Tedros Abraha, ed., tr., Mäṣḥafä Sawiros zä-Esmunayn: la versione geʿez del Kitāb al-īḍāḥ di Severo di Ašmūnayn (X/XI sec.): Omelie I-III. Introduzione, edizione del testo etiopico con traduzione italiana annotata, Patrologia Orientalis, 56/2 (247) (Turnhout: Brepols, 2017). 216 pp. (133–348)

1.

• prima edizione dell’opera*

 There were so far no printed editions of the work; the first is the one here reviewed

2.

• precedente edizione critica del testo*

                                                                       NO

• precedente edizione scientifica di riferimento*

                                                                       NO

3.

• testo dell’edizione critica e ragioni della scelta ecdotica*

The edition marks an essential but also obvious progress, in that it provides the first edition of the first part (three out of twelve homilies) of a large theological work arranged in the form of a dialogue between a master and a disciple. The work is of some importance for the literary and theological tradition of Ethiopian and Eritrean Christianity and the edition puts at disposal of the scholars the first portion of a long text that is often quoted in the exegetical tradition and was not accessible before in any form, unless directly from the manuscripts.

4.

• criteri di edizione*

The edition reflects in some aspects the standard (or lack of standard) of other editions of Gǝʿǝz texts, but it has also some progressive aspects that must be duly underlined as a positive development in the field. (For an overview of current practices in editing Gǝʿǝz texts, see Bausi 2016, 2020a, 2020b.) The most important and progressive feature of this edition is the abandonment of the base-manuscript method, followed by the author in other contributions, in favour of an edition of explicitly reconstructive character. The reconstructive character concerns both the formal and the textual aspect. The editor, in fact, aims at achieving the Gǝʿǝz text, which is a translation from the Arabic (see below concerning the question of the dating of the translation), and he actually assumes, without declaring it, the Arabic text as the virtual archetype, as far as this is attainable (see p. [23]/155: ‘L’analisi dell’attestazione manoscritta gǝʿǝz del KĪ, porterebbe ad escludere che ci siano state più traduzioni dall’arabo. I codici studiati in questa sede discendono da una sola versione, in seguito “trattata” dai copisti, soprattutto per quanto riguarda la grammatica ma anche su qualche punto teologico […] Lo sforzo primario di questa edizione consiste nel cercare di restituire un testo gǝ’ǝz più vicino possibile alla matrice araba’). The considerable risk in this lack of distinction is that of calquing the edited Gǝʿǝz text upon the Arabic model, also when agreement with the Arabic is the result of a secondary development, either in a branch or in a single witness of the tradition. Ideally, the reconstruction should instead aim at the earliest version of the Gǝʿǝz text, not at the Arabic model used by the Gǝʿǝz translator nor any other Arabic text.

This confusion surfaces in other passages of the Introduction, where the authors (e.g. on p. [15]/147) appears to use interchangeably the KĪ (abbreviation of Kitāb al-īḍāḥ, title of the Arabic work) and KH (Kǝbrä Haymanot, one of the titles of the Mäṣḥafä Sawiros) abbreviations: the Qerǝllos is ‘[i]l testo teologico di riferimento più autorevole e influente per la Chiesa Ortodossa Etio-Eritrea […] di cui si riscontrano molte impronte [sic] e citazioni anche nel KĪ’, so that it seems that the Gǝʿǝz Qerǝllos was used directly in the Arabic. The same on p. [7]/139: ‘Il KH è stato esplicitamente concepito come difesa della fede “Ortodossa Giacobita”’, an intention that can only be attributed to the original Arabic text.

The editor intends to provide a normalized orthography according to the spelling as canonized in August Dillmann’s Lexicon (Dillmann 1865), which has essentially an etymological base and distinguishes roots phonemes which in the course of time the scribal practice has merged, following the phonetic neutralization that took place in Amharic (see ibid.: ‘Laddove non ci sono implicazioni per l’integrità del testo e per la traduzione italiana […], l’ortografia è stata uniformata alle forme standard’). This attention to the formal aspect is extended to the morphology, for which the editor intends to preserve calques and solecisms which are due to the influence of the model (pp. [23–24]/155–156: ‘Il testo nel corpo rappresenta le lezioni più vicine all’originale arabo. Ad esempio, a fronte di varianti quali Spirito Santo al femminile ed al maschile, la precedenza è stata assegnata alla forma femminile perché corrisponde all’originale. Il cambiamento al maschile è un intervento intenzionale del copista abissino per il quale “il genere” dello Spirito Santo non è femminile’). In some cases, inconsistencies are not explained, neither in the apparatus nor in the commentary: see for example (p. [32]/164.4): ‘ዐሠርቱ፡ ወክልኤቱ፡’, ‘twelve’, nominative, but in agreement with ‘ድርሳናተ፡’, ‘homilies’, accusative, in the text (unless this is the result of spelling out the numerals).

The text is arranged in paragraphs of varying length, numbered separately for each of the three homilies (55 §§ in the first homily, pp. [32–79]/164–211; 92 §§ in the second, pp. [80–167]/212–299; 40 §§ in the third, pp. [168–211]/300–343), which is of some advantage for internal references. Punctuation also follows the Arabic tradition, which serves as reference (p. 23/155: ‘L’arbitrarietà della punteggiatura nei manoscritti gǝ’ǝz ha reso necessario il ricorso ai manoscritti arabi e alle due edizioni arabe per la scansione del testo’). In some cases, however, the perfect parallelism between text and translation is not respected (see for example the lines 13–14 on [40]/172, at the end of § 1 in the edition, which correspond to line 15 in the translation on p. [41]/173, at the beginning of § 12).

In one case at least, for the sake of structuring, the editor has introduced an element that does not belong to the tradition: the word መቅድም። (p. [32]/164, l. 1) at the very beginning, translated ‘Proemio’, does not belong to the witnesses nor does it appear in the tradition, and is an editorial addition which is not marked as such in the apparatus. It can be considered an homage to a nineteenth-century tradition, where additions of this kind were not unusual; yet, they should not appear in a modern scholarly edition. Another matter is the insertion of subtitles for sections, in the translation only, which reasons of clarity and easier access to the content can justify.

5.

• presenza di approfondimenti filologici forniti, ad esempio, nella nota al testo (o in altri scritti come introduzione o prefazione)

Yes. The edition is provided with an extensive Introduction (pp. [7–25]/139–157) that deals with several aspects of the work: the author; the Arabic model and its textual tradition (manuscripts and the two main printed editions), of which the Gǝʿǝz text edited is a translation; the genesis and the structure of the Gǝʿǝz text (see below concerning the issue of the date of the translation); the textual tradition of the Gǝʿǝz text, and its witnesses, used (nine manuscripts) and not used (eight manuscripts) in the edition, and the reasons why; the methods of the edition and of the translation; the sigla used in the edition and in the commentary to the translation; a bibliography.

For a more detailed consideration of the recensio, see the section of additional remarks below.

6.

• fonti archivistiche consultate dal curatore*

oppure

• biblioteche o luoghi presso i quali il curatore ha consultato i testimoni usati per l’edizione*

There is no consistent statement concerning which manuscripts were consulted directly and which in digital, microfilm, or other surrogate forms, but on the basis of the author’s remarks and acknowledgements, the following appears a realistic estimation.

