In recent years, scholarly interest in the historical epic Iohannis seu de bellis Libycis, composed in eight books by the African poet Flavius Cresconius Gorippus,Footnote 1 has grown significantly. The poem recounts the victorious campaign waged by the Romans under the command of the magister militum vacans, John Troglita, against the rebellious Moors in AD 546–48. Through flashbacks, the narrative also covers the years 533–45. It thus includes Belisarius’s famous invasion and the subsequent struggles to retain the newly-reconquered Africa within the empire. Owing to its highly detailed account, the Iohannis – though a poetic composition – stands as a crucial source for the history of late-Vandal and Byzantine Africa.
A separate issue in research on Gorippus’s Iohannis encompasses its manuscript tradition, the history of its editions, the ensuing debate regarding its content and, consequently, a reflection on what actually constitutes the historical source underpinning our knowledge. The poem has survived in only a single fourteenth-century manuscript, Codex Trivultianus 686, currently held at the Trivulziana Library in the Castello Sforzesco in Milan.Footnote 2 The copyist of the manuscript was the unfulfilled Tuscan poet Giovanni De Bonis di Arezzo (ca. 1350–1404), who drew upon this little-known poem in his own literary work – a circumstance that ultimately contributed to the unexpected preservation of the Iohannis and its subsequent discovery among his writings in 1814 by Pietro Mazzucchelli.Footnote 3 De Bonis transcribed the Codex Trivultianus solely for his own use in a semi-Gothic minuscule, executing it exceptionally carelessly and with clear influences from the vernacular. None of the previous editors of the Iohannis has been a specialist in Gothic palaeography; hence, critical remarks regarding the quality of Giovanni De Bonis’s handwriting seem to stem from a lack of expertise amongst his subsequent readers.Footnote 4
De Bonis was not a historian. It seems that he was only moderately interested in this aspect of the work (suffice it to say that he used the Iohannis primarily to praise Gian Galeazzo Visconti in the poem Viscontina),Footnote 5 which accounts for the numerous simplifications, errors and misreadings of individual words, including, in particular, the names of characters. On the other hand, even a correct reading of this handwriting often makes no logical sense. Unfortunately, we do not know what the original manuscript, copied by De Bonis, looked like, what its state of preservation was, or how faithfully it reproduced the text of the Iohannis. Many of the errors attributed to Giovanni De Bonis may, in fact, have originated in the copy he transcribed.
Starting with the editio princeps (1820) of Pietro Mazzucchelli – an editor and scholar of great diligence, endowed with a truly remarkable philological instinct – subsequent editorsFootnote 6 proposed conjectures and emendations that significantly altered the text of the Iohannis, often failing to indicate not only the proposals of their predecessors but also the readings transmitted in the Codex Trivultianus. Consequently, a scholar wishing to refer to the work of the African epic poet is sometimes forced to consult all the critical editions (which are not all that numerous), the manuscript itself (recently digitised, though in a far from satisfactory manner)Footnote 7 and all available partial editions with commentaries, articles and reviews, as their authors sometimes introduce further changes to the text or, even more frequently, reverse the unjustified emendations of their predecessors.
To illustrate the above points, I shall cite a few examples concerning the names of historical figures and toponyms relating to the sites of battles mentioned exclusively by Gorippus; therefore, the correct interpretation of this information has significant implications for contemporary historical debate.
