Mari3 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 2018 Tommaso Mari The Latin Translations of the Acts of the Council of Chalcedon Tommaso Mari HE COUNCIL OF CHALCEDON (A.D. 451) was a turning point in the history of the church. It was summoned by Marcian, the Eastern Roman Emperor, and its main goal was to undo the effects of the Council of Ephesus II (449), which had been held under Theodosius II. Hundreds of bishops and lesser clergy from the eastern part of the Roman Empire attended the council; they were joined by very few clergymen from the western part of the Roman Empire, including three delegates of Pope Leo I (Pascasinus bishop of Lilybaeum, Lu- censius bishop of Asculum, and the Roman presbyter Boniface).1 Of the Council of Chalcedon we possess the Acts, which consist of the minutes of the Council, alongside a selection of letters and related documents. The medieval manuscript tra- dition has preserved a Greek version of the Acts and three Latin translations dating to the sixth century. All these have been published in the 1930s by E. Schwartz in the Acta Conciliorum Oecumenicorum.2 It is well known that translation activity from Greek into Latin flourished in Late Antiquity, particularly as far 1 In writing Pascasinus and Lucensius I follow the Latin primary sources (in the Greek sources we read similarly Πασκασῖνος and Λουκίνσιος); on the other hand, many modern scholars write Paschasinus and Lucentius. 2 E. Schwartz, Acta conciliorum oecumenicorum II Concilium universale Chalcedonense (Berlin/Leipzig 1933–1937). Volume II.1 contains the Greek Acts, II.3 the Latin Acts. I cite the text of the Acts according to the page and line number of Schwartz’s edition. An English translation of the Acts is in R. Price and M. Gaddis, The Acts of the Council of Chalcedon (Liverpool 2005). In this paper, all translations are my own. T TOMMASO MARI 127 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 as religious texts are concerned.3 In this paper, I am going to offer an overview of the sixth-century Latin translations of the Acts of the Council of Chalcedon; in particular, I am going to focus on the characteristics of the translations (their quality and reliability), the work of the translators (their approach to the source text and their translation techniques), and the historical context in which these translations were produced (mainly insofar as it influenced the work of the translators). Special attention will be paid to the statements of the Roman delegates and the way these were translated. The Acts: language and textual history The Council was first and foremost a Greek matter.4 Its location and attendees made it so. From both the Greek and Latin versions of the Acts it appears clearly that the norm was for most attendees to speak Greek. The utterances of the Roman delegates are explicitly marked as having been expressed in Latin and translated into Greek by interpreters;5 on the other 3 For a brief overview of translations from Greek into Latin in Late Antiquity see A. Rigolio, “Translation of Greek Texts in Late Antiquity,” in G. K. Giannakis et al., Encyclopedia of Ancient Greek Language and Linguistics (Leiden/Boston 2014) III 436–441, esp. 436–437. On this topic, a fundamen- tal work is the monograph of H. Marti, Übersetzer der Augustin-Zeit. Interpretation von Selbstzeugnissen (Munich 1974). A special focus on religious texts is in C. Rapp, “Hagiography and Monastic Literature between Greek East and Latin West in Late Antiquity,” Cristianità d’Occidente e cristianità d’Oriente (Spoleto 2004) 1221–1281. More recently, the topic of translations in connection with multilingual competence has been explored by T. Denecker, Ideas on Language in Early Latin Christianity (Leiden/Boston 2017) 158–169. 4 This aspect, with regard to the Councils that took place under the reign of Theodosius II, is discussed in F. Millar, A Greek Roman Empire. Power and Belief under Theodosius II (Berkeley/London 2006). 5 Not all of their statements are said to have been uttered in Latin and translated into Greek. For example, the very first statement of the chief of the Roman delegation Pascasinus, par. 4–5 of session 1 (ACO II.1 65.15–22), is marked as having been made in Latin, but his next few ones (session 1 par. 7, 10) and those of the other Roman delegate Lucensius (session 1 par. 9, 12) are not. This hardly means that the Roman delegates switched from Latin to Greek; it stands to reason that sometimes the notaries thought it sufficient to 128 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 hand, no linguistic remark is made on the myriad statements of the bishops who came from the Greek-speaking world. As exceptions are more likely to be signalled than rules, one can reasonably conclude that Latin was the exception and Greek was the rule. It is not entirely clear how the minutes of the Council were produced.6 We know that the proceedings were ready right after the end of the gathering; the Roman delegates returned to Rome with some documents and the rest were sent to Pope Leo I by Anatolius, archbishop of Constantinople.7 We do not have this version of the proceedings, but we know from Pope Leo’s re- action that it was written (for the most part) in Greek: for in March 453 he wrote to his representative in Constantinople, Julian of Cos, lamenting that he did not sufficiently understand the content of the proceedings because of the linguistic barrier; therefore, he asked Julian to provide him with a Latin translation of the full proceedings.8 For the rest, the only Latin materials relating to Chalcedon that Leo had access to were some of the Roman delegates’ Latin statements, the Emperor Marcian’s Latin speech in the sixth session, and a few translated texts and short summaries of other sessions.9 There is no evidence that Pope Leo’s request of a full translation was ever satisfied during his lifetime. The official version of the Acts was published between 454 and 455 in Constantinople.10 This version included a selection of specify the language of the statement at the earliest opportunity and simply took it for granted later on. 6 Price and Gaddis, The Acts I 74–78. 7 Letter of Anatolius of Constantinople to Pope Leo, December 451 (ACO II.1 448.24–28). 8 Letter 113 of Pope Leo to Julian of Cos, 11 March 453 (ACO II.4 66.35—67.6, ep. 60). 9 Cf. ACO II.3.1 V. Some of these texts are transmitted in the Collectio Vaticana (ACO II.2.2), a collection of materials from Chalcedon assembled shortly after the Council for Pope Leo (cf. ACO II.2.2 X–XIV). 10 Price and Gaddis, The Acts I 78–83. TOMMASO MARI 129 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 letters about the Council written by significant personalities of the time: the Pope, the Eastern and Western Emperors and their associates, etc. It is believed that the general orientation of this edition was to undermine the role of the Roman see to the ad- vantage of the see of Constantinople. This determined a series of editorial interventions that obscured and distorted to some degree the events of the Council of Chalcedon.11 It is important to note that this edition must have reflected quite faithfully the situation of ‘unbalanced’ bilingualism of the Council; aside from the Greek statements and frame, it contained the Latin state- ments of the Roman delegates alongside their translation into Greek.12 The Greek Acts as we have them now are the result of a later revision in which the Latin materials were excised. However, they still retain evidence that they once included the Latin statements. Here follows one of several examples (ACO II.1 65.15–19, session 1 par. 4–5): *** τούτων τοίνυν ῾Ελληνιστὶ ἑρμηνευθέντων διὰ Βερονικιανοῦ τοῦ καθωσιωμένου σηκρεταρίου τοῦ θείου κονσιστορίου Πασκα- σῖνος ὁ εὐλαβέστατος ἐπίσκοπος καὶ φύλαξ τοῦ ἀποστολικοῦ θρόνου, ὡς ἔστη ἐν μέσωι μετὰ καἱ τῶν συνελθόντων αὐτῶι, ἔφη· Τοῦ μακαριωτάτου καὶ ἀποστολικοῦ ἐπισκόπου τῆς Ῥωμαίων πόλεως […]. These words having been translated into Greek by Veronicianus, the hallowed secretary of the divine consistory, Pascasinus, the most devout bishop and guardian of the apostolic see, took his stand in the centre together with his companions and said: “From the most blessed and apostolic bishop of the city of Rome […].” If we focus on the first word of the Greek text, τούτων, it is evi- dent that this lacks an antecedent (whence the editor Schwartz posited a lacuna, indicated by the asterisks, just before the pro- noun). The antecedent consisted of the statement of bishop 11 Price and Gaddis, The Acts I 80–82. See also R. Price, “Truth, Omission, and Fiction in the Acts of Chalcedon,” in R. Price and M. Whitby (eds.), Chalcedon in Context: Church Councils 400–700 (Liverpool 2009) 92–106. 