Of the nine manuscripts used in the edition (ABCDGHSTV), the editor examined directly two manuscripts (G, in Eritrea, Asmara, and V, in Città del Vaticano), and probably a third one (S, in Ethiopia, Addis Ababa). Of the others, there are some which are little or not accessible at all in their physical form, and it is likely that the editor, tacitly, examined digital or analogical surrogates. Of the further eight manuscripts not used in the edition, the editor examined directly two manuscripts (Israel, Jerusalem, Ethiopian Archbishopric, JE 342E, and Eritrea, Asmara, Cathedral Church of Maryam Ṣǝyon).

7.

• tipo di apparato/i*: genetico / evolutivo / varianti alternative

• posizione dell’apparato*: a piè di pagina / a fondo volume

The edition has a rich apparatus of alternative variants. The apparatus is placed below the text, on the even pages, while the translation, with footnotes of commentary, holds the odd pages. The internal margins of text and translation have line numbers (each fifth line has the number). The outer margins of the even pages indicate folia and columns of manuscripts G and V, signalled with an asterisk in the text.

The editor does not discuss the ratio of the apparatus, which appears to be a negative one. The reference to the variant is not given according to line numbers, but with footnotes references in the text, numbered separately for each page, and, in case a variant includes more than one word, with additional signs in the text used to mark the beginning and end of the passage. A positive or a mixed apparatus would have been more convenient, as appears, for example from this case (p. [34]/166, n. 18): ‘ABDGHST ፍካሬ፡’, with a variant apparently supported only by CV. The system of additional signs is not always clear as well: for example, in this case (p. [32]/164, n. 1): ‘C omette la sezione contenuta fra questo asterisco e il seguente’, it is not clear where the next asterisk is. The next asterisk indicates the change of column in MS G (‘G 3b’, in the margin), at p. [32]/164.5 of § 2, which continues until p. [34]/166.2; a footnote to the translation (p. [33]/167) indicates instead: ‘C E om. §§ 1–2’, with E being an Arabic witness. In the end, it is impossible to understand how long is the omission in C. Since the apparatus is negative, this has further consequences. Moreover, since there is no apparatus of available witnesses for each page or paragraphs, and MS B is discontinuous, it is very difficult, even by subtraction, to catch the actual textual basis of each reading: see for example the apparatus on p. [74]/206.20, n. 15: ‘B des.’ (also transl. on p. [75]/207, n. 136: ‘B termina in questo punto’) and p. [82]/214.28, n. 22: ‘B inc.’ (also transl. on p. [85]/217, n. 12: ‘Inizio di B dopo la brusca interruzione nel primo capitolo, § 51’), with several pages between § I.51 and § II.5 where the reader could presuppose that tacitly MS B supports all readings.

The syntax of the apparatus is as follows: after the reference number (corresponding to a footnote) follow the siglum or sigla of the witnesses which have a variant reading, and then the variant reading; if more variants for the same reading are present, a single space after the first variant separates the following siglum or sigla, after which in turn the second variant follows, for example (p. [32]/164): ‘22 A T ያመ” B ያሙቶሙ’. This witness-variant sequence does not appear to be convenient and is against the commonly used rules (see for example the variant-siglum sequence in the Neste-Aland New Testament editions (Aland et al. 2012) or in Rahlfs’s LXX (Rahlfs and Hanhart 2006), and in many other standard editions). Moreover, a longer space between variants to the same reading in the text, if not also a punctuation mark (just a comma), would help much. The same string would then appear in this form: 22 ያመ” A T,   ያሙቶሙ B. Numbered notes are separated by an em-dash (—). Also to be noted, not only in the apparatus, but in the whole book, is the practice of not putting the so-called word divider (፡) after the last (or an isolated) Gǝʿǝz word (if it were used, we would have had ‘ያሙቶሙ፡ B’). If this sign in origin was in fact used to separate words, it is true that the use in manuscripts prescribes that also the last word has it, so that there is no reason not to put it after every Gǝʿǝz word. The economy and clarity of the apparatus is not ideal: for example, in this case (p. [32]/164, n. 20), for the reading ወያበጥል፡ (translated ‘rende vano’), the apparatus has: ‘A T add. ቃሎ S ወይበጥል፡ ቃሉ’; the translation also has a corresponding footnote (p. [33]/165, n. 5): ‘A S add.: la parola di’; but ‘la parola di’ is present not only in A (ቃሎ፡) and S (ቃሉ፡), but also in T, which has the same reading as A. Moreover, the addition ቃሎ፡ (‘la parola di’) with the resulting text ‘rende vana la parola di colui’ is found not only in A T, but also in S, which has ‘la parola di’, albeit in a different case (with nominative suffix pronoun ቃሉ፡ instead of accusative suffix pronoun ቃሎ፡), as required by the concurring variant ይበጥል፡ ‘è vano’, instead of ያበጥል፡ ‘rende vano’. Punctuation is not noted, unless some sporadic cases, the reason for which is not clear, see for example this erratical occurrence (p. [40]/172, n. 1): ‘1 T ። ። ።’.

8.

• presenza di note / commento al testo*

• tipo di note / commento*

• posizione delle note / commento*

An annotated translation accompanies the edition, with explanation of some passages and in particular with consideration of the Arabic tradition (for which see below), which is given in extenso in not a few cases (there are 148, 215, and 340 footnotes for the three homilies). A disturbing aspect, however, is that sigla used for the Gǝʿǝz witnesses in the apparatus and in the notes to the translation, and sigla used for Arabic witnesses (only in the notes to the translation) have exactly the same aspect: ABCDGHSTV are used for the Gǝʿǝz witnesses and EMPQY are used for the Arabic witnesses; the use of additional signs or exponents (for example, Ear) would have made the notation more user-friendly.

There are patent inconsistencies between the data provided by the commentary and the apparatus: for example, the apparatus (p. [48]/180, n. 5) has ‘CDHV add. መሐሪ’, but the corresponding translation (p. [49]/181, n. 44) has ‘V add. “misericordioso”’, while in fact, according to the apparatus, also CDH share the reading ‘misericordioso’ (መሐሪ፡); similar case with other MSS on p. [48]/180, nn. 22–24 and p. [49]/181, n. 49.

9.

• presenza di altri scritti e loro descrizione/utilità/specificità*

The edition is provided with a substantial Introduction, where literary, philological, and linguistic phenomena are considered. See here more in detail (italicized script reflects the use of the editor): p. [7]/139: ‘Ringraziamenti’; p. [8]/140: ‘Accenni prosopografici sull’autore’; pp. [8–9]/140–141: ‘Kitāb al-īḍāḥ (KĪ), “Il Libro della delucidazione (della fede)”’; pp. [9–10]/141–142: ‘Le fonti del Kitāb al-īḍāh: Fonti dichiarate; Fonti implicite’; pp. [11–12]/143–144: ‘Edizioni del Kitāb al-īḍāḥ’; p. [12]/144: ‘Tradizione manoscritta del KĪ’; pp. [12–14]/144–146: ‘Genesi e Sitz im Leben della versione gǝʾǝz del KĪ’; p. [14]/146: ‘Struttura del testo’ (all in all, five lines); pp. [15–16]/147–148: ‘Influsso del KĪ nella Chiesa Ortodossa di Etiopia e di Eritrea’; pp. [16–19]/148–151: ‘Manoscritti gǝʾǝz collazionati per la presente edizione’; pp. [19–20]/151–152: ‘Codici parzialmente collazionati e codici esclusi dall’apparato critico’; pp. [21–23]/153–156: ‘Rapporto fra il KĪ e la sua versione gǝʾǝz’; pp. [23–24]/155–156: ‘Metodo di edizione’; p. [24]/156: ‘Metodo di traduzione’; p. [25]/157: ‘Sigle dei manoscritti e dell’edizione araba; Sigle dei manoscritti gǝʾǝz; Abbreviazioni e sigle usate nell’apparato critico e nelle note a piè di pagina della traduzione italiana’; pp. [27–31]/159–163: ‘Bibliografia’, which also includes some Abbreviazioni for more frequently quoted works; text with apparatus (on even pages) and annotated translation follows on pp. [32–211]/164–343; pp. [213–216]/345–348: ‘Sommario’.