Thus, according to the widely-used edition by James Diggle and F.R.D. Goodyear (Reference James Diggle1970), as well as modern translations based upon it,Footnote 8 verse Ioh. I 62 (Figure 1) contains the reading: congressus Nabedes fretus uirtute feroci. This refers to an important Persian commander, ‘second only to the king’, who was defeated by the Romans at Nisibis in 541 (cf. Procop., Hist. bell. II 18, 9; PLRE IIIb, p. 909). In Codex Trivultianus 686,Footnote 9 however, the reading is nec habet (actually an overwritten letter ‘n’ and the abbreviation ‘ht’), which Pietro Mazzucchelli also retained in his editio princeps. He further divided the entire line into two sentences by inserting a full stop between the words fretus and uirtute: Parthoque a rege secundos congressus nec habet. Fretus uirtute feroci, amisit socias, ipso superante, cateruas. It was not until the next editor of the poem, Immanuel Bekker (Reference Bekker1836), that the change from nec habet to Nabedes was proposed, and all subsequent editors adopted his emendation. It should be noted that this is the only place in the entire Iohannis where this Persian commander is mentioned. Although Bekker’s emendation, based on the parallel (though in many respects also factually different) Greek text of Procopius, seems convincing, it is a classic attempt to align a problematic passage with another source. The danger inherent in this method is, on the one hand, the tendency to regard unique information as textual corruption or later interpolations, and, on the other, the tendency to diminish the source value of a work whose narrative becomes subordinate to the text with which it is compared.
Ioh. I 62.
A striking example of the consequences of this practice is the paradoxographic work, the Collectanea rerum memorabilium, by Gaius Julius Solinus (third century AD), whom the eminent historian and editor Theodor Mommsen regarded as a thoughtless compiler of older works, foremost among them Pliny the Elder’s Natural History. Consequently, Mommsen treated information not attested elsewhere as medieval interpolations and removed it from his edition (still the standard one to this day), thereby preventing it from entering scholarly circulation. What is more, he reconciled textual variants with the wording of texts that Solinus was supposedly merely transcribing, depriving the author of his unique style. The discovery of several minor late-antique fragments of Solinus’s work has, however, confirmed the reliability of at least part of the rejected tradition. A new critical edition of the work is currently being prepared under the supervision of Giuseppe Zecchini.Footnote 10
To return to the Iohannis. The editors’ interventions went even further in verse I 89 (Figure 2): Mermeroen. Campis Romanum Persa sequentem, which in the manuscript clearly begins with the words memor enim (written using a common abbreviation). In this instance, too, the reference is to a Persian commander (Pers. Mihr-Mihroe, PLRE IIIA, pp. 884–85), known from other sources, including, in particular, Procopius’s History of the Wars. On this occasion, the editor of the editio princeps, Pietro Mazzucchelli, was able to link the words memor enim – which did not fit the context – with a name that had appeared in the text of the poem slightly earlier (line 75: mermeroes) and again later (line 98: the clearly corrupted reading mermeroos). This speaks all the more to his credit, given that in the very same line 89, the reading per fusa vel perfusa – corrected by Mazzucchelli to effusa – was not emended to Persa until 1879 by Joseph Partsch, who noted Giovanni De Bonis’s characteristic practice of writing s as f (e.g. in Ioh. V 297 and VI 33). Furthermore, Gorippus thus became the only known author to record the capture of Mermeroes by the Romans at Dara. In line 77, Mazzucchelli altered the manuscript’s transmission from claras to Daras (Figure 3) – a place name that likewise appears only once in the poem – an emendation on which all subsequent editors have agreed.
Ioh. I 89.
Ioh. I 75–77.
Leaving aside the significance of the above observations and the scholarly debate surrounding them, the source on which researchers rely in this case is not the manuscript text itself, but rather the emendations introduced in the critical editions. This phenomenon is illustrated even more clearly by line I 100 (Figure 4). According to the unanimous consensus of editors, this verse contains the name of Emperor Justinian’s special envoy to the East, who was sent to summon the poem’s protagonist, John Troglita, back to the capital. In the manuscript, however, we find the phrase urbi cui, which Partsch (Reference Partsch1879) read as urbi cuius and emended to Urbicius. This emendation, serving as the sole attestation of this figure’s name and activity, was subsequently adopted as the basis for an entry in the influential Prosopography of the Later Roman Empire (Vol. IIIb, Cambridge 1992, pp. 1,393–94). Yet, this reading is not significantly more plausible than the one proposed by Mazzucchelli – Urbicapus, which was later adopted by Bekker (Reference Bekker1836) – at least from a textual perspective, as Mazzucchelli’s ingenious proposal drew upon his classical erudition, specifically his reading of Plautus’s Miles gloriosus (composed some seven centuries before the Iohannis), where the word appears as a derisive epithet for the protagonist (line 1,055: exprome benignum ex te ingenium, urbicape, occisor regum).Footnote 11
Ioh. I 100.
