Eustathius of Thessalonica, Commentary on the Odyssey cover image


Eustathius (c. 1115–1195) was archbishop of Thessalonica, and before that a successful teacher and rhetorician in Constantinople during the reign of Manuel I Komnenos. He was remembered as the most learned man of his day, and his philological work represents one of the greatest surviving monuments of Byzantine classical scholarship. In connection with his activities as teacher and orator, Eustathius produced what he termed parekbolai on the Odyssey and the Iliad,1 Dionysius Periegetes,2 Pindar (lost except for the preface),3 Aristophanes (lost except for a few fragments),4 and perhaps Oppian (lost).5 These works are not “commentaries” in the sense of explanatory or critical notes on a text, but collections of miscellaneous material in some way associated with phrases and words in the ancient poems. They were composed via partially overlapping processes of gradual accumulation of entries that must have been produced over the course of a number of years before being finally compiled into the texts we have.6

The Commentaries are heavily reliant on the ancient scholia on the Homeric poems. They also refer constantly, both explicitly and tacitly, to Athenaeus’ Deipnosophistai, which was known to Eustathius through a long-lost copy of the Epitome of the text.7 Perhaps more important from a modern perspective, Eustathius had access to a wealth of ancient secondary material, including many grammarians and lexicographers, that was held in libraries in Constantinople and Thessalonica and later destroyed. He is thus our last—and in many cases our only—means of access to this literature and the long tradition of learning it represents.

The editio princeps of the Commentary on the Odyssey was printed in Rome by Nicolaus Majoranus in 1542–1550, as the third volume of a set that also included the Commentary on the Iliad. Page- and line-numbers from the Rome edition represent the standard modern means of citing the text. A Basel edition of the text of 1560–1569 represents little more than a clumsy reprinting of the editio princeps. Modern readers generally know the Commentary through the two-volume edition of Johann Gottfried Stallbaum (Leipzig 1825–1826). Stallbaum’s work was not based on original manuscript work. It accordingly offers no critical apparatus, and contains its own unique crop of errors, omissions and unnoted small corrections of the text. Our new edition thus represents the result of the first systematic collation of the manuscripts of the Commentary on the Odyssey in almost five hundred years, and the first true critical edition of the text.8 It is also accompanied by the first translation of the text into any modern language,9 as well as by a full set of notes and apparatuses that trace Eustathius’ sources and identify the ancient texts to which he refers.

Our project has its origins in Cullhed’s 2014 Uppsala University doctoral dissertation, which treated only Books 1–2 of the Commentary.10 Each of us brings his own background and interests to our new edition. But we have conceived this from the first as a fully cooperative effort. Whatever praise or criticism our text, notes and apparatus are thought to deserve, should accordingly be directed to us equally and impartially.

The Text

All known copies of the Commentary on the Odyssey descend from two surviving twelfth-century manuscripts that were written and annotated in the same hand and finished in Thessalonica: P = Par. gr. 2702 (Paris, Bibliothèque Nationale) and M = Marc. gr. 460, coll. 330 (Venice, Biblioteca Nazionale Marciana).11 Both descend from a lost master copy (α), which was probably similar to L = Laur. plut. LIX, 2 and 3 (Biblioteca Laurenziana, Florence), containing Eustathius’ Commentary on the Iliad. L is written and annotated by the same hand as M and P, and includes over 3000 substantial additions made in the margins or on slips of oriental paper sewn into the binding. It was clearly a working copy produced by Eustathius himself or perhaps more likely by an assistant working under his supervision. M and P also contain later interlinear and marginal additions and corrections in the same hand as that of the main text, although these are not as numerous as in L. In other words, M and P represent a stage of textual development that is analogous not to L but to a stage after L.

The relationship between M, P and α is complicated by the fact that, when the author or his assistant produced M, he recycled quires from previously disassembled copies that predated P and filled in the gaps with new text.

In 1395.60–1482.23 (folia 1–38 in M), M was copied from α after P, and was thoroughly corrected with further additions. In 1482.23–1537.39 (folia 39–62 in M), M consists of quires copied from α before P. There are accordingly marginal and interlinear additions in P not found at the corresponding point in M. In 1537.39–1852.41 (folia 63–201 in M), M was copied from α after P. It therefore incorporates marginal additions in P into the main text and offers further additions and corrections not found in P. The corrections and revisions in M, however, break off after fol. 95v (1611.18). In 1852.57–1917.20 (fol. 202–251 in M), M again consists of quires from a copy of α that predated P. There is accordingly a marginal addition in P which was not incorporated into M.

The principle of this edition is to print the latest and most fully revised version of the text according to the specific conditions in each section. As in van der Valk’s edition of the Commentary on the Iliad, we mark material that was added after revisions by placing it within double square brackets ⟦…⟧. Marginal and interlinear corrections of errors are not marked this way, however, but are integrated into the text and noted in the apparatus criticus. Interlinear notes that are not corrections or additions, but part of the text itself as variant readings, are printed with a slash between the readings (e.g. 1385.51 συναλιφῇ/συναλοιφῇ; 1401.56 ἡ τροφὴ τοῦ/τῷ ἀνθρώπου/ῳ ἀγαθὸν ἀναγκαῖον). Emendation is limited to mechanical errors, and no attempt has been made to correct illogical or misguided statements. Thus the discussion of the word ἄγαυρος (“splendid”) at p. 1444.7–8 reads: εἰ δέ τι μετέχει τοῦ τοιούτου ἀγαυροῦ τὸ ἀγαύρισμα, ὅπερ ἐστὶ πάλης εἶδος, οὐκ ἂν εἴη ἀκριβῶς ἀποφήνασθαι (“It would be impossible to prove precisely whether agaúrisma, which is a form of wrestling, has anything to do with this agaurós”). ἀγαύρισμα is a palaeographically explicable corruption of ἀγκυρισμα (cf. Hsch. α 582; Suda α 261), but this is a mistake by Eustathius or his source, not a mechanical error.