10.

• ulteriori eventuali considerazioni

Recensio of the Gǝʿǝz tradition

The editor does not always provide data on the manuscript witnesses in a clear form, which makes difficult to follow some of his considerations. A real recensio is missing, so also any comprehensive hypothesis on the relationship of the witnesses, even though some principles are discussed and applied, like the eliminatio codicum descriptorum. Since the editor presents his edition as a reconstructive one, one has then to assume that he evaluated each single reading with a look at the proximity of the Gǝʿǝz text to the Arabic, although no general conclusion is drawn from this evidence, even ex post, aside from a vague evaluation of the value and tenor of the single witnesses. Yet, the sequence of texts in some multiple-text manuscripts could have provided indications for a first hypothesis.

The recensio thus simply consists of two lists of used and non-used manuscripts (see above). In two previous contributions the editor had already provided preliminary lists. The first (see Tedros Abraha 2006, 628) listed fifteen manuscripts, that is, all manuscripts mentioned in the Introduction aside from the manuscript in Qoma Fasilädas and the manuscript UNESCO from Dima Giyorgis, plus an Amharic translation (without any details on its whereabouts) that is lacking in the list provided in the edition. The second (Tedros Abraha 2010, 640b–641a) adds the MS Qoma Fasiladas and again the Amharic version without any further detail. Little details are also found in a third contribution (Tedros Abraha 2017b, 427–430).

The editor finally seals the concise presentation of the Gǝʿǝz manuscripts with a remark that an editor should never voice (p. [20]/152): ‘Ritengo che eventuali scoperte di altri codici non porterà [sic] novità che cambieranno sostanzialmente i dati sulla traduzione dall’arabo’. Honestly, who can say that?

Here follows the list of the Gǝʿǝz manuscripts, used and not used, mentioned in the Introduction.

(A) Paris, Bibliothèque nationale de France, Éthiopien, d’Abbadie 76 (see Chaîne 1912, 50–51; Conti Rossini 1914, 152, no. 130); 129 ff., 18th (Chaîne) or 18th/19th (Conti Rossini) c.; contains on ff. 1r ff. the Mäṣḥafä Sawiros; on ff. 121r ff. a Homily of St John Chrysostomos on the annunciation of the angel Gabriel to Mary; on ff. 126v ff. Praises in honour of the Virgin composed by the monk Yǝsḥaq. For this manuscript the editor notes the retention of some elements of contents. This feature would separate this manuscript, along with the others with the same retention, from all the other innovating manuscripts; these latter would depend from a sub-archetpye, even though the editor does not argue in these terms. In his words (p. [16]/148), ‘La pneumatologia, in particolare la processione dello Spirito Santo dal Padre e dal Figlio, espressa nelle copie più antiche del decimo trattato del KĪ [Arabic Vorlage of the Mäṣḥafä Sawiros], mantenuta in A è un elemento che permette di collocarlo fra le prime versioni del KH [Kǝbrä Haymanot, an altenative name of the Mäṣḥafä Sawiros]’. What does ‘prime versioni’ mean? were there more recensions of this work, in Arabic, and in Gǝʿǝz as well? since from other statements this does not seem the case, and all witnesses attest to one Gǝʿǝz translation only, this remark is quite obscure and points to a lack of appropriate terminology. As for the other statement, ‘Quanto alla natura del testo, si notano alcune aggiunte rispetto al resto della tradizione testuale. La morfologia e la sintassi di questo codice riflette [sic] una fase del gǝ’ǝz post-medievale, quando si abbandonano alcune delle forme tradizionali’, as long as the date of origin of the translation is not ascertained, if these forms could have belonged or not to the first translation and to the primary text of the work remains unclear.

(B) London, British Library, Orient. 815 (see Wright 1877, 226–227, no. 339); 121 ff., second half of the 17th c.; contains on ff. 26ra–115 the Mäṣḥafä Sawiros, followed (from the catalogue, editore silente) on f. 116r by a theological treatise in Amharic and on f. 118r by a short hymn to St Täklä Haymanot; the initial leaves, instead, contain chronological texts and tables and are due to a different hand of the 19th c. so that this manuscript results to be a composite. For this manuscript the editor remarks (p. [16]/148) that it is

interessante per la sua vicinanza alle versioni più prossime alla matrice araba e per alcuni aspetti paleografici. Il primo trattato è inficiato da ampie lacune; termina bruscamente al f. 34vb ben prima della conclusione del trattato per poi continuare ex abrupto con parte del secondo trattato. Si verifica la medesima situazione alla fine del trattato secondo ove mancano alcuni fogli. Ricorrono ampie lacune nel terzo trattato che peraltro si interrompe ex abrupto. All’origine delle lacune si può ipotizzare che le condizioni fisiche del codice fossero piuttosto precarie e che i fogli mancanti siano caduti e che poi siano andati persi o dislocati.

As appears, these observations do not approach the obvious question whether all these lacunae are due to material loss in this manuscript (which should be evident) or in any of its ancestors, that is, in the previous tradition. An answer provides the old catalogue itself (Wright 1877, 226), qualifying the manuscript as ‘Defective in various places, and apparently copied from a disarranged Ms.’. Concerning the relationship, the manuscript is said to have (p. [17]/149) some interesting phenomena, which appear to be homoiuteleta common both with MS G and with MS C (‘Meritano attenzione alcuni fenomeni: B G salto dal primo መወልዲት “ostetrica” al seguente (quindi B e G risultano essere uniti dalle medesime omissioni dovute ad una svista; gli omoioteleuti (soprattutto nel cap. II), identici a quelli del manoscritto C; le varianti fantasiose, e.g. ብሔረ per ኄር (cap. III)’, etc.), whereas of the latter MS C the editor says (ibid.) that ‘Spesso, nelle omissioni, concorda con il codice G’. So, there are juxtaposed observations without any evaluation of how to interpret these phenomena in a rational way. Obviously, taken as the editor poses it, one does not know what to do of the shared homoioteleuta of MSS B C and B G, which are obviously polygenetic.