As has already been mentioned, it is unfortunately unclear whether the corruption of proper names in the Codex Trivultianus 686 stemmed from a substandard exemplar or from the copyist’s limited knowledge. On the one hand, one can observe, that Giovanni De Bonis correctly recorded well-known names such as Justinian (Iustinianus), John (Iohannes) and Gregory (Gregorius); indeed, in the margin next to the incorrect reading gragorius in Ioh. IV 487, the copyist himself noted the correct form. On the other hand, however, problems consistently arise with the transcription of less recognisable names and terms. For instance, in Ioh. I 367 (bellisarius – Figure 5), the name may have been associated by De Bonis – or by the scribe of his exemplar – with the Latin word bellum. This is further suggested by Ioh. I 387 (Figure 6), where the commander’s name is written with a single ‘l’, and Ioh. III 313 (libelisarius; Figure 7). Consequently, when dealing with such uncommon names, it is indispensable to evaluate critically the conjectures and emendations proposed by successive editors.
Ioh. I 367.
Ioh. I 387.
Ioh. I 313.
Based on the above examples, one might conclude that the essential tool for a researcher wishing to make use of the Iohannis text should be the latest critical edition by Diggle and Goodyear, in which such issues are noted in the apparatus criticus. However, this is not always sufficient. The critical apparatus of each of the few existing editions presents only a selection of readings and proposals from earlier scholars – specifically those that the current editors deemed significant. While this is an understandable practice, particularly given commercial or editorial constraints, the situation differs for a poem preserved in only a single manuscript of highly questionable accuracy. In such a case, reckoning with all past emendations and critical remarks (which are, after all, very few) would not significantly expand the apparatus. By doing so, readers of the Iohannis would gain access to alternative interpretations or textual corrections, even if these were not ultimately incorporated into the main body of the new edition. Line Ioh. III 318 (Figure 8) serves as an ideal example to support this claim.
Ioh. III 318.
The manuscript appears to transmit the reading cellas uatari (in minuscule) as the name of one of the sites of skirmishes with the Moors, where John Troglita was said to have distinguished himself. Nonetheless, the critical apparatus in the Diggle–Goodyear edition omits Mazzucchelli’s suggestion, according to whom the first word did not refer to a place name but to the inhabitants of the Vatari region (te cultor Vatari miro spectabat amore). The German philologist Paul Haupt held a similar view (Opuscula III, Leipzig 1876, p. 626), proposing the emendation tellus (te tellus Vatari miro spectabat amore). Shortly afterwards, Joseph Partsch (Reference Partsch1879) concluded that the poet was referring to the locality of Scalae Veteres, commonly identified with the site mentioned in Procopius’s Hist. bell. IV 17, 3: ἐν χωρίῳ ὃ δὴ Σκάλας Βέτερες καλοῦσι Ῥωμαῖοι – the theatre of a battle fought by the praefectus praetorio Africae, Germanus, against Stotzas’s rebels in the spring of AD 537. Partsch’s reading, Cellas Vatari, was subsequently adopted by Michael Petschenig (Reference Petschenig1886) – without any reference in the notes to the earlier proposals by Mazzucchelli and Haupt – and later accepted by Diggle and Goodyear. It must be admitted that this way of interpreting the Trivultianus 686 text is entirely justified from a palaeographical perspective. In this manner, however, the word cellas becomes a classic hapax legomenon, which increases the probability that Mazzucchelli’s and Haupt’s intuitions were correct and that we are dealing here with a corrupted term referring to the locality of Vatari, though not necessarily one easily restored by a common word such as tellus or cultor.