12 E. Schwartz, “Zweisprachigkeit in den Konzilsakten,” Philologus 88 (1933) 245–253, esp. 247–248. 130 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 Pascasinus in Latin. The same situation is observed time and again.13 We shall come back to this below. The sixth-century Latin translations As far as we know, the first Latin translations of the full Acts of the Council of Chalcedon were produced in the mid-sixth century in Constantinople.14 This was the time of the so-called Three-Chapter Controversy.15 The Council of Chalcedon had not been accepted by all churches. The Emperor Justinian aimed to achieve unity between Chalcedonians and Non-Chal- cedonians. Among the things he did to reconcile them was to condemn in 543/4 some writings of Theodore of Mopsuestia, Theodoret of Cyrrhus, and Ibas of Edessa (the “Three Chap- ters”). That angered Chalcedonians, for these figures had been accepted as orthodox by the Council of Chalcedon. Pro- Chalcedonian resistance was strong among bishops from the Latin-speaking world. The debate about Chalcedon and the fact that most westerners could not easily read Greek made it urgent to produce Latin translations. For example, the African bishop Facundus of Hermiane, one of the leaders of the Chalcedonian resistance in Constantinople, wrote the Pro defensione trium 13 ACO II.1 114.19–24 and 28–31 (session 1 par. 273 and 274–275); 119.33—120.4 (session 1 par. 330, a citation from the Acts of an earlier Council); 120.22–24 (session 1 par. 336); 204.22–33 (session 2 par. 4); 319.24–28 (session 5 par. 9); 364.26–36 (session 8 par. 6–7); 467.30–36 (ses- sion on Photius and Eustathius par. 49–50). A couple of times the sentence referring to the translation follows the translated statement: 289.20–33 (session 4 par. 6–7) and 308.11–16 (session 4 par. 38–39). In one case the translated statement is simply introduced by δι᾽ ἑρμηνέως εἶπεν “he said through an interpreter” (206.38, session 2 par. 18). The same also applies to translated letters, in which case one finds the expression Τῶν Ῥωμαικῶν τούτων ἡ ἑρμηνεία ἐστὶ τὰ ἐπαγόμενα Ἑλληνιστί: ACO II.1 3.1; 4.31; 5.1, 29; 6.19; 8.4; 9.14. 14 Price and Gaddis, The Acts I 83–85. 15 H. Chadwick, The Church in Ancient Society. From Galilee to Gregory the Great (Oxford 2001) 612–627; A. Grillmeier with T. Hainthaler, Jesus der Christus im Glauben der Kirche II.2 Die Kirche von Konstantinopel im 6. Jahrhundert (Freiburg/ Basel/Vienna 2004) 431–484. TOMMASO MARI 131 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 capitulorum in 546–548 and filled it with Latin citations from the Acts of the Council of Chalcedon.16 Pope Vigilius too, who resided in Constantinople against his will from 547 to 555, cited the Acts in his writings, although he ignored Greek. In a letter that he wrote in 553, Vigilius referred to his men qui eiusdem linguae [i.e. Greek] uidentur habere notitiam as those thanks to whom he could read the Acts of the Council of Chalcedon.17 Schwartz was of the opinion that pro-Chalcedonians in Constantinople translated more and more of the Acts according to their needs and eventually, after the Council of Constantinople in 553, completed the translation which we now call the Versio antiqua (Φa in Schwartz’s edition).18 Popes of the Early Middle Ages knew this version. The Versio antiqua was soon revised and enriched with the translation of some pre- and post-conciliar letters included in the Greek Acts. This edition is known as the Versio antiqua correcta (Φc in Schwartz). The identity of the editors is unknown to us. Their motivations, however, seemed to be quite transparent to Schwartz: for example, the displacement of the canons after the last session instead of the sixth, where they actually belonged, has been interpreted as an attempt to confer greater authority on those sessions that the editors relied upon for their defense of the “Three Chapters.”19 The third edition is known as the Versio Rustici (Φr in Schwartz).20 This is the one we know the most about, for it is a very ‘personal’ edition, its editor having left substantial traces of his work in it. This was the Roman Deacon Rusticus, the nephew of Pope Vigilius and an important theologian of the mid-sixth century; he was a strenuous defender of the “Three 16 ACO II.3.1 VI–VII. 17 Collectio Avellana 83 (p.236 Guenther). 18 ACO II.3.1 VI–VII. 19 ACO II.3.3 XI. 20 The Versio Rustici is the basis of Schwartz’s edition of the Latin Acts. The differences from the Versio antiqua and Versio antiqua correcta are recorded in his apparatus. 132 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 Chapters” and the author of the theological dialogue Contra acephalos.21 After Pope Vigilius joined Justinian in the condem- nation of the “Three Chapters,” Rusticus turned against Vigil- ius. This cost him excommunication and exile to Egypt. Before the death of Justinian (565) he returned to Constantinople. In the monastery of the Acoemete monks, he found the tranquillity and materials he needed to carry on his battle. There he took up the revision of the Latin translation of the Acts of Ephesus I (431) and of Chalcedon, which he included in his most important work, the Synodicon. From his subscriptiones we know that he worked at the translation of the Acts of Chalcedon from Feb- ruary 564 to March 565, and gave it the last touch in April 566.22 We shall look at the details of this edition below. However, I should anticipate that Rusticus revised the Versio antiqua correcta against Greek manuscripts of the Acts, in particular one codex that he found in the Acoemete monastery.23 Another interesting aspect of Rusticus’ edition is the scholarly apparatus that ac- companies it in some manuscripts, the adnotationes Rustici, which allow us an insight into Rusticus’ philological and theological work.24 In what follows, I shall compare the three Latin translations of the Acts with each other and with the Greek Acts (Γ). My focus is going to be on the process and results of the translation; I shall not concern myself with issues of content or structure, such as which materials appear across the different versions and how they are organized. However, it is crucial to keep in mind that 21 S. Petri, Rusticus Diaconus: Contra Acephalos (Turnhout 2013). 22 ACO II.3.3 XIII–XIV. 23 Cf. Rusticus’ inscriptio to the first session of the Acts (ACO II.3 27.2): RVSTICVS EX LATINIS ET GRAECIS EXEMPLIS MAXIME ACOEMIT(ENSIS) MONAST(ERII) EMENDAVI. Schwartz discusses this at ACO II.3.3 XIV–XVII. 24 In ACO II.3 Rusticus’ notes are reported in a dedicated section of the critical apparatus. They are also published in PL Suppl. 