(C) London, British Library, Orient. 773 (see Wright 1877, 224–225, no. 337); 219 ff., AD 1682–1706; contains on ff. 2r–147v the Mäshafä Sawiros; on ff. 148r ff. Concerning the Prayer of Longinos; on ff. 150v ff. the Life of the Syrian monk Barsauma; on ff. 164r ff. the Homily of Cyril of Jerusalem on St Simeon receiving Our Lord; on ff. 167r ff. the Life of Severos of Antiochia, followed by its own colophon (f. 218v). The editor remarks that MS G ‘Spesso, nelle omissioni, concorda con il codice G’.

(D) EMML (= Addis Ababa and Collegeville, MN, Hill Monastic Manuscript Library, Ethiopian Manuscript Microfilm Library, microfilmed from Ankobärr, Church of Miṭaq Täklä Haymanot) no. 2145 (a microfilm copy: see Getatchew Haile and Macomber 1982, 256–257); 123 ff., 20th c.; contains only the Mäṣḥafä Sawiros. The editor observes (p. [17]/149): ‘Secondo i catalogatori il codice apparterrebbe al XIX sec. ma la grammatica (e.g. recensisce/aggiorna gli accusativi con ቦ፤ አልቦ፤ አኮ), l’intervento editoriale che nell’intenzione del copista, al termine di una sezione, scansionerebbe il testo tramite una serie di punti a forma di croce aksumita suggeriscono che è più verosimile una data di copiatura ancora più bassa’. In this case, the editor makes the assumption that the tradition is regularly subjected to a process of updating and correction or pseudo-correction, so that every copying would reflect a determinable state of update, and there would be a consistent and regular correspondence between form and age. This assumption, in fact, is only a hypothesis that is not substantiated by any thorough study. If this is possible, it cannot yet be assumed as a regular process and is no serious basis for the evaluation of a single manuscript. The amount of errors in this manuscript raises some moralistic concern in the editor (ibid.): ‘Il testo è inficiato da tante, troppe, omissioni (imputabili ad un copista indolente?), sviste, errori di copiatura (e.g. ወሞተ፡ ዮሴፍ፡ ወኮነ፡ ፈርዖን፡ ያአምሮ፡ ወተሰምየ፡ ፈርዖን፡ ዘየአምሮ። “E Giuseppe morì e vi fu un faraone (che) lo conosceva e fu preposto un faraone che lo conosceva” (III, 4–5). Pur riconoscendo la fatica di copiare un libro così lungo e ripetitivo negli argomenti, non si può sottacere che il codice è stato trascritto da un copista mediocre’. Also note (ibid.) that the translation of ተሰምየ፡ with ‘fu preposto’ is wrong: ‘fu preposto’ would be ተሠይመ፡ (täśäymä), which is in fact frequently confused in the manuscript tradition with ተሰምየ፡ (täsämyä). What more surprises, however, is that, in the end (ibid.), ‘Il valore di questo codice per la storia della trasmissione del KH può consistere nel fatto che rafforza l’attestazione di un gruppo di manoscritti con un certo tipo di testo, e.g., con C sembra essere un codice gemello, anche se in qualche misura, “gemello sfigurato”. Non risultano invece lezioni utili per fare confronti con la matrice araba’. Neither which manuscripts this group consists of, nor what is the sense of posing the question of the relationship with the Arabic model is clear.

(G) Asmara, Gäǧǧärät, convento dei Frati Minori Cappuccini, MS 45, 170 ff., 18th c.; the editor, who seems to have examined this manuscript on the spot, does not report anything else on the contents of this uncatalogued manuscript, so that we assume it contains only the Mäṣḥafä Sawiros. Aside from terminological approximation (p. [18]/150, ‘coperchio’ for ‘piatto’, ‘cover’), the editor provides an important element: the manuscript has a retention that was already noted for MS A, which would determine (editore silente) a sub-archetype from which all other manuscripts descend (ibid.): ‘G è uno dei pochi testimoni che preserva la dottrina della doppia processione dello Spirito Santo del KĪ’: yet, even in noting this, one has to remark the attribution of this conservative feature to the Arabic model (‘KĪ’), while it should be ascribed to the early Gǝʿǝz version.

(H) London, Britsh Library, Orient. 738 (see Wright 1877, 204–205, no. 314); 180 ff., first half of the 18th c.; contains on ff. 5v-140v the Commentary of St John Chrysostomos on the Letter to the Hebrews; on ff. 141r–180r the first five Homelies of the Mäṣḥafä Sawiros, the last one imperfect). According to the editor (p. [18]/150), it is a ‘recensione abbastanza accurata di testi precedenti’, but we do not know which. Among the linguistic features, there is the later, nominative construction with the existential ቦ፡ and አልቦ፡ instead of the accusative.

(S) EMML no. 1195 (microfilmed from Addis Ababa, Holy Trinity Cathedral; see Getatchew Haile and Macomber 1979, 152–153); 269 pp., AD 1967; contains on pp. 249–261 a Homily of St John Chrysostomos on the annunciation of the angel Gabriel to Mary; on pp. 261–268 Praises in honour of the Virgin St John Chrysostomos, which could be attributed to a monk Yǝsḥaq. According to the editor, this is a ‘buon testimone della versione più vicina al testo arabo’. The editor remarks that the manuscript has a colophon dated 5 Ṭǝrr 1959 = 13 January 1967, but does not note any other detail. See below on the question of the dating of the Gǝʿǝz translation.

(T) Ṭānāsee 65 = Dabra Māryām 7 (Berlin, Staatsbibliothek Preußischer Kulturbesitz, Orientabteilung, microfilmed from Lake Ṭana, Däbrä Maryam; see Hammerschmidt 1977, 61–63); 180 ff., 18th c.; contains on ff. 2ra–36vb the Gospel of Mark, in a first block; in a second block, by the same hand as the first block, on ff. 37ra–163va the Mäṣḥafä Sawiros; and on ff. 163vb–180rc the Mäzgäbä Haymanot (‘Treasure of the faith’). The editor assumes that this manuscript is strictly related to MS A based on decisive evidence, that besides retention (which would not demonstrate any relationship), also includes innovations (pp. [18–19]/150–151), ‘molto vicino al manoscritto A con cui, oltre alle strette affinità testuali condivide le omissioni, le aggiunte e gli errori, T appartiene alla stessa famiglia di manoscritti che preservano la lezione: “la processione dello Spirito Santo procede da loto due, ossia, dal Padre e dal Figlio”’.

(V) Città del Vaticano, Biblioteca Apostolica Vaticana, Cerulli etiopici 122 (see Raineri 2004, 88–89); 209 ff., 19th c.; contains on ff. 5 ff. the Qalemǝnṭos (‘Book of Clement’); on ff. 77v–163r the Mäṣḥafä Sawiros; on ff. 164r–207r the Book of Maccabees. This manuscript (p. [19]/151) ‘conserva la versione madre della versione gǝʿǝz’, which, in the editor’s terminology, seems to mean that the manuscript is a correct witness.