The context of Gorippus’s account would also point to Partsch’s conclusion that the poet had the same toponym in mind as Procopius, though he further mentions another skirmish unknown to the Byzantine historian, near Autenti. Notwithstanding, the place named Scalae Veteres is otherwise unattested, whereas Vatari is documented on the famous Tabula Peutingeriana (IV 3) and in the inscription CIL VIII 22899. This refers to a locality on the border between Numidia and Proconsular Africa, on the road from Carthage to Sétif, at the junction with the road leading to Theveste – perhaps identical to present-day Fedj es-Siouda. Its identification with Procopius’s place name is also suggested in the article by Christian Courtois in the influential Realencyclopädie (Vatari, RE VIII A.1, 1955, col. 488).Footnote 12 Still, this does not mean that the reading transmitted by the Iohannis manuscript should be aligned precisely with the Greek account of the famous historian, even if he was referring to the Latin form of the place name as he had heard it. This would lead us to an interesting paradox: a historian writing in Greek records a Latin name with a specific meaning (‘Old Stairs’), while an African poet writing in Latin records a name whose possible meaning remains unknown to us, although according to some scholars it may have had a Libyan etymology. This is further suggested by similar toponyms and ethnonyms, such as the town of Cilium, the Chellenses tribe attested in inscription CIL VIII 16352 found at Cellae, mons Transcellensis (Amm. Marc. XXIX 5, 20), limes caputcellensis (Not. Dign. occ. 25, 32 and 30, 18) and finally Urceliana manus (Ioh. VI 390), which, by the way, raises similar doubts as Cellas.Footnote 13 However, the hypothesis that the name Cellas is of Libyan origin is dependent on the view that its reading is correct.
Meanwhile, an examination of the critical apparatus in Procopius’s own work reveals that he did not necessarily employ the name Scalae Veteres.Footnote 14 Specifically, from Jakob Haury’s standard edition,Footnote 15 it appears that the most important fourteenth-century manuscripts containing the History of the Wars transmit readings strikingly similar to Gorippus’s (Codex Trivultianus) version: καλασβατάρες V (Codicis Vaticani Graec. 152 pars prior), καλλασβατάρας P (Cod. Parisinus Graec. 1702) and καλασβατάρας O (Cod. Ottobonianus Graec. 82). The reading Σκάλας Βέτερες is a later proposal by Joseph Scaliger contained in the much later manuscript α (Cod. Leidensis), which Haury, as he himself notes, knew only at second hand through Wilhelm Dindorf’s edition from 1833.Footnote 16 This emendation was also accepted by Wilhelm von Christ, the initiator of Haury’s edition. However, if one consults Dindorf’s edition – which was in fact a compilation of even earlier editions – one finds that he opted for a different reading from Haury: Kαλλασβατάρας (and it was this variant that Partsch might have had in mind in 1879). What is more, in Claude Maltret’s (Maltretus) Latin translation of 1662, included below the main text of the Dindorf edition, the place name is rendered as Scalas veteres, which in turn suggests that the correct form of the toponym was Scalae. Apart from Scaliger’s conjecture, Dindorf also noted in his critical apparatus another reading by Raffaello Maffei (Raphael Volaterranus) from his 1506 Latin translation of The Persian War and The Vandal War: Calabastores.Footnote 17
Diggle and Goodyear, in their edition of the Iohannis (following Partsch and subsequently Petschenig), therefore stop halfway. They retain both the variant Vatari – attested in other sources – and the variant Cellas, which indeed appears to be present in the manuscript, but at the same time seems to echo the problematic Greek variant from Procopius’s History of the Wars. A complete list of the rejected emendations would not ultimately alter the text of the manuscript itself; yet, for researchers, readers, or translators relying on such an edition, it would serve as a valuable tool and a starting point for reflection on the actual name of what is, in other respects, a significant site in Byzantine Africa.