4 pp.546–597 as a running text. As always happens with marginalia, not all of Rusticus’ ad- notationes have been preserved by the manuscript tradition. See ACO II.3.3 XXI–XXIII for an overview of Rusticus’ adnotationes. TOMMASO MARI 133 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 the Latin translations are at times more reliable than the extant Greek version, not least because the Greek Acts underwent further revision after they were translated into Latin (probably in the seventh century).25 In particular, if one looks at the content, the Latin Acts include materials that have been excised from the Greek Acts.26 How to translate: uerbum e uerbo or sensus de sensu? As was mentioned above, translations played an important role in the Greco-Roman culture of Late Antiquity, and it will be helpful to say something about the translation techniques in this context.27 It is well known that the Romans distinguished between two kinds of translation: literal (uerbum e uerbo) and non- literal (sensus de sensu). The type of translation depended very much on the type of text to be translated. In his famous Letter 57, Jerome argued for translating uerbum e uerbo the Scripture and sensus de sensu all other texts. Administrative texts were generally the object of literal translations. That was a way for translators to safeguard the readers and themselves. This was generally the case of conciliar translations as well.28 Translation samples (1): the ‘narrative’ frame The first sample of translation is the very beginning of the Acts of the Council of Chalcedon. This is the introductory paragraph of the ‘narrative’ frame. I shall present the four versions side by side in columnar fashion (Γ = Greek version, ACO II.1 55.2–6; 25 Price and Gaddis, The Acts I 82–83; ACO II.1.1 VII–VIII and II.1.3 XXIX–XXX. 26 For example, the Latin translations include the minutes from the Council of Ephesus I that were read out at the first session of Chalcedon (session 1 par. 911–945); these have been excised from the extant Greek Acts as they were deemed superfluous. 27 See S. Brock, “Aspects of Translation Technique in Antiquity,” GRBS 20 (1979) 69–87, and B. Rochette, “Du grec au latin et du latin au grec. Les problèmes de la traduction dans l’antiquité gréco-latine,” Latomus 54 (1995) 245–261, esp. 249–250. 28 B. Rochette, Le latin dans le monde grec. Recherches sur la diffusion de la langue et des lettres latines dans le provinces hellénophones de l’Empire romain (Brussels 1997) 150–151. 134 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 Φa = Versio antiqua, Φc = Versio antiqua correcta, Φr = Versio Rustici, all at ACO II.3 p.27.5–8). Each line has one word or one trans- lation unit; where a version omits a word or translation unit that other versions have or moves it (relatively) far away, I mark the line with a hyphen; if the content of different versions does not match across lines I underline the text, even where it is a matter of word order; when in the discussion I mention the line number, I mean the line number of my layout and not of the editions. 5 10 15 20 25 30 Γ Ὑπατείαι τοῦ δεσπότου ἡμῶν Μαρκιανοῦ τοῦ αἰωνίου αὐγούστου καὶ τοῦ δηλωθησομένου τῆι πρὸ ὀκτὼ Εἰδῶν Ὀκτωβρίων – ἐν Χαλκηδόνι κατὰ κέλευσιν τοῦ θειοτάτου καὶ εὐσεβεστάτου δεσπότου ἡμῶν Μαρκιανοῦ τοῦ αἰωνίου αὐγούστου συνελθόντων ἐν τῆι ἁγιωτάτηι ἐκκλησίαι τῆς ἁγίας μάρτυρος Εὐφημίας τῶν ἐνδοξοτά- των ἀρχόντων […] Φa Consulatu domni nostri Marciani perpetui augusti et qui fuerit nuntiatus, sub die octauo Iduum Octobrium indictione IIII Calchedona secundum praeceptionem sacratissimi et piissimi domni nostri Marciani perpetui augusti congregatis in sancta ecclesia sanctae martyris Euphimiae gloriosissimis iudicibus […] Φc Consulatu domni nostri Marciani perpetui augusti et qui fuerit nuntiatus, sub die octauo Iduum Octobrium indictione IIII Calchedona secundum praeceptionem sacratissimi et piissimi domni nostri Marciani perpetui augusti congregatis in sancta ecclesia sanctae martyris Euphimiae gloriosissimis iudicibus […] Φr Consulatu domni nostri Marciani perpetui augusti et qui fuerit nuntiatus, sub die octauo Iduum Octobrium indictione IIII Calchedona secundum praeceptionem sacratissimi et piissimi domni nostri Marciani perpetui augusti congregatis in sancta ecclesia sanctae martyris Euphimiae gloriosissimis iudicibus […] TOMMASO MARI 135 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 The columnar layout makes it easy to see that this is by and large a literal translation. To almost each word in the Greek ver- sion there corresponds one in the Latin versions, and the word order is respected. In modern times, one often associates literal translation with the work of unskilful translators and unidiomatic results in the target language. That is clearly not the case here: the translator(s) of Φa did a good job of producing an idiomatic text and avoiding infelicities. For example, the formulaic τοῦ δηλωθησομένου (lines 7–8) has been translated with a periphrasis, qui fuerit nuntiatus, which is also a formula in Latin texts; the genitive absolute συνελθόντων … τῶν ἐνδοξοτάτων ἀρχόντων (24–32) is aptly rendered with an ablative absolute, congregatis … gloriosissimis iudicibus. In this case, the editor(s) of Φc and Rusticus were happy with the translation of Φa and did not feel like they had to make any changes to it. At 12 all three Latin translations indicate the indiction (indictione IIII), a piece of chronological information that is missing from the Greek version. It might be that this detail was not in the Greek original but was supplied by the editors of Φa. This may be confirmed by the fact that Rusticus notes in the margins of his text that he did not find this information in his Greek Acoemete codex (Acumit(ensis) non habet).29 The other note of Rusticus on this passage concerns the Latin translation sacratissimi (16), said of the Emperor Marcian; Rusticus evidently found θειοτάτου in the Acoemete manuscript, thought that diuinissimi was a more ac- curate translation than sacratissimi of the previous Latin versions, and wrote Acumit(ensis) diuinissimi in the margins. In both the case of indictione IIII and that of sacratissimi, Rusticus spotted a diver- gence between the Latin translation he was revising (the Versio antiqua correcta) and the Greek codex he was using for comparison (the Acoemete manuscript). However, he did not change the Latin translation based on the Greek (i.e. by deleting indictione IIII and by changing sacratissimi to diuinissimi), but only provided his readership with information about these uariae lectiones.30 In 29 Both this and the following adnotationes are preserved by the manuscripts Paris, BNF, Lat. 11611 and Lat. 1458 (both written in ninth-century France). 30 See ACO II.3.3 XXI for similar cases. 136 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 philological terms, he supplied a critical apparatus. Translation samples (2): the first statement of Dioscorus We have considered the case of a literal yet adequate transla- tion. The degree of faithfulness varies, and translations of oral statements are less literal than those of the ‘narrative’ frame, on account of the latter being rather formulaic. Let us consider, for example, the first statement of Dioscorus of Alexandria, one of the protagonists of the Council (session 1 par. 18). This time I present only three texts, for the Versio antiqua and Versio antiqua correcta do not show differences and can be presented together, as happens fairly often (Γ at ACO II.1 67.20–24, Φ at II.3 42.14– 18): 5 10 15 20 25 Γ Ἐκέλευσεν ὁ εὐσεβέστατος βασιλεὺς σύνοδον συγκροτηθῆναι, καὶ δὴ συγκεκρότηται κατὰ θεῖον νεῦμα τοῦ εὐσεβεστάτου βασιλέως ἡμῶν. – περὶ δὲ τῶν πεπραγμένων διὰ τὸν γενόμενον ἐπίσκοπον τῆς Κωνσταντινουπόλεως ἁγίας ἐκκλησίας Φλαβιανὸν ὑπομνήματά εἰσι πεπραγμένα ἐν τῆι ἁγίαι συνόδωι καὶ Φac Praecepit piissimus imperator synodum celebrari, quae – secundum sacrum nutum – – – congregata est. de his autem quae gesta sunt propter Constantinopolitanae sanctae ecclesiae quondam episcopum Flauianum, monumenta sunt acta in sancto concilio et Φr Praecepit piissimus imperator synodum celebrari, quae – secundum sacrum nutum piissimi imperatoris nostri congregata est. de his autem quae gesta sunt propter Constantinopolitanae sanctae ecclesiae quondam episcopum Flauianum, monumenta sunt acta in sancto concilio et TOMMASO MARI 137 