Further manuscripts which were not used in the edition are summarily described, with some explanation why they were not used. The editor states (p. [19]/151), ‘Dopo un ventennale vaglio di tutti i manoscritti ai quali è stato possibile accedere, sono state escluse le copie delle copie, e i testimoni privi di elementi utili alla ricostruzione della matrice gǝʾǝz’. How these aspects, which are reasonable, actually work in determining the selection of the manuscripts, does not appear clearly. These manuscripts (ibid.) ‘[a]vrebbero reso inultimente [sic] ingombrante l’apparato critico. Nei codici esclusi, spesso si riscontrano varianti consistenti in palesi modifiche del textus receptus […] Al centro dell’interesse si pone il testo e non i manoscritti’. This statement is alarming and contradictory: if the editor pursues the reconstruction of the earliest translation (‘Lo sforzo primario di questa edizione consiste nel cercare di restituire un testo gǝ’ǝz più vicino possibile alla matrice araba’), the textus receptus does not match this approach, since, by definition, the textus receptus is a standardized textual form emerged in the course of the tradition. The editor uses this concept in a way that is not understandable.

(1) London, British Library, Orient. 771 (see Wright 1877, 225–226, no. 338); 139 ff., AD 1721–1730; contains exactly the same texts as MS C (Orient. 773): on ff. 1r ff. the Mäṣḥafä Sawiros; on ff. 92r ff. Concerning the Prayer of Longinos; on ff. 94r ff. the Life of the Syrian monk Barsauma; on ff. 103r ff. the Homily of Cyril of Jerusalem on St Simeon receiving Our Lord; on ff. 105v ff. the Life of Severos of Antiochia. The manuscript is a copy of MS C, as noted by Wright 1877, 226 (‘This volume is evidently a copy taken from Orient. 77.’) and accepted by the editor, who has rightly eliminated this witness from the apparatus.

(2) London, British Library, Orient. 772 (see Wright 1877, 220, no. 330); 128 ff., first half of the 18th c.; contains on ff. 3r ff. the Ancoratus of Epiphanius; on ff. 47r ff. the Mäṣḥafä Sawiros. For this manuscript the editor does not provide any reason for its elimination.

(3) EMML no. 2149 (microfilmed from Tägulä and Bulga, Church of Miṭaq Täklä Haymanot; see Getatchew Haile and Macomber 1982, 260); AD 1779–1800; contains on ff. 2r–146v the Mäṣḥafä Sawiros. Also for this manuscript the editor does not provide any reason for its elimination.

(4) Paris, Bibliothèque nationale de France, Éthiopien, d’Abbadie 125 (see Chaîne 1912, 80; Conti Rossini 1914, 152–153, no. 131); 209 ff., 18th c.; contains on ff. 1v ff. the Mäṣḥafä Sawiros; on ff. 79r the Homily of St John Chrysostomos on the incarnation; on ff. 82v ff. Praises in honour of the Virgin composed by the monk Yǝsḥaq; on ff. 85r ff. the Aksimaros (Hexaemeron) of Epiphanius; on ff. 98r the Gädlä Addam (‘Life of Adam’); on ff. 152r ff. the Ancoratus of Epiphanius; on ff. 191v the Mätlo sǝrʿat (prorogation of the ordinance written after the deposition of Nestorius); on ff. 193r ff. the Ṭǝntä Haymanot (‘Beginning of the faith’). For this manuscript the editor does not provide any reason for its elimination. The sequence clearly points to a relationship with MS A, but the argument is not evoked by the editor.

(5) Jerusalem, Ethiopian Archbishopric, JE 342E (see Isaac 1984–1986, 69); 17th or 18th c. This manuscript has also the theological retention concerning the pneumatology, like manuscripts A and T. Also for this manuscript, which he could access and evaluate, the editor does not provide any reason for its elimination.

(6) Asmara, Cathedral Church of Maryam Ṣǝyon, no shelf mark; AD 1895/1986. Also for this manuscript, which he could access and evaluate, the editor does not provide any reason for its elimination.

(7) Bägemdǝr, Qoma Fasilädäs, no shelf mark (see Wion 2002, 287–288; unfortunately, no figure corresponds in the publication to the ‘photo no. 1’ quoted on p. 287; later restudied in Wion 2012, 329–333); 15th c.; contains the Mäṣḥafä Sawiros; ends on f. 115 with ‘some prayers for Mary composed by the monk Yishaq, which must be read on 29 Magabit [sic]’ (Wion 2002, 287–288): with good reasons, Anaïs Wion has pointed out that this manuscript, that could not be examined more in detail but she dated on palaeographic grounds to the fifteenth century, could represent ‘the “canonical” composition’ of the Mäṣḥafä Sawiros. The editor is extremely laconic concerning this manuscript, which is not accessible at the moment. In the list (p. [20]/152) he only provides a one-line mention and a reference to the study by Wion, without any other detail or comment, while he had already briefly mentioned that he could not access it (p. [14]146, n. 1: ‘Dopo ripetuti tentativi di accedere al codice di Qoma Fasilädäs, a malincuore, mi sono rassegnato a procedere con la pubblicazione del presente lavoro’). For further aspects related to the dating of the manuscript, and consequently of the Mäṣḥafä Sawiros, see below.

(8) Goǧǧam, Dima Giyorgis, MS 46 = UNESCO 10.74: 238 pp., 18th c. (microfilm with a typewritten unpublished description; the film has been digitised, including a preceding typwritten description, by the project EMIP: Ethiopian Manuscript Imaging Project, directed in 2005–2011 by Steve Delamarter); contains on ff. i–vi (numbering of EMIP) the Names and teaching of the 72 disciples of Christ; on ff. 1 ff. the Mäṣḥafä Sawiros starts on ff. 1 followed by the Sǝrʿatä Betä krǝstiyan (‘Order of the Church’) attributed St John Chrysostomos. The editor does not mention the digitisation and does not provide any reason for the elimination of this manuscript.

Dating of the Gǝʿǝz version

The dating of the work is of course important in itself, as an essential element of cultural history. However, the dating of the work also has technical philological implications for all those cases, mentioned by the author, where the tradition shows conflicting evidence of more or less archaic forms. These forms should be restituted or not according to the hypothesis on the date of the work, in case, besides contents, also aspects of form are considered.

The editor mentions on various occasions evidence that points to a translation of the Mäṣḥafä Sawiros at the latest from to the fifteenth century, as the palaeography of the MS Qoma Fasilädäs demonstrates (see above). Yet, repeatedly and without any firm ground, he proposes at the same time the hypothesis that the work could have been translated in the seventeenth century (p. [14]/146):

La maggior parte dei testimoni manoscritti del Mäṣḥafä Sawiros descritti in questa sede, appartengono al periodo e all’area geografica gondarini. L’epoca gondarina iniziò con il trasferimento e stabilizzazione della corte imperiale etiopica a Gondar. Ciò avvenne nel quarto anno del regno di Fasil nel 1635–36, e si protrasse fino al regno di Tewodros II (1855–1868). Si tratta di un periodo durante il quale, insieme alle ben documentate turbolenze politiche e religiose, l’Etiopia conobbe una grande effervescenza culturale che giustifica l’ipotesi, se non di prima traduzione del Mäṣḥafä Sawiros, almeno di una sua ampia copiatura e circolazione.