The greatest problem for editors, however, is posed by names that are not attested anywhere else, particularly those of unambiguously non-Latin origin. The numerous Germanic names in the poem – such as those of Vandals or Germanic officers in the Roman army (e.g., Guntarith, Putzintulus, Fronimuth)Footnote 18 – intermingled with Latin, Armenian and even Persian names, sometimes lead to confusion as to who is being referred to at any given moment. For example, in Ioh. VI 520–22 (Figure 9), an ink stain has damaged the already barely legible text in the Trivultianus manuscript. In Diggle and Goodyear’s reading, this passage recounts the battle struggles of four of John Troglita’s officers: his namesake John, Putzintulus, Geisirith and Sinduit (ductoris senior felix non sorte, Iohannes. / At laeuum tenuere latus Putzintulus ingens, / Geisirith arcitenens et celsis Sinduit armis). Notably, the words in the first line between duc- and sorte are completely obscured. In Pietro Mazzucchelli’s editio princeps, where line 520 belongs to Book V (a structural issue discussed below), the verse reads: Ductor erat peditum Felix, non sorte, Iohannes. The editor thus appears to introduce an additional officer named Felix. Diggle and Goodyear do not even mention this suggestion, following instead Joseph Partsch’s proposal: ductoris, senior, felix (Diggle himself proposed a further correction: simili felix instead of senior felix). The names of the other commanders have also been modified. In the manuscript, one finds the readings butzuntulus (corrected to Putzintulus by Mazzucchelli), grisirith (Geisirich by Mazzucchelli; Geisirith by Partsch and subsequent editors) and induit, which Mazzucchelli emended to inclitus. Although this latter emendation was adopted by Bekker, there is no mention of it in the Diggle–Goodyear edition. In Mazzucchelli’s interpretation, therefore, there would be no mention whatsoever of an additional officer’s name. The reading Sinduit was only introduced later in Partsch’s edition.
Ioh. VI 520–22.
It is worth noting that even the last two Vandal kings are referred to in the Iohannis by slightly different variants of their names from those found in contemporary works. Specifically, in the Codex Trivultianus 686, the first of these rulers is referred to as Hildimer (Ioh. III 198 and 219), whilst the second is referred to as Geilamir (Ioh. I 381 and III 17).Footnote 19 Due to the lack of comparative material, it is impossible to determine whether the poet himself preferred this spelling, or whether it is the result of a careless transcription by the manuscript’s copyist, Giovanni De Bonis. It is, however, essential to observe that in both cases the sources provide us with several different variants of these names, probably resulting mainly from the uncertainty of Greek and Latin authors regarding the transcription of Germanic anthroponyms. In the case of Hilderic, apart from the form Hildericus – which predominates in modern scholarship – the following variants exist: the Greek Ἰλδέρικος (Procopius) and the Latin Hildirix (on coins and epigraphs), Hilderix (Victor of Tonnona, in the accusative Hildericum; see also the genitive Hilderici in the Acta of the Concilium Carthaginense AD 525) and Hildirit (Victor of Vita). These variants are recorded in The Prosopography of the Later Roman Empire.Footnote 20 Unfortunately, the authors of the study did not take into account that these forms may also vary within the manuscripts transmitting the content of the sources; in critical editions, and even more so in translations, such variants are frequently standardised or emended. For example, in Theodor Mommsen’s critical edition of Jordanes’ Getica (1882), the reading Ilderich is found, while Antonino Grillone (2017) favours the variant transmitted by the c-manuscript family – Hildericus. In turn, in two versions of the local African work Laterculus regum Vandalorum et Alanorum (recensio Augiensis and recensio Hispana, ed. Becker-Kötter), the variants Hiltirix, Hildirix and Hildrix coexist. Alongside the latter, in the very same version of the text, Hilderico appears in the dative. In the case of Gelimer (the variant used by Procopius of Caesarea), the form Geilamir (also Geilamer, Gheilamir) – presumably closer in sound to the Vandalic original – is equally common in the sources and was probably penned by Gorippus himself.
When dealing with the even greater number of Moorish names in the poem – which are often indeclinable – the situation becomes exceptionally complex. For example, in Ioh. V 338 (line 934 in Book IV according to Mazzucchelli and Bekker; Ioh. IV–V 338 in Partsch; line 290 in Book V according to Petschenig and Diggle–Goodyear; Figure 10), the editions by Mazzucchelli and Bekker describe the brutal death of a warrior named Athiman: Extremis Athiman singultibus ore uocabat. Partsch, however, modified this name to Authima. Petschenig hesitantly agreed with this correction, labelling it a crux philologorum, while Diggle removed the warrior’s name altogether, reading the passage instead as haud iam and relating it to another warrior, Roffas, mentioned earlier (stomachoque peremptus / flumina rubra uomit Roffas atque ilia pulsans / extremis haud iam singultibus ore uacabat). Personally, I would suggest reading the word attamen here, or possibly autem hians, which would alter the interpretation accordingly.