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 30 παρακαλῶ ταῦτα ἀναγνωσθῆναι. ea postulo recenseri. ea postulo recenseri. A telling mark of literal translations is respecting the word order of the source language. Here we see that in a couple of cases. In the first sentence (1–3), the Latin keeps the verb-subject order (praecepit piissimus imperator) of the Greek (Ἐκέλευσεν ὁ εὐσεβέστατος βασιλεὺς): the verb-subject order is typical of Koine Greek but not of Latin, where the verb tends to be in the final position.31 In the second sentence, εἰσι πεπραγμένα (25) is translated keeping the same word order, sunt acta, which is less natural a word order in Latin than acta sunt—see congregata est and gesta sunt at 14 and 16: where the sixth-century editors came across synthetic forms in the Greek (συγκεκρότηται and πεπραγ- μένων at 7 and 16, respectively), they chose the more typically Latin word order for the analytic passives, with the verb at the end. But the Latin translation does not always respect the syntax and word order of the source language: in a few cases, the translators took some liberties to make the Latin more idiomatic. The coordinated καὶ δὴ συγκεκρότηται (6–7), for example, is rendered by a relative with the verb in final position, in the Latin fashion (quae … congregata est 6–14). The participle τῶν πεπραγ- μένων (16), which could have been rendered by the participle gestis (governed by de), is also rendered by a relative (his … quae gesta sunt 15–16). Finally, the Latin gets rid of the hyperbaton ἐπίσκοπον … Φλαβιανὸν (19–23) by moving the genitives before the accusative (18–23): Constantinopolitanae sanctae ecclesiae quondam episcopum Flauianum. From a semantic point of view, this translation is quite ac- curate. There are only a few minor imprecisions that would not have jeopardized the understanding of the readers. For example, συγκροτηθῆναι is translated with celebrari in the first instance (5), but συγκεκρότηται (7) is rendered by the more fitting congregata est (14). For θεῖον “divine” the Latin translations have sacrum (9). 31 G. Horrocks, Greek: A History of the Language and its Speakers2 (Chichester 2010) 108–109. 138 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 Also, the translators must have thought it unnecessary to trans- late the particle δή (6). The most striking difference between the Latin versions is the absence from the Φa and Φc of the translation for τοῦ εὐσε- βεστάτου βασιλέως ἡμῶν “of our most pious emperor” (11–13). It seems unlikely that this expression was independently omitted in the later manuscript tradition of Φa and Φc. The mistake must have come about during the translation itself, or even before— possibly the editor of Φa used a Greek text that already omitted the expression. What is certain is that here Rusticus took ad- vantage of his Greek exemplar to fill in the gap: he supplied piissimi imperatoris nostri but did not comment on the omission— quite the opposite of the cases seen above (the indiction and the translation of θειοτάτου), where Rusticus noted the issues but did not correct the text. Shortcomings of the Latin translations: different types of errors We have seen that the Latin translations are by and large reliable. The readers of the Latin translations were in a good position to understand what was written in the Greek Acts of Chalcedon. Surely there were errors, though. Some of the errors made by the translators of Φa were corrected by later editors, some were not. In certain cases, later editors introduced errors where they thought they were making corrections. Here I pre- sent some examples of the different types of errors that could mar the understanding of the proceedings of Chalcedon for those who had access only to the Latin Acts. I distinguish be- tween semantic and syntactic errors.32 The most frequent type of semantic mistake derives not from the selection of the wrong Latin translation of a Greek word but from the incorrect reading of a Greek word, or from textual corruptions in the Greek source manuscript. Let us consider the example of session 11 par. 53. In the Greek version (ACO II.1 411.20–21) the bishops of Asia state: ἐπεὶ εἴ τις ὧδε χειροτονη- θείη, καὶ τὰ παιδία ἡμῶν ἀποθνήισκει καὶ ἡ πόλις ἀπόλλυται 32 I follow the distinction of S. Lundström, Lexicon errorum interpretum La- tinorum (Uppsala 1983) 10–12. TOMMASO MARI 139 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 “For if someone should be consecrated here, both our children will perish and the city will be ruined.” Φa translated the con- ditional sentence as nam si aliquis hic ordinetur, correctly (ACO II.3 500.20–21). The editor(s) of Φc, however, must have read πείσει “he/she will persuade” instead of ἐπεὶ εἴ, because this version (followed by Φr) has suadeat instead of nam si. This also prompted the translators to change ordinetur to ordinari, so that the resulting sentence suadeat aliquis hic ordinari is quite distant from the original Greek. And to think that the editors meant that as a correction! Another example is in the the draft of Canon 4 at session 6 (par. 17). This decrees, “that no one is to build a monastery contrary to the will of the bishop of the city.” For “build,” the original Greek at ACO II.1 353.3 has οἰκοδομεῖν. Φa translated it as aedificare, correctly (II.3 438.12). The editors of Φc, however, read οἰκονομεῖν for οἰκοδομεῖν and translated it as disponere “ar- range” (II.3 437.14). Eventually Rusticus reverted it to aedificare, showing that he read the correct Greek. As for syntactic errors, the most frequent type derives from syntactic calques. At session 1 par. 643 John the presbyter wrote about Eutyches that “he suffers from the disease of the heretics” (νοσεῖν αὐτὸν τὰ τῶν αἱρετικῶν πάθη, ACO II.1 159.8). This is rendered by Φa with aegrotare eum haereticorum passione (II.3 143.15 app.). The editors of Φc corrected the ablative passione to the accusative passiones, following the Greek πάθη. But in doing so, they produced a syntactic calque which does not work in Latin, for Greek νοσέω can govern the accusative, but Latin aegroto cannot. Rusticus did not realize this and kept the accusative. Interestingly enough, the same sentence at ACO II.3 145.16 is translated correctly in all versions with πάθη becoming the ablative perfidia. While the syntactic calque of the previous case probably did not compromise the understanding of the passage, in other con- texts overly literal translations could obscure syntactic relations. In the synod held at Constantinople in 449 and reported in the first session of Chalcedon, at par. 729 the bishop of Constan- tinople asks about a paper containing the creed of Nicaea (ACO II.1 167.24–25); for “document” he uses ὁ χάρτης. The deacon 140 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 Eleusinius comments that the paper should have been accepted; he refers to the paper with the masculine pronoun αὐτόν. In all Latin versions (ACO II.3 154.10–12), ὁ χάρτης is translated with the feminine chartula, but αὐτόν is mechanically translated with the masculine eum; therefore, the relation between chartula and eum is lost. Latin, Greek, Latin again: the statements of the Roman delegates We have seen above that the Roman delegates spoke in Latin at the Council, and that their statements were translated into Greek by interpreters and recorded in both Latin and Greek in the original Acts. We have also seen that the extant Greek Acts include no texts in Latin; at some point in time these must have been excised. There is evidence, however, that the Latin trans- lators of the Acts had access to some of the original Latin materials. In what follows we shall see how. The condemnation of Dioscorus As a rule, the Latin translations of the statements of the Roman delegates are retroversions from Greek. That is to say, the Greek translations of the Latin statements were re-translated into Latin. The following