The same position the editor takes in an article (never mentioned in the edition) published the same year as the edition (see Tedros Abraha 2017b, 427–430, esp. 428), suggesting that the Mäṣḥafä Sawiros ‘was probably translated around the end’ of the seventeenth century, even though the editor himself mentions here that the pictures of the MS Qoma Fasilädäs he was able to see reconfirm a dating of the manuscript from the fifteenth century. On this ground, one does not understand the reiterated doubts or uncertainty concerning the dating of the work, which appears repetition of outdated hypotheses presented in his first contribution (Tedros Abraha 2006), when, understandably, the evidence of the MS Qoma Fasilädäs could not yet be considered (Tedros Abraha 2017b, 428): ‘The translation and/or the spread of the Mäṣḥafä Sawiros would fit well into the religious scenario of Ethiopia in the aftermath of the failed Catholic mission (1555–1634). The book would have supplied to the local Church the arguments to counter the “opponents of the Jacobite creed”’.

There is also another evidence about which the editor is completely reticent, whatever is the position one woud take on it. As also mentioned by Wion 2002, 287, the catalogue of MS C (EMML 1195) gives a subscriptio to the Mäṣḥafä Sawiros with a date to AD 1385/1396 (see Getatchew Haile and Macomber 1979, 152), here fully edited and translated from the manuscript, now accessible online on the website of the HMML (https://www.vhmml.org/):

(p. 249a) ወውእቱ፡ ተፈጸመ<፡ >በዕለተ፡ ዐርብ፡ አመ፭ለጥር፡ ወርኅ፡ ወበዓመተ፡ ፲፻ ወ፬፻ ወ፲ወ፪ ዓመት፡ ለሰማዕታት፡ ንጹሐን፡ በረከቶሙ፡ ተሀሉ፡ ምስለ፡ አቡነ፡ ኀብተ፡ ማርያም፡ ወምስለ፡ ንጉሥነ፡ ኃይለ<፡ >ሥላሴ፡ ወምስለ<፡ >ጸሐፊ<፡ >ገብረ፡ ኢየሱስ፡ ለዓለመ፡ ዓለም፡ አሜን። ወበዓመተ፡ ፲፻ ወ፫፻ ወ፹ወ፰ እምትስብእቱ<፡ >ለእግዚእነ፡ ኢየሱስ<፡ >ክርስቶስ፡ ሎቱ፡ ስብሐት፡ ወሣህሉ፡ ላዕሌነ፡ አሜን። ወበዓመት<፡ >፷፻ ወ፰፻ ወ፹ወ፰ ለ(p. 249b)ፍጥረተ፡ አቡነ፡ አዳም፡ በረከቱ፡ ተሀሉ፡ ምስለ፡ ሕዝበ፡ ክርስቲያን፡ ለዓለመ<፡ >ዓለም፡ አሜን።።

And that (work) was completed in the day of Friday, the 5 of Ṭǝrr and in the year 1412 (mistake for 1112) year of the pure Martyrs. Let their blessing be with Abunä Ḫabtä Maryam and with our king Ḫaylä Śǝllase and with the scribe Gäbrä Iyäsus, for ever and ever, amen. And in the year 1388 from the incarnation of our Lord Jesus Christ, to him glory, and (let) his mercy (be) upon us, amen. And in the year 6888 of the creation of our father Adam, let his blessing be with the Christian people, for ever and ever, amen.

The 5 of Ṭǝrr of 1388 (year of Mercy, that is 1 January AD 1396) was a Saturday and not a Friday, and therefore something is wrong in this colophon; it is likely that the day refers to the copying of the manuscript. Rightly Wion (Wion 2002, 286) points out that since other subscriptiones date Gǝʿǝz works to AD 1395/1396, there can be a mythical element that has caused this fictitious attribution. Yet, there are two (and not only one, as noted by Wion 2002, 287) attributions to a period before the seventeenth century of the Mäṣḥafä Sawiros in notes recorded by Ernst Hammeschmidt in his catalogue of the Lake Ṭana manuscripts: the first one, on the inner side of the cover of MS T (Ṭānāsee 65 = Dabra Māryām 7, see Hammerschmidt 1977, 63), states: ደረ። ደብረ ማርያም። ሣዊሮስ፡ በአፄ፡ በዕደ፡ ማርያም፡ ዘመን። (‘Dära. Däbrä Maryam. Śawiros, at the time of Aṣ́e Baʿǝdä Maryam’, that is, AD 1468–1478); the second one, erroneously place on the inner side of the cover of MS Ṭānāsee 59 = Dabra Māryām 1, but actually referring to Ṭānāsee 65 = Dabra Māryām 7 (see Hammerschmidt 1977, 52), states: ደረ፡ ደሴት፡ ውስጥ፡ ደብረ ማርያም። ወንጌል፡ ማርቆስና፡ ሰዊሮስ፡ በአጼ፡ አምደ፡ ጽዮን፡ ዘመን። (‘In the Dära isle, Däbrä Maryam. Gospel of Mark and Säwiros, at the time of Aṣ́e ʿAmdä Ṣǝyon’, that is, AD 1314–1344). As Hammerschmidt points out, the content of the note matches MS T. Obviously, no palaeographic attribution of that manuscript to the fourteenth century is possible.

The Arabic tradition

An essential and valuable aspect of this edition of a Gǝʿǝz text is the systematic consideration of the Arabic tradition, from which the Gǝʿǝz text depends, even though, as is common, what is available in print is not a reliable critical edition. The editor does not provide any estimation of the number of available Arabic manuscripts (he provides references to literature, though), but has taken the pain of considering, besides one printed edition (E = Abū al-Bišr Sāwīrus ibn al-Muqaffaʿ 1925), also four manuscripts (MPQY, the earliest one from the thirteenth century), which are listed in the sigla section and are used in the notes to the translation. Besides the manuscripts of the complete work (pp. [15–16]/147–148), there exists also an abridgment in Arabic that was translated into Gǝʿǝz, in one single manuscript (Paris, Bibliothèque nationale de France, Éthiopen, d’Abbadie 122; see Conti Rossini 1914, no. 35, published by Cerulli 1960, 189–211 (text), 215–236 (transl.), on which the editor, on p. 148, n. 34, appears to bring some important new hypotheses).

That the editor must use the Arabic cautiously is demonstrated by the passage (pp. [13–14]/145–146) added in the Gǝʿǝz translation falsely attributed to Dioscorus (who largely antedated Jacob Baradaeus), which supports the claim of the Ethiopians to no subordinate role among the other Monophysite (or Miaphysite) Christian communities (as declared by the pseudo-nicaean canon that limits the prerogatives of the metropolitan of Ethiopia, see in the Fǝtḥa nägäśt, Guidi 1899, 30). This addition states that ‘Quanto agli Etiopi sono Giacobiti insieme a noi fin dall’inizio, prima di costoro (dei Siriani e degli Armeni)’, ‘As to the Ethiopians, they are Jacobite along with us, before them (i.e. the Syrian and the Armenian)’. It is difficult, however, to agree that the anti-historical attribution to Dioscorus marks something that is not acceptable, if we think of the theological meaning of being Jacobite, which is the essential element here. Be it as it is, it appears likely that since the earliest stage, the Gǝʿǝz text was not only translated, but also adapted to some extent.