Ioh. V 336–38.
The preceding examples illustrate not only a divergent numbering of verses but also a fundamental difference in book division. In the history of research into the text of the Iohannis, such a change has occurred twice. In the Codex Trivultianus 686, the books are not numbered numerically; instead, they are separated by blank spaces and distinctive gallows-shaped markers above the first word of each book. Mazzucchelli counted seven such books, of which Book IV comprised as many as 1,171 verses. Since the Codex Budensis, known to us only indirectly from a mention by the humanist Johannes Cuspinianus in the early sixteenth century, contained a poem in eight books,Footnote 21 Mazzucchelli assumed that Book VIII was missing from his manuscript, and that its content was devoted to a description of John Troglita’s triumphant return following his victory over the Moors.Footnote 22 This division was also retained in Bekker’s Reference Bekker1836 edition.
In 1879, however, the German scholar Gustav Löwe reported that he had identified, in the Veronese florilegium Flores moralium auctoritatum (dated 1329), 22 further verses omitted from the Trivultianus manuscript (including three passages pertaining to Book VI). Based on the numbering of the books in the Veronese florilegium and the logical – though contrary to ancient practice – inference that a new book should open with a description of the battle itself, Löwe divided the massive Book IV into two smaller sections after verse 644, resulting in a total of eight books, in accordance with Cuspinianus’s testimony. His findings were adopted that very same year in Joseph Partsch’s edition (which, however, retained the old line numbering and continued to combine Books IV and V), as well as by subsequent editors and translators.Footnote 23
It was not until the Trivultianus manuscript was re-examined at the beginning of the 21st century by Giulia Caramico and Peter Riedlberger that it became apparent that both Mazzucchelli and Partsch, as well as Diggle and Goodyear – namely, the only editors who had consulted Giovanni De Bonis’s manuscript while preparing their editions (the latter two only via microfilm) – simply failed to notice his characteristic, though in this case less distinct, mark above the first word at the top of folio 33v. Apart from such markings, De Bonis, as already mentioned, separated the individual books by a line break. Nevertheless, Book V begins on a new page, meaning that a line break was redundant, which made the transition to the next section of the poem easy to overlook. Consequently, this book actually begins with line 597 of Book IV in older editions (bellorum at princeps nocturno tempore lerna), a restructuring that introduces a more rational balance of content across the entire work (yielding a line distribution per book of: 581, 488, 460, 596, 575, 773, 542, 656).Footnote 24 Other scholars quickly accepted this discovery; however, it has necessitated the use of dual line numbering in parts of Books IV and V until a new edition is published to supersede that of Diggle and Goodyear.Footnote 25
The shift in verse numbering is not the only alteration to the text’s layout that future editors and, indeed, all scholars should take into account. In Ioh. IV 200–204 of the Codex Trivultianus 686 (Figure 11), the first line is clearly truncated (Dux simul Marturius illis), while in the left margin near line 204 the copyist has added the words postquam medio. Pietro Mazzucchelli had already proposed using the marginal note to reconstruct the missing line (Dux simul e mediis postquam Marturius illis). Partsch merely noted the presence of the lacuna, suggesting that more lines had been omitted from the text, in which, according to him, the poet described the death of another Roman commander, John, son of Sisiniolus. Finally, Petschenig, and after him Goodyear, decided to insert verses 35–37, which describe this event, in place of the lacuna: dux simul / ereptus †medius† felici morte Iohannes / occubuit: fastus domini non pertulit hostis / sorte pari, enses nec timuit captiuus eriles / Marturius illis. Benjamin Goldlust rightly rejected this drastic alteration in his recent edition of Book IV, while conceding that multiple lines had indeed been omitted from the original text at this point.Footnote 26
Ioh. IV 200–204.