text is a very clear example. This is the sentence of condemnation of Dioscorus delivered by the Roman delegates during the third session (the second in the Greek Acts), par. 94. We are lucky that the original Latin text was brought to Pope Leo, who attached it to his letter 103 addressed to the Gallic bishops (ACO II.4 155–156, ep. 112). By looking at the original Latin (II.4 155.34–156.2), the Greek translation (II.1 224.24–27), and the Latin retroversions (II.3 304.22–25)33 we can appreciate the translation process in its entirety—of course, the following analysis is valid only assuming that Pope Leo and the later tradition did not substantially alter the text delivered by the Roman delegates.34 33 As the differences between Φa, Φc, and Φr are not significant here, I present only Φa. 34 In fact, a corruption in the Latin text preserved by Leo can be found in the lacuna of ACO II.4 156.16, corresponding to over twenty words in the Greek translation. TOMMASO MARI 141 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 5 10 15 20 25 Original Latin Manifesta sunt quae Dioscorus – Alexandrinae – urbis antistes – contra canonum disciplinam regulasque ecclesiasticas inlicita praesumptione commisit, sicut praeterita gestorum – – seu praesens declarauit assertio. Γ Δῆλα γεγένηται τὰ τετολμημένα Διοσκόρωι τῶι γενομένωι τῆς Ἀλεξανδρέων μεγάλης ἐκκλησίας ἐπισκόπωι – κατὰ τῆς τῶν κανόνων τάξεως καὶ τῆς ἐκκλησιαστικῆς καταστάσεως – – – ἔκ τε τῶν ἤδη ἐξετασθέντων ἐν τῶι πρώτωι συνεδρίωι καὶ ἀπὸ τῶν σήμερον πεπραγμένων. Φa Manifesta facta sunt quae a Dioscoro quondam Alexandriae magnae ciuitatis episcopo commissa sunt aduersus regularem ordinem et ecclesiasticam disciplinam, – – – tam ex his quae dudum examinata sunt in priore consessu, et ex his quae hodie acta sunt. These first few lines of the statement are very instructive about translation processes. The Acts do not tell us whether the Latin statement, which was presumably read out, was translated into Greek on the spot or if the translation was prepared in advance. The Greek translation is not slavish. For example, it has some words that are not in the Latin:35 Dioscorus is τῶι γενομένωι … ἐπισκόπωι “formerly bishop” (5–9), whereas in the original Latin he is simply antistes “bishop,” as if he had not been deposed yet. In fact, he was a bishop until the sentence of condemnation was 35 One cannot rule out, in principle, that the omission or addition of some words is a mere accident of the later manuscript tradition. However, since that is ultimately impossible to ascertain, I shall trust the text established by Schwartz. 142 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 pronounced; the deposition was effective right away, and it might be that the Greek translators took that factor into account: when they translated this sentence, Dioscorus was already an ex- bishop. In the Greek version, the church of Alexandria in Egypt is described as μεγάλης (7), an adjective that is absent from the original—and the original talks of “city,” not “church.” By contrast, the Greek translation also omits something that is in the original: Dioscorus perpetrates his crimes inlicita praesumptione “with lawless audacity” in the Latin texts (16–17), but not in the Greek one. The Latin syntax is also adapted to the Greek idiom. Just as in the previous Latin translations we have appreciated that Greek participles are rendered by relative clauses in Latin, here a Latin relative clause is rendered by a participle in Greek: quae … commisit (3–18) becomes τὰ τετολμημένα (3). Even more to the point, the final comparative clause (sicut … assertio) is completely rearranged in the Greek version (19–26). All of this is so much more evident if one compares the original Latin text with the sixth-century Latin retroversions from the Greek. For example, one could hardly tell that tam ex his quae dudum examinata sunt in priore consessu et ex his quae hodie acta sunt of Φa (19–26) goes back to sicut praeterita gestorum seu praesens declarauit assertio. In other instances, however, the Latin translators did a better job of producing a text that was close to the original one. That is the case of the relative at 3 ff., for τὰ τετολμημένα is retroverted as quae … commissa sunt, which is close to the original quae … commisit (although in the retroversion the verb is not in the final position as in the original). In the brief section we have seen, the differences between the original and the translations mostly concern the form and not so much the substance. That is to say, the receiving end of the translation process(es) had access to fairly reliable translations of the original speech as far as its content was concerned—with the exception of a few words that were added or went missing—and their global understanding of the message of the Roman dele- gates was not distorted. Interestingly, where the Roman delegates mention papal primacy, the texts diverge a little more: in particular, the Greek (ACO II.1 225.14–17) is less prolix—or the original Latin (ACO TOMMASO MARI 143 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 II.4 156.21–24) is more prolix, depending on the point of view. The Latin retroversions do not show substantial differences, so I shall present all three of them as one (Φ at ACO II.3 305.21–24): 5 10 15 20 25 Original Latin unde sanctus ac beatissimus papa caput uniuersalis ecclesiae Leo per nos uicarios suos sancta synodo consentiente Petri apostoli praeditus dignitate, qui – ecclesiae fundamentum et petra fidei et caelestis regni ianitor nuncupatur Γ ὅθεν ὁ ἁγιώτατος καὶ μακαριώτατος ἀρχιεπίσκοπος τῆς μεγάλης καὶ πρεσβυτέρας Ῥώμης Λέων δι’ ἡμῶν – καὶ τῆς παρούσης ἁγιωτάτης συνόδου μετὰ τοῦ τρισμακαριω- τάτου καὶ πανευφήμου Πέτρου τοῦ ἀποστόλου, ὅς ἐστι πέτρα καὶ κρηπὶς τῆς καθολικῆς ἐκκλησίας καὶ τῆς ὀρθοδόξου πίστεως ὁ θεμέλιος – Φ unde sanctissimus et beatissimus archiepiscopus magnae senioris Romae Leo per nos – et per praesentem sanctam synodum una cum ter beatissi- mo et omni laude digno Petro apostolo, qui est petra et crepido catholicae ecclesiae et rectae fidei fundamentum – In the text transmitted in his own letter, Pope Leo is identified as “the head of the universal church” (caput uniuersalis ecclesiae, 5– 7), but in the Greek translation he is down-graded to “arch- bishop of great and senior Rome.” Who exactly is innovating here? Is Leo magnifying his titles in his letter to the Gallic bishops? Or is it rather the Greek translators who downplay the role of the Roman see within the universal church? One has to take into account that the stretch of time following the Council of Chalcedon was a period of friction between Rome and Con- stantinople over the idea of the Roman primacy within the 144 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 church.36 I have mentioned above that the official edition of the Acts promoted under the Emperor Marcian aimed to undercut the role of the Roman see at the Council. The downgrading of Pope Leo’s standing in this passage fits in well with that ten- dency. What is ironic here is that the sixth-century Latin trans- lators, who were actively pro-Roman, were tricked by the Greek translations into downgrading the Pope, while they probably would have been very happy to call him “the head of the uni- versal church.” In the present case, we are lucky enough to have the original Latin and we can tell that the words of the Romans were distorted in the Greek translation and, consequently, in the Latin retroversion. The question now is: how many more times did that happen without our being able to tell? Remnants of Latin in Greek manuscripts: Rusticus’ testimony This question brings us to a new topic, the remnants of sup- posedly original Latin in some Greek manuscripts of the Acts up to the sixth century and the translators’ approach to it. We have seen that the extant Greek manuscripts of the Acts do not have any Latin. But thanks to Rusticus we know that some Latin materials survived to his day. For in some of his adnotationes he writes that he found a few Latin interlocutiones in his Acoemete codex and in other unspecified Greek manuscripts,37 and in some notes he also transcribes them. Here is a list of statements that Rusticus found written in Latin in his Greek codices (all in ACO II.3; cf. II.3.3 XVIII): session 1 par. 9 (Lucensius) at 40.16 (Acoemete);38 session 4 par. 6 (Pascasinus) at 364.17 (Acoemete);39 session 4 par. 28 36 See S. Wessel, Leo the Great and the Spiritual Rebuilding of a Universal Rome (Leiden/Boston 2008) 285–321. 