Language and translation

The Italian translation is written in a clear and fluid language. The editor (and translator) has found brilliant solutions for the most difficult tournures, and has rightly opted for rendering the sense rather than the letter; see an example (p. [40]172.28–29): ‘ቃልከ፡ ሎሙ፡ ዘይጤይቅ፡’, translated as (p. [41]/173.29–30): ‘la tua parola sia per essi di convincimento’, where the frequently used verb አጠየቀ፡ receives the appropriate meaning, while the same verb is rendered differently in other contexts, for example (p. [39]1/71.2–3) ‘per assicurare’, or (p. 171.22–23) ‘Ci hanno dimostrato’. There might be cases when one disagrees (for example, p. [36]/168.3): ‘አሐዱ፡ ገጽ፡’, translated as (p. [37]/169.2) ‘un solo volto’, whereas one would expect ‘una sola persona’ (‘only one person’), but the translation is definitely one of the strong points of this edition.

The book is written in Italian, by an Eritrean-born author educated in the Italian school. Yet, still admiring the genius of a polyglot who can master several European, Ethiopian Semitic, and Christian oriental languages, there are not a few passages that betray an exotic flavour: see for example (p. [8]/140) ‘Accenni prosopografici sull’autore’, for ‘Cenni prosopografici’ etc.; (p. [10]/142) ‘Pur non essendoci citazioni esplicite dei Padri della Chiesa […] è chiaro che l’autore ne possedeva una loro [sic] ampia conoscenza’, where ‘loro’ is superfluous and not acceptable in the written language; (p. [10]/142) ‘Sawirus si dimostra a suo agio con la letteratura agiografica, ben percettibile nel retropensiero [sic] del KĪ, che per quanto concerne la pratica della vita cristiana riflette una forte impronta monastica’, where ‘retropensiero’ seems to stay for ‘retroterra’, ‘background’, while ‘retropensiero’ (arrière-pensée) has a different meaning. Other examples are marked with ‘sic’ in this review. In the case of the word (p. [36]/168.25) ‘ጳጦስ፡’, rendered in the translation (p. [37]/169.24) with ‘ṗaṭos’, and explained in the footnotes as a ‘calco’, one wonders whether the author does not understand what ‘calco’ means in Italian, or rather lacks understanding of linguistic interference: this is in fact not a ‘calque’, but exactly the opposite of a calque, that is, a ‘loanword’, or ‘borrowing’ in the narrow sense (p. [37]/169, n. 18: ‘Ṗaṭos è un calco del greco βάτος, “roveto” (Es. 3,2)’).

Without being able to carry out a systematic check, I quote the following passage as an example where the translation does not reflect the Gǝʿǝz text, but is the result of an interpretation which is not made explicit in the constitutio textus nor in the apparatus (p. [38]/170.18): ‘በውስተ፡ ሃይማኖተ፡ ጉባኤ፡’, translated as (p. [39]/171.20): ‘nel sinodo della fede’; as noted in the footnote (ibid., n. 21), the Arabic has في الامانة لاجامعة, which is explained as ‘nel credo comune’. In fact, the literal translation of the Gǝʿǝz is ‘nella fede del sinodo’ (‘in the faith of the synod’), which presupposes a slightly different Arabic text (for example, في امانة لاجامعة), whereas the translation ‘nel sinodo della fede’ is not grammatically justified.

The text is full of interesting passages that provide abundant material for the study of medieval (post-Aksumite) Gǝʿǝz: see for example (p. [36]/168.5) the construction ‘እስከ፡ ኮኑ፡ የዐጽቦሙ፡’, (p. [37]/169.4) ‘a tal punto che fu ad essi difficile’, with the auxiliary verb konä in the plural, in agreement with the suffix pronoun of the singular verb in the konä-periphrastic construction; the same konä-periphrastic construction (p. [34]/166.24) ‘ወኮኑ፡ ኢይሰምዑ፡’, (p. [35]/167.26) ‘e non sentirono’, would be better translated as ‘e non stavano a sentire’.

Transcriptions

The accuracy of transcription is an essential aspect of the quality of a work where several languages are used (besides Italian and other European languages, Gǝʿǝz, Arabic, and Amharic as well). There are a few errors, see for example from the first pages of the book (p. [8]/140, twice): ‘Abū al-Barākat Ibn Kabar’, instead of ‘Abū al-Barakāt Ibn Kabar’, and ‘Anbā Mīẖa’īl’, instead of ‘Anbā Mīẖā’īl’, but, as a rule, these errors do not seriously affect the work. The same for other mistakes such as, for example (p. [11]/143, n. 13): ‘arabisher’ instead of ‘arabischer’; (p. [11]/143.6): ‘pontenziali’ instead of ‘potenziali’; and so also in the text (p. [34]/164.6): ‘ዘገረ፡’ instead of ‘ዘሀገረ፡’.

Overall synthetic evaluation

A fair overall evaluation of the critical edition with annotated translation by Tedros Abraha of the first three homilies of the Mäṣḥafä Sawiros of Sāwīrus ibn al-Muqaffaʿ is difficult. The edition is the result of a considerable amount of work and has the merit of providing the first edition of an unpublished text established through the analysis of large part of the manuscript tradition, and is furnished with a rich apparatus of variants. The careful translation makes the text accessible to the non-specialists, and, more than that, is an essential complement to the edition, since it provides the ultimate interpretatio of the edited text. The systematic use of the Arabic text, documented in the footnotes to the translation, offers large material to clarify discrepancies, misunderstandings, and the making of the Gǝʿǝz version. This is all in all an important and valuable contribution to the study of the Gǝʿǝz literary heritage.

The method of the edition, however, shows contradictory aspects. Let’s note first that the editor shows a surprising reticence—with a manuscript tradition dating from the seventeenth century, with the only exception of one manuscript, unfortunately not accessible, dating from the fifteenth century (MS Qoma Fasilädäs)—in accepting that the terminus non post quem of the translation in Gǝʿǝz must be the fifteenth century. On the positive side, the editor has carried out an evaluation of the single witnesses, applied at least to some extent a correct eliminatio codicum descriptorum, even though the reasons for some eliminations are not specified, and has not assumed any ‘better’ or ‘older’ manuscript as base for the edition in the absence of convincing reasons for this choice. On the negative side—aside from the ample room for technical improvement of the apparatus—a clear recensio, at whatever level of the tradition, is missing: there is neither any attempt at making a sense of the sparse observations concerning common errors and conservations—the latter, observed for manuscripts AGTV, but also MS Jerusalem JE 342E, which would demonstrate the existence of an innovative family common to the non-conservative witnesses—nor any sense is made of the correspondence in the sequence of texts observed for other manuscripts (AS, but also MS d’Abbadie 125, and MS Qoma Fasilädäs), which cannot be accidental and could have suggested a first genealogical hypothesis to be proved on more substantial evidence.

Finally, the lack of distinction in the objective to be reached between the Gǝʿǝz text translated from the Arabic, and the Arabic text as it is attested, which is used as a virtual archetype, risks to justify textual choices that calque the edited Gǝʿǝz text upon the Arabic model, also when the agreement with the Arabic is the result of a secondary development.

Bibliographical references

Abū al-Bišr Sāwīrus ibn al-Muqaffaʿ 1925. Kitāb al-durr al-ṯamīn fī īḍāḥ al-dīn, ed. Murqus Ǧirǧis (Cairo: n.pub., 1925).

Aland, Barbara, Kurt Aland, Johannes Karavidopoulos, Carlo M. Martini, and Bruce M. Metzger, eds, 2012. Novum Testamentum Graece Based on the work of Eberhard and Erwin Nestle, (28th edn, Edited by the Institute for New Testament Textual Research, Münster / Westphalia under the direction of Holger Strutwolf) (Stuttgart: Deutsche Bibelgesellschaft, 2012).

