ex- plicative synonym(s)—a role that is always fulfilled in Es by the article and never by the connective. It seems much more likely, therefore, that καί is an error of visual origin, due to an anticipation of the second καί which connects the explicative synonyms. The question, however, whether χ corrected καί by changing it to τό, or simply copied it from σ , is not easy to decide, but probably correction would have required more alertness from Cristoforo than what he normally showed. Given this situation, we are compelled to turn to indirect evidence. One relevant passage that may serve has already been touched upon: the complete omission of 31 consecutive items in Ed, starting with χρίω and ending with χρώς (a supposed page B in σ ). As we saw, both the first and last entries of the section fall at mid page in Es, and there is no sign whatsoever in the manuscript that might induce a scribe to leave out the subsequent 31 entries. If we assume that Cristo- foro made his copy χ from Es, it is wholly incomprehensible why he decided at the lemma χρίω, when he later made his Florentine copy (Ed), that he would stop copying the next 31 items, and after this section why he made another decision at the lemma χρώς that from that point on he would resume copying the text. A similar independence of χ from Es is indicated by the different ways the three manuscripts handle the lacuna be- 834 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 tween ΑΛΙΞ ALLEX and ΑΝΔΡΑΧΝΗ PORCACLA in H. Es fills the gap with 122 entries (presumably taken from a humanist vocabulary): 22 words beginning with αλ-, 46 with αµ-, 54 with αν-.66 What is worth stressing, however, is that the interpolated words in Es cannot be distinguished from the surrounding original ones. They simply do not stand out at all: they start and end at mid pages and are written in the same size and style as the rest. The only sign that may cause some suspicion is that their alphabetization is not without minor faults, but this is not a particularly striking feature, because similar imperfections in the arrangement occur elsewhere, too. How do W2 an Ed deal with the lacuna, then? The Floren- tine manuscript keeps the whole hiatus intact without putting in a single word, while the Vienna codex introduces two sequences of entries that are identical with two parts of the interpolation in Es, specifically all the 22 words beginning with αλ-, none of those beginning with αµ-, and the first ten be- ginning with αν-, in sum, 32 out of 122 items. In other words, he takes over first one passage consisting of 22 items, then omits a stretch of 46 entries, then takes another stretch of words consisting of 10 from the interpolated material, and omits again a run of 44 items. Both decisions are inexplicable again on the assumption that they (and their common ancestor χ ) were copied directly from Es. As we saw, there is nothing in the text of the Escorial man- uscript that may have reasonably prompted a scribe to omit, on one occasion, every interpolated word as in the case of Ed, or to select, on another occasion, just those bits that were selected in the case of W2. We should, therefore, posit a master manu- script in which it was manifest where the interpolated entries started and where they ended, and the fact of interpolation was also evident; on the other hand, any further attempt to re- construct the arrangement of this lost copy even more closely seems already rather hazardous. Thus, the three different ways 66 Note that the number 122 strongly suggests again that they may have filled 4 pages in σ , two pages consisting of 31 lines, two of 30. GÁBOR BOLONYAI 835 ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 of handling the lacuna suggest a common ancestor to which Es and χ had independent access. Es opted for incorporating the whole interpolation into the vocabulary, whereas Ed opted for a complete omission, W2 for a partial one. The relation of all the manuscripts can be summarized in the following stemma: H | σ (N. Sagundino’s lost copy) Es χ (B. Bursa’s copy) (C. Persona’s own lost copy) | W1 (J. Pannonius’ copy) W2 Ed (Persona’s copy made (Persona’s copy made for Bishop of Caffa) for Lapacci) 9. After Benedetto’s death Finally, the fate of Benedictus’ dictionary after his death. As is testified by the next possessor’s ex libris, the book was owned by Diego Hurtado de Mendoza.67 It is very likely that he pur- chased it during his Italian sojourn when he was the Spanish ambassador in Venice (1539–1547). Mendoza systematically searched for ancient manuscripts, especially Greek ones, and in his will, compelled by the king, he donated his collection to the Escorial Library. Concerning the acquisition, one detail may merit our attention. Mendoza also came into possession of a book that was copied by Benedetto’s elder brother Sebastiano. The manuscript (T. III. 19), which contains Cicero’s De officiis, 67 P. A. Revilla, Catálogo de los Códices Griegos de la Biblioteca de El Escorial I (Madrid 1936) 253; Ötvös, “Janus Pannonius” 96. 836 BENEDICTUS AND HIS GREEK-LATIN DICTIONARY ————— Greek, Roman, and Byzantine Studies 57 (2017) 792–836 was made in 1424 in Venice and is not without palaeographi- cal interest. Zamponi, acclaiming him “un copista straordi- nario,” assigns an important place to Sebastiano Borsa (as his name appears in the colophon)68 in the development of humanist script. He considers him an important representative of the Venetian experimental school that combined the chan- cery script with antique elements, a development comparable to the Florentine innovations.69 It is conceivable that the two manuscripts of the Bursa brothers were bought together by the great Spanish book collector, who may also be credited with a second layer70 of marginal notes, written by an evidently later hand, found in Benedetto’s dictionary.71 March, 2017 Greek Department Eötvös Loránd University Múzeum krt.4/F Budapest, Hungary H-1088 bolonyai.gabor@btk.elte.hu 68 A. Hobson, Renaissance Book Collecting: Jean Grolier and Diego Hurtado de Mendoza (Cambridge 1999) 88. 69 S. Zamponi, “La scrittura umanistica,” Archiv für Diplomatik 50 (2004) 478–479. 70 Thiermann, in Les manuscrits 659. 71 For the illutrations of Escorial Σ I.12 I am grateful to the Real Biblioteca Monasterio del Escorial, for those of Edili 219 to the Biblioteca Medicea Laurenziana, and for those of ÖNB Lat. 3370 and Suppl.gr. 47 to the Österreichische Nationalbibliothek. This study is part of a project on the reception of Greek culture in the Renaissance supported by the Hungarian Scientific Research Fund - OTKA (project number: K 112283).