37 Schwartz, ACO II.3.3 XIV–XIX, lists the references to Greek MSS. 38 Here Rusticus talks of singulas interlocutiones, not simply of interlocutionem: this suggests that also the following statements of the Roman delegates (1.10, 1.12) were in Latin in the Acoemete codex (and perhaps the previous ones too, 1.5 and 1.7?). They were probably lost in the manuscript tradition. 39 Schwartz writes in the critical apparatus that another annotation of the TOMMASO MARI 145 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 (Pascasinus) at 374.27 (unspecified codex); session 5 par. 9 (Pascasinus) at 390.4 (unspecified codex); session 6 par. 3 (the Emperor Marcian’s speech) at 409.10 (Acoemete). Here, on the other hand, are the signatures that he found written in Latin:40 session 1 par. 945 nos. 2, 4, 5 (Arcadius, Projectus, and Philip at Ephesus I) at 228.16, 18, 19 (Acoemete);41 session 1 par. 945 no. 49 (Senecio at Ephesus I) at 230.4 (Acoemete); session 1 par. 945 no. 172 (Felix at Ephesus I) at 234.14 (arguably Acoemete); session 3 par. 971 (Pascasinus) at 331.1–3 (Acoemete); session 6 par. 91–3 (Pascasinus, Lucensius, Boniface) at 415.28 ff. The first statement of bishop Lucensius of Asculum What did Rusticus think of these materials? Did he take them to be the original Latin statements and signatures? In his notes he does not say that explicitly. The use he made of these, moreover, is not entirely consistent.42 In the very first case, session 1 par. 9 (Lucensius’ statement), Rusticus transcribed the Latin text that he found in the Acoemete codex in the margin of his text, next to the translated statement of Lucensius.43 That is, he did not replace the retroversion with the supposedly original statement, but simply used it for comparison in his ‘critical same kind referred to session 4 par. 92-4 at 365.17 (another statement of Pascasinus). 40 The Latin-speaking bishops could sign in their own language at the Councils. 41 Parts of the Acts of Ephesus I (431) were read out at Chalcedon and recorded in the Acts of that Council. These and the following signatures are preserved in Latin in the Greek Acts of the Council of Ephesus (cf. ACO I.1.7 111, 112, 113, 116). The corresponding section has been excised from the Greek Acts of Chalcedon, so it is impossible for us to verify in which language they were written there. 42 See the case of Pascasinus’ statements in the fourth session at par. 6 (ACO II.3 364.16–26) and par. 9 nos. 2–4 (365.16–24). Either Rusticus did not transcribe these or they were lost in the later manuscript tradition. 43 That is so in Par.lat. 11611 and 1458. On the other hand, the scribe of the Codex Veronensis 58 (or the scribe of its model) substitutes this for Lucensius’ statement in the body of the text. 146 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 apparatus’. Rusticus introduces Lucensius’ statement with the caption singulas interlocutiones a foris ita posui ut Acumitensium codex Graecus latine continet, where a foris seems to indicate that he found these statements in the margins of his Acoemete codex.44 I provide side by side four versions of this text: the one that Rusticus wrote in the margin (ACO II.3 40.16–19 apparatus), the Greek translation (II.1 65.30–32), and the Latin translations in Φac and Φr (II.3 40.16–19).45 The questions we will have to answer are: is the text in the adnotatio Rustici the original Latin statement? Is the Greek its translation and, if yes, what kind of translation is it? What is the relationship between the Latin translations and the text in the adnotatio Rustici? 5 10 15 20 25 Adn. Rustici Iudicii sui – – redditurus est rationem. – cum personam iudicandi non haberet, praesumpsit. – synodum ausus est facere sine auctoritate sedis apostolicae, quod numquam licuit, numquam factum est. Γ Τῆς ἰδίας κρίσεως – – λόγον ἀποδότω. πρόσωπον γὰρ ἥρπασεν τοῦ κρίνειν, ὅπερ οὐκ ἐκέκτητο· – σύνοδον ἐτόλμησεν ποιῆσαι ἐπιτροπῆς δίχα τοῦ ἀποστολικοῦ θρόνου, ὅπερ οὐδέποτε γέγονεν οὐδὲ ἐξὸν γενέσθαι. Φac Iudicii sui necesse est eum dare rationem, quia cum personam nec iudicandi haberet, subripuit et synodum ausus est facere sine auctoritate sedis apostolicae, quod numquam factum est nec fieri licuit. Φr Iudicii sui necesse est eum dare rationem, quia cum nec personam iudicandi haberet, subrepit et synodum ausus est facere sine auctoritate sedis apostolicae, quod numquam factum est nec fieri licuit. 44 Cf. ACO II.3.3 XVIII. 45 Here I present Φac together because they do not show differences. TOMMASO MARI 147 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 Some elements appear to show that the Greek text is a translation of the statement found by Rusticus, hence that the latter is the original. For example, the compound ἀποδότω (6) seems to be translated from the compound redditurus of the adnotatio Rustici (cf. the simplex dare of Φ); there is asyndeton between ἐκέκτητο and σύνοδον (13–15), as in the adnotatio Rustici between praesumpsit and synodum (while Φ have subripuit/subrepit et synodum); πρόσωπον … τοῦ κρίνειν “the role of a judge” (7–10) is not idiomatic and would be well explained as a translation effect of personam iudicandi. Φac and consequently the text in the main body of Φr are basically retroversions from the Greek. That is revealed by the following elements: quia (7) translates γὰρ (8), while there is no causal conjunction in the original Latin; subripuit/subrepit (13) is a translation of ἥρπασεν (9), not a variation on praesumpsit; the word order of the quod-relative at 22–26 is the same in the Greek and Φ and different from that of the adnotatio Rustici; also, fieri licuit in Φ (26) is closer to ἐξὸν γενέσθαι than to the mere licuit of the original Latin. There are several cases, however, in which both Φ and the adnotatio Rustici unexpectedly agree against the Greek. Some might simply be due to the natural Latin idiom: that is the case of iudicii sui (with the adjective in postnominal position) versus the Greek ἰδίας κρίσεως (1–2), and rationem (6) in postverbal position versus the Greek λόγον in preverbal position. Other similarities, however, are more striking: for example, at 8–12 both Latin texts have a cum-clause while the Greek has a relative with ὅπερ; also, the main clause is postponed to the subordinate, unlike in Greek. Finally, at 15–21 synodum … apostolicae runs in exactly the same way in both Latin versions; admittedly, this is no different from the Greek version, but one might expect that in the retroversion from Greek some differences would have arisen between the retroversion and the original Latin. What can we learn from the analysis of this passage? While one has to take into account that errors might have arisen in the manuscript tradition, one can also reasonably assume that the statement preserved by the adnotatio Rustici is the original one and 148 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 that it was recorded in the original bilingual Acts. The greater simplicity and more marked tendence to parataxis and asyn- deton are compatible with it being an oral statement, and one would hardly see why copyists would want to produce a less elegant text, if this was a secondary version. If things are so, it would follow that the Greek version is its translation as recorded in the proceedings. Of course