Ambu, Martina 2018–2019. ‘Review of Tedros Abraha, Mäṣḥafä Sawiros zä-Esmunayn: la versione geʿez del Kitāb al-īḍāḥ di Severo di Ašmūnayn (X/XI sec.): Omelie I–III. Introduzione, edizione del testo etiopico con traduzione italiana annotata, Patrologia Orientalis, 56/2 (247) (Turnhout: Brepols, 2017))’, Annales d’Éthiopie, 32 (2018–2019), 319–322.

Bausi, Alessandro 2016. ‘On editing and normalizing Ethiopic texts’, in Alessandro Bausi and Eugenia Sokolinski, eds, 150 Years after Dillmann’s Lexicon: Perspectives and Challenges of Gǝʿǝz Studies, Supplement to Aethiopica, 5 (Wiesbaden: Harrassowitz Verlag, 2016), 43–102.

Bausi, Alessandro 2020a. ‘7.5 Ethiopic’, in Philippe Roelli, ed., Handbook of Stemmatology: History, Methodology, Digital Approaches, De Gruyter Reference (Berlin–New York, NY: De Gruyter, 2020), 479–493.

Bausi, Alessandro 2020b. ‘Tradizione e prassi editoriale dei testi etiopici: un breve sguardo d’insieme’, AION (filol.) Annali dell’Università degli Studi di Napoli “L’Orientale”, 42/1 (= Atti del Seminario internazionale ‘Filologie del Mediterraneo’, Archivio di Stato di Napoli, 16-17 ottobre 2018) (2020), 184–207.

Cerulli, Enrico 1960. Scritti teologici etiopici dei secoli XVI-XVII, II: La Storia dei Quattro Concili ed altri opuscoli monofisiti, Studi e Testi, 204 (Città del Vaticano: Biblioteca Apostolica Vaticana, 1960).

Chaîne, Marius 1912. Catalogue des manuscrits éthiopiens de la Collection Antoine d’Abbadie (Paris: Imprimerie Nationale, 1912).

Conti Rossini, Carlo 1914. Notice sur les manuscrits éthiopiens de la Collection d’Abbadie, Extrait du Journal Asiatique (Paris: Imprimerie nationale, 1914).

Dillmann, Christian Friedrich August 1865. Lexicon linguae aethiopicae, cum indice latino. Adiectum est vocabularium tigre dialecti septentrionalis compilatum a W. Munziger (Lipsiae: T. O. Weigel, 1865).

Getatchew Haile and William F. Macomber 1979. A catalogue of Ethiopian manuscripts microfilmed for the Ethiopian Manuscript Microfilm Library, Addis Ababa and for the Hill Monastic Manuscript Microfilm Library, Collegeville, Vol. IV: Project Numbers 1101-1500 (Collegeville, MN: Hill Monastic Manuscript Library, 1979).

Getatchew Haile and William F. Macomber 1982. A catalogue of Ethiopian manuscripts microfilmed for the Ethiopian Manuscript Microfilm Library, Addis Ababa, and for the Hill Monastic Manuscript Library, Collegeville, Vol. VI: Project Numbers 2001-2500 (Collegeville, MN: Hill Monastic Manuscript Library, 1982).

Guidi, Ignazio, ed., 1899. Il “Fetha Nagast” o “Legislazione dei Re”: codice ecclesiastico e civile di Abissinia, tr. Ignazio Guidi, II, Pubblicazioni scientifiche del R. Istituto Orientale in Napoli, 3 (Roma: Casa Editrice Italiana, 1899).

Hammerschmidt, Ernst 1977. Äthiopische Handschriften vom Ṭānāsee, II: Die Handschriften von Dabra Māryām und von Rēmā, Verzeichnis der orientalischen Handschriften in Deutschland, 20/2 (Wiesbaden: Steiner, 1977).

Isaac, Ephraim 1984–1986. ‘Shelf List of Ethiopian Manuscripts in the Monasteries of the Ethiopian Patriarchate of Jerusalem’, Rassegna di Studi Etiopici, 30 (1984–1986), 53–80.

Rahlfs, Alfred and Robert Hanhart, eds, 2006. Septuaginta (Stuttgart: Deutsche Bibelgesellschaft, 2006).

Raineri, Osvaldo 2004. Enrico Cerulli. Inventario dei manoscritti Cerulli etiopici, Studi e Testi, 420, Cataloghi sommari e inventari dei fondi manoscritti, 8 (Città del Vaticano: Biblioteca Apostolica Vaticana, 2004).

Tedros Abraha 2006. ‘The Ethiopic Version of the Mäṣḥafä Sawiros’, in Siegbert Uhlig, Maria Bulakh, Denis Nosnitsin, and Thomas Rave, eds, Proceedings of the XV^th International Conference of Ethiopian Studies, Hamburg, July 20–25, 2003, Aethiopistische Forschungen, 65 (Wiesbaden: Harrassowitz Verlag, 2006), 628–637.

Tedros Abraha 2010. ‘Severos of Ašmunayn’, in Siegbert Uhlig and Alessandro Bausi, eds, Encyclopaedia Aethiopica, IV (Wiesbaden: Harrassowitz Verlag, 2010), 640a–641b.

Tedros Abraha 2017a. Mäṣḥafä Sawiros zä-Esmunayn: la versione geʿez del Kitāb al-īḍāḥ di Severo di Ašmūnayn (X/XI sec.): Omelie I–III. Introduzione, edizione del testo etiopico con traduzione italiana annotata, ed., tr. Tedros Abraha, Patrologia Orientalis, 56/2 (247) (Turnhout: Brepols, 2017).

Tedros Abraha 2017b. ‘Sāwīrus ibn al-Muqaffaʿ in Gǝʿǝz Literature: A Provisional Reappraisal’, Le Muséon, 130/3–4 (2017), 421–444.

Tedros Abraha 2020. Mäṣḥafä Sawiros zä-Esmunayn: la versione geʿez del Kitāb al-īḍāḥ di Severo di Ašmūnayn (X/XI sec.): Omelie IV–VIII. Introduzione, edizione del testo etiopico con traduzione italiana annotata, ed., tr. Tedros Abraha, Patrologia Orientalis, 57/2 (250) (Turnhout: Brepols, 2020).

Wion, Anaïs 2002. ‘The Manuscripts Library of Qoma Fasilädäs Monastery”’, in Baye Yimam, Richard Pankhurst, David Chapple, Yonas Admassu, Alula Pankhurst, and Birhanu Teferra, eds, Ethiopian Studies at the End of the Second Millennium: Proceedings of the XIV^th International Conference of Ethiopian Studies, November 6–11, 2000, Addis Ababa, I (Addis Ababa: Institute of Ethiopian Studies, Addis Ababa University, 2002), 275–300.

Wion, Anaïs 2012. Paradis pour une reine. Le monastère de Qoma Fasilädäs, Éthiopie, XVII^e siècle, Histoire ancienne et médiévale, 112 (Paris: Publications de la Sorbonne, 2012).

Wright, William 1877. Catalogue of the Ethiopic Manuscripts in the British Museum Acquired since the Year 1847 (London: Gilbert and Rivington, 1877).