The gaps in the version of the Iohannis transmitted by the Codex Trivultianus 686 have at times provided a basis for further theories. Thus, some scholars, in an attempt to classify Gorippus among the authors who viewed Justinian’s activities in Africa unfavourably – in line with the tone of the final passages of Book IV of Procopius’s History of the Wars – noted the brevity and restraint of the praise directed at the emperor.Footnote 27 Maria Cesa, on the other hand, attempted to explain this silence by highlighting the subject of the poem – the praises of John Troglita – rather than Belisarius or Justinian himself.Footnote 28 Yet, in the Trivultianus manuscript, a single folio is missing, containing approximately 62 lines following line 22 of Book I. In the latest and most widely-used edition of Diggle–Goodyear, the two lines following this one are simply left blank, and a lacuna is noted in the critical apparatus. Nonetheless, Maria Assunta Vinchesi, in her edition of Book I (Reference Vinchesi1983), supplemented this gap with lines probably adapted from this very passage by Giovanni De Bonis in his own literary output. This reconstruction reveals that what is missing from the manuscript at this point is precisely the elegant eulogy of the Emperor Justinian.
This article provides an overview of selected issues concerning the text of the Iohannis, which is in dire need of a new critical edition. Such an undertaking must take into account the existing tradition of editions and commentaries, the considerable progress made in Gorippus scholarship in recent years, and a thorough revision of earlier interventions in the poet’s text. It is no coincidence that many of the changes proposed by the editors of individual books of the work consist precisely in reverting to manuscript variants or to Mazzucchelli’s conservative proposals.Footnote 29 However, a literal reading of the manuscript text (which should involve not only seasoned philologists but also specialists in Gothic paleography) is not always viable. In my remarks so far, so as not to multiply the examples, I have also not mentioned the problem of metre. After all, the Iohannis is, above all, a poetic work, a heroic epic of the historical variety, written in regular dactylic hexameter – a fact often overlooked by scholars primarily interested in its potential as a first-rate historical source. Many of the conjectures and emendations proposed by commentators may therefore convey the intended meaning. Still, they do not fit the metre, which means that they certainly do not reflect the words of the poem’s author. Finally, there remains a fundamental problem – the extent to which the Codex Trivultianus 686, transcribed by Giovanni De Bonis, faithfully reflects the content of the late-antique poem, and thus, ultimately, which source researchers should in fact rely upon.
Still, the main aim of this article was not merely to draw attention to the complexity of the editorial history of the Iohannis, nor, even less so, to articulate expectations regarding a new critical edition (or editions) of this poem, but above all to reflect on the subject matter of a historian of antiquity. It would be a truism to state that a classicist should consult the sources, which, concerning most ancient works, are understood to be the original text in a critical edition. The case of the Iohannis demonstrates how misleading it can be to rely solely on the decisions of a single editor, whose work, much like that of a translator, is substantially an interpretation of the text as it has been transmitted over time. To researchers who do not engage in the editing of textual sources daily, the spelling of a historical figure’s name or a place name may seem a marginal issue. Yet, the example of Urbicius demonstrates that contemporary historical reconstruction is sometimes based not on the text as it appears in the manuscript, but precisely on the conjectures of its editors and commentators. The example of Cellas Vatari, in turn, illustrates the broader phenomenon of one source’s subordination to another. Philological questions concerning the textual reconstruction of the epic by the sixth-century poet are therefore of paramount importance to the historian, and indeed to any scholar wishing to utilise this poem as a vital source for the history of Africa during the decline of Vandal rule and the establishment of the Byzantine regime.
Acknowledgement
I am deeply grateful to Antoni Czachor from the Manuscripts Department of the Jagiellonian Library in Cracow for his expert guidance on my study of the digitised version of the Matritensis 686 manuscript. Furthermore, I would like to thank prof. Sławomir Sprawski and the anonymous reviewers for their valuable comments, which greatly improved the structure and overall coherence of this paper. I express my gratitude also to the management of the Trivulziana Library for granting permission to use the scans of the Trivultianus 686 manuscript.
Open access