one cannot tell for sure whether these were exactly the interpreter’s words or the result of later editing. This Greek translation is similar in its characteristics to that of the sentence of condemnation of Dioscorus seen above (session 3 par. 94); it is faithful enough but not quite literal. As for sixth-century Latin translations, although Φa and the later editions are essentially retroversions from the Greek, the ele- ments shared with the original Latin paint a more complex scenario: the first Latin translators of the Acts also had access to (some of) the original Latin statements.46 One cannot take for granted that they realized that these Latin texts were the originals. Unlike what we moderns would feel normal, they chose not to simply write down the Latin speeches, but to use them as an aid for their retroversion from the Greek. After all, their task was to translate the Acts from Greek.47 Rusticus’ approach is even more a case in point, and we can see more clearly what he did. He transcribed the text of Φc in his edition and corrected it based on his own sense of the language (he changed personam nec iudicandi to nec personam iudicandi and subripuit to subrepit). He did not use the original Latin to correct his version, but referred to it only as a term of comparison. 46 ACO II.1.3 XXIII; cf. E. Schwartz, “Der sechste nicaenische Kanon auf der Synode von Chalkedon,“ SBBerl 27 (1930) 611–640, esp. 622–623. A famous case discussed by Schwartz is that of session 16 on the privileges of the see of Constantinople. 47 In a private communication, Prof. Richard Price informed me of a ninth-century parallel in the work of Anastasius Bibliothecarius as a translator of the Acts of Nicaea II (787). Anastasius found in the Roman archives the originals of two letters from Pope Hadrian. He gave one in the original Latin but translated the other from the Greek, flagging parts where the Greek departed from the Latin. TOMMASO MARI 149 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 The signatures of the Roman delegates at Chalcedon (third session) We can observe variations on that pattern in some lists of signatures of Chalcedon.48 In his Acoemete codex, Rusticus found the Latin signature of Pascasinus, the chief of the Roman delegation, at session 3 par. 97 no. 1;49 however, he did not transcribe it in the margin, probably because it did not have substantial differences from that of Φc. Here Γ is at ACO II.1 230.14–16, Φ at II.3 331.1–3: 5 10 15 20 25 Γ Πασκασῖνος ἐπίσκοπος – Λιλυβαίων τῆς Σικελῶν ἐπαρχίας ἐπέχων τὸν τόπον – – – – τοῦ ἀρχιεπισκόπου τῆς μεγάλης Ῥώμης Λέοντος ὥρισα ἅμα τῆι ἁγιωτάτηι συνόδωι – ἐπὶ τῆι καθαιρέσει Διοσκόρου ὁμοῦ – – καὶ ὑπέγραψα Φac Pascasinus episcopus ecclesiae Lilibeo – – et uicarius beatissimi atque apostolici uniuersalis ecclesiae papae urbis Romae – – – sanctae synodo praesidens in Dioscori damnatione consensu uniuersalis concilii subscripsi Φr Pascasinus episcopus ecclesiae Lilybetanae – – uice beatissimi atque apostolici uniuersalis ecclesiae papae urbis Romae Leonis – – sanctae synodo praesidentis in Dioscori damnatione consensu uniuersalis concilii subscripsi 48 In the list of signatures from the Council of Ephesus I that was read out at session 1 par. 945, Rusticus found five signatures in Latin, but they are not as interesting for our purposes. 49 In his adnotatio he wrote: et ista interloquutio Latine iacet in codice Acumitensium. 150 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 What is most striking here is the extent to which Φac and Φr agree against the Greek. In particular, the way the Pope is referred to at 8–14 is much more magniloquent in the Latin versions, which define him as “the most blessed and apostolic (man) of the universal church, Pope of the city of Rome, Leo,” while in the Greek he is more simply the “archbishop of great Rome Leo.” At the same time, the Latin versions give a prom- inent position to Pascasinus (to the Pope in Φr), who is “presiding over the holy council” (16–20):50 the word for “presiding over” is simply omitted in the Greek version. We have observed the same phenomenon above, in comparing the original version and the Greek translation of the sentence of condemnation of Dioscorus pronounced by the Roman delegates (session 3 par. 94 nos. 1–3): the Greek version omitted the most magniloquent attributes of the Pope found in the original Latin text (caput uniuersalis ecclesiae), thus undermining the Pope’s primacy. In that case, the Latin versions are direct translations of the Greek and so they too omit the most flattering epithets for the Pope. Here, to the contrary, the flattery made it into the Latin Acts, meaning that already the editors of the Versio antiqua had access to the original signature of Pascasinus, and that the Latin materials that Rusticus found in the Acoemete codex matched that. The signatures of the Roman delegates at the sixth session The next passage to be examined presents a slightly different situation. This is the list of signatories of session 6 par. 9, confirming the Definition of Faith of the Council. Here Rusticus notes that the signatures of the three Roman delegates are written in Latin in the Acoemete codex (hae suscriptiones tres sic Latine continentur in codice Acumitensium ut hic). I consider only the first and more interesting signature, that of bishop Pascasinus. Compared with the previous cases, we have one more witness here: the Collectio Dionysiana, a Latin version of conciliar decrees and canons produced in the early sixth century by the monk 50 Pascasinus did preside over the third session. In the Pope’s mind, how- ever, he should have presided over the other sessions as well, but that did not happen; the Council was chaired by the officers of the court of Constan- tinople instead. TOMMASO MARI 151 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 Dionysius Exiguus (ACO II.2 157.27–29). Γ is at II.1 337.17–19, Φ at II.3 415.28–30: 5 10 15 Γ Πασκασῖνος ἐπίσκοπος ἐπέχων τὸν τόπον τοῦ δεσπότου μου τοῦ μακαριωτάτου καὶ ἀποστολικοῦ τῆς οἰκουμενικῆς ἐκκλησίας ἐπισκόπου πόλεως Ῥώμης Λέοντος ἐν τῆι συνόδωι τοῖς Ἑλληνιστὶ ἀναγνωσθεῖσιν ὑπέγραψα Φa Pascasinus episcopus tenens locum domini mei beatissimi atque apostolici uniuersalis ecclesiae papae urbis Romae Leonis in concilio his quae graeca lecta sunt suscripsi Φc Pascasinus episcopus uicarius – domini mei beatissimi atque apostolici – – papae – – Leonis – – statuens suscripsi Φr Pascasinus episcopus uice – domini mei beatissimi atque apostolici uniuersalis ecclesiae papae urbis Romae Leonis synodo praesidens statui consensi et suscripsi Here Φa is clearly translated from the Greek; Φc presents some differences and omissions (but at least the omissions might have arisen in the medieval manuscript tradition). There is one striking difference between the Versio Rustici and ΓΦa: this time it lies in Pascasinus claiming for himself the role of president of the Council in the Versio Rustici (synodo praesidens statui consensi et suscripsi, 16–19), as opposed to a more passive role in ΓΦa (ἐν τῆι συνόδωι τοῖς Ἑλληνιστὶ ἀναγνωσθεῖσιν ὑπέγραψα = in concilio his quae graeca lecta sunt suscripsi ). The genuiness of the signature found by Rusticus in the Acoemete codex seems confirmed by comparison with the more ancient Collectio Dionysiana.51 Once again, as in the case of the condemnation of Dioscorus seen above, the Greek version undercut the prestige of the Roman faction and at least the early Latin translation fell for it. Rusticus, 51 One can exclude that Rusticus drew this information from the Collectio Dionysiana, because he provides a better text of the next two signatures. 152 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 thanks to his philological research, restored the genuine text and with it the prestige of the Roman delegation.52 The speech of the Emperor Marcian at the sixth session The Emperor Marcian presided over the sixth session of the Council and delivered a speech in Latin and then in Greek.53 It was probably the importance of this occasion that prompted Rusticus to go out of his way to provide as much evidence as he could. Let us consider the content included in the different ver- sions. The extant Greek Acts have only the Greek speech; however, they retain traces of the Latin speech, because the Greek one is preceded by the caption “Translation of the Latin address” (῾Ερμηνεία τῆς ῾Ρωμαικῆς προσφωνήσεως).54 The Ver- sio antiqua and Versio antiqua correcta have only the Latin translation of the Greek speech. Rusticus found the original Latin speech in his Acoemete codex55 and transcribed that in the first place;56 then he transcribed the Latin translation of the Greek speech that he found in the Versio antiqua correcta, having corrected it against the Greek.57 52 In this case, however, it was incorrect of Pascasinus to claim the presidency of the Council, for the Emperor Marcian presided over the sixth session. But Pope Leo expected that Pascasinus should preside over the whole Council, so Pascasinus must have signed according to the Pope’s expecta- tions. 53 ACO II.1 335.19–21: προσεφώνησεν τὰ ὑποτεταγμένα πρότερον ῾Ρω- μαιστὶ καὶ μετὰ τὴν ῾Ρωμαικὴν προσφώνησιν ῾Ελληνιστί “He delivered the following address in Latin first, and after the address in Latin then in Greek.” 54 This cannot refer to the following Greek speech, which is introduced by its own caption: ῾Ο θειότατος … Μαρκιανὸς … ῾Ελληνιστὶ προσεφώνησεν οὕτως. 55 The adnotatio Rustici in Par.lat. 11611 reads: in Acumitorum uero quodice (sic) mox Latine scriptum est ita (ACO II.3 409.12 app.). 56 There can be no doubt that the version found by Rusticus in the Acoemete codex is the original Latin speech of Marcian, for the same version is transcribed almost without variants in the Collectio Vaticana (ACO II.2.2 97.19—98.16). 57 In his marginal adnotatio, he commented thus: a signo isto totum sic emendaui ad Graecam proprietatem. TOMMASO MARI 153 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 We have observed in the case of the first statement of Lu- censius at the first session (par. 9) that while the translators of the Versio antiqua essentially retroverted the statement from the Greek, they also had access to the original statement in Latin and used it in their retroversion. This is the only way to explain the striking similarities between the Versio antiqua (and correcta) and the original Latin statement found by Rusticus. The case of Marcian’s speech is even more evident. Let us consider the first few lines of it in four versions: the original Latin one as preserved by Rusticus and the Collectio Vaticana (ACO II.3 409.12–14 = II.2 97.19–20); the original Greek speech as preserved in the Greek Acts (ACO II.1 335.27–30); the Latin translation of the Greek speech of the Versio Antiqua/Correcta (ACO II.3 410.14–15 apparatus), and the same Latin translation as corrected by Rusticus against the original Greek (ACO II.3 410.14–15): 5 10 15 Original Latin Vbi primum – – diuino iudicio ad imperium sumus electi, inter tantas necessitates rei publicae – nulla nos magis causa constrinxit – Γ Ἐν προοιμίοις τῆς ἡμετέρας βασιλείας, θείαι ψήφωι ἐπ’ αὐτὴν αἱρεθέντες, πρὸ τῶν ἄλλων ἁπασῶν καὶ ἀναγκαιοτάτων φροντίδων οὐδὲν οὕτω καὶ βουλῆς καὶ σπουδῆς ἄξιον ἐνομίσαμεν Φac Vbi primum – – diuino iudicio ad imperium sumus electi, ante alias omnes publicas et necessarias curas(Φa)/cau- sas(Φc) nihil sic et consilio et studio dignum putauimus Φr Vbi primum – – diuino iudicio ad imperium sumus electi, ante alias omnes publicas et summe necessarias causas nihil sic et consilio et studio dignum putauimus In Φ the main clause ante alias … putauimus (7–16) is clearly translated from the Greek, just as one would expect. That is the case for most of Marcian’s speech. However, the initial clause from ubi to electi (1–6) is not a translation of the Greek, but a faithful transcription of the original Latin just as reported by Rusticus and the Collectio Vaticana. This shows that the editors of Φa had access to at least parts of the original Latin speech of 154 THE LATIN TRANSLATIONS OF CHALCEDON ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 Marcian and occasionally used it.58 An interesting question is why this happens only occasionally. The answer to this can only be speculative at this stage. One could imagine, for example, that the first translators resorted to the Latin original where their Greek manuscript was damaged or they could not make sense of the Greek text. If we turn to Rusticus’ text, finally, we can appreciate that in order to emend Φc he resorted to the Greek: for example, while Φac translate the superlative ἀναγκαιοτάτων with the positive necessarias, Rusticus restores the idea of the superlative by writing summe necessarias (9–10). Conclusions This overview of the sixth-century Latin translations of the Acts of Chalcedon has illustrated that the Latin Acts are by and large reliable: the translators did a good job of producing a translation that was faithful to the Greek, yet most of the times idiomatic in Latin. Translation mistakes obviously occur but they rarely compromise the understanding of the text; they can be divided into semantic ones (wrong meaning or wrong word translated) and syntactic ones (calques that either are unidiomatic in Latin or produce broken syntax). Oral statements are translated a little more freely than the more formulaic ‘narrative’ frame. The first translation was revised twice, but the subsequent editors did not produce an altogether different text; the differences are in the details. Each of the later editors corrected the previous version against Greek manuscripts. The Greek Acts originally included some parts in Latin, especially the statements of the Roman delegates. In the Latin versions, such parts were normally retroverted from Greek, like the rest of the Acts. Here one has to consider that the Greek Acts were produced amid tensions between Constantinople and Rome and favoured the former, while the Latin translators supported the Roman see. The Greek translations of the Latin 58 There are a few more passages in which the Versio antiqua and the original Latin clearly agree against the Greek (all in ACO II.3): 410.23–24 app. (Φa) = 409.19 (original); 410.30–32 app. (Φa) = 409.24–26 (original); 411.1–3 (Φa) = 409.27–29 (original). TOMMASO MARI 155 ————— Greek, Roman, and Byzantine Studies 58 (2018) 126–155 originals were quite free and at times distorted their message by undercutting the role of the Pope and his delegates; as a con- sequence, the Latin retroversions too could involuntarily distort their message. However, the sixth-century translators had access to some Latin originals, of which they took advantage in different ways: at times they used them as an aid to the retro- version, less often they transcribed them instead of producing a retroversion from the Greek. In this way they sometimes man- aged to bypass the filter of the Greek and restore the prestige of the Roman see.59 December, 2017 Otto-Friedrich-Universität Bamberg tommaso.mari@uni-bamberg.de 59 This article is part of a project that has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 677638. Previous versions of this paper have been presented at the Shifting Frontiers in Late Antiquity 12th Biennial Conference at Yale and at the 19th International Colloquium on Latin Linguistics in Munich. For this work, I am grateful to many: Peter Riedlberger has greatly helped during the various stages of the paper’s preparation; Luisa Andriollo, Maria Constantinou, Thomas Graumann, and Geoffrey Horrocks have read and commented on previous drafts; Kent Rigsby and an anonymous reader have contributed to the final version with useful suggestions. Finally, I wish to thank Kristin Hanna for helping with my English. All remaining mistakes are my responsi- bility alone.