Accents on enclitics have been normalized, and the punctuation offered is syntactic, as opposed to the elocutionary punctuation of the Byzantine manuscript. The text is accompanied by references to relevant verses in the Odyssey, three note apparatuses keyed to pages and lines in the Rome edition, and a facing English translation.

The function of the apparatus criticus is primarily to indicate that a textual error has been corrected in the edited version, or that a textual error or variant is present in one of the two manuscripts.

Whenever Eustathius explicitly mentions or cites a source, a reference is provided in the citation apparatus. Sources used tacitly, along with further comparanda, are noted in the source apparatus. A full list of parallels is generally not given when this can be found in recent editions of Eustathius’ source, such as Erbse’s edition of the scholia on the Iliad or Pontani’s edition of the scholia on the Odyssey up to Book 8.

The edition follows the page and line breaks in the Rome edition. Pages in Stallbaum’s edition and folia in P and M are also indicated. The text is accompanied by an English translation. Eustathius’ parekbolai are impromptu notes and extracts. His Greek is often inelegant and repetitive, his arguments and analyses wandering and obscure, his language technical, and his discussions so deeply embedded in previous scholarly controversies that their point can be difficult to comprehend at first. The translation makes no systematic effort to conceal these characteristics. Instead, it offers an English version of the text that is intended to be simultaneously as true to the Greek original and as clear as possible. The reader may accordingly be forced at times to read through a section more than once in order to understand exactly what is being argued and why. Owing to the lexicographic function of Eustathius’ commentaries, Greek words are often transliterated rather than translated. Explanations of these words are added within square brackets, unless Eustathius himself explains them in the same section. Greek words that form part of the syntax, but whose phonology or etymology is significant for the argument, are translated, with a transliteration added within square brackets. Round parentheses indicate explanatory supplements of the translation. The most familiar ancient names are given their standard Latinized forms in the translation (thus “Aeschylus” rather than “Aiskhylos”), while others are transliterated directly from the Greek. This is a compromise that will satisfy no one, but is the best that we can offer.

1. See v.d.Valk I–IV.

2. K. Müller, Geographi Graeci minores, vol. 2, Paris 1861.

3. A. Kambylis, Eustathios von Thessalonike: Prooimion zum Pindarkommentar, Göttingen 1991.

4. W.J.W. Koster & D. Holwerda “De Eustathio, Tzetza, Moschopulo, Planude Aristophanis commentatoribus I”, Mnemosyne 7, 1954, 136–156; D. Holwerda, “De Tzetza in Eustathii reprehensiones incurrenti”, Mnemosyne 13, 1960, 323–326.

5. A. Dyck, “Did Eustathius compose a commentary on Oppian’s Halieutica?”, Classical Philology 77, 1982, 153–54; cf. v.d.Valk I xci and II l. A manuscript allegedly containing scholia on Greek epigrams by Eustathius was destroyed in the fire of 1671 at the Escorial; see J. Iriarte, Regiae bibliothecae Matritensis codices Graeci mss, Madrid 1769, I 277.

6. For recent work on Eustathius, see F. Pontani, V. Katsaros and V. Sarris (eds.), Reading Eustathios of Thessalonike, Berlin 2017; F. Pontani, “Scholarship in the Byzantine Empire (529–1453)”, in F. Montanari (ed.), History of Ancient Greek Scholarship: From the Beginnings to the End of the Byzantine Age, Leiden 2020, 460–467.

7. See M. van der Valk, “Eustathius and the Epitome of Athenaeus”, Mnemosyne 39, 1986, 400.

8. For an edition of the proem that laid important groundwork for this project, see F. Pontani, “Il proemio al ‘Commento all’Odissea’ di Eustazio di Tessalonica (con appunti sulla tradizione del testo)”, Bollettino dei classici 21, 2000, 5–58.

9. The Spanish humanist Vicente Mariner translated the text into Latin between 1622 and 1623 (Madrid, Biblioteca Nacional, Matr. lat. 9861–9862).

10. E. Cullhed, Eustathios of Thessalonike: Parekbolai on Homer’s Odyssey 1–2: Proekdosis, Uppsala 2014, later published as Eustathios of Thessalonike: Commentary on Homer’s Odyssey. Volume 1: On rhapsodies A–B, Acta Universitatis Upsaliensis: Studia Byzantina, Uppsala 2016. For reviews, see T. Dorandi, Byzantine and Modern Greek Studies 43, 2019, 316–318 and S. Lampakes, Byzantina Symmeikta 29, 2019, 475–476.

11. For a more complete account of what follows, see F. Pontani, “Il proemio al ‘Commento all’Odissea’ di Eustazio di Tessalonica” (n. 8); E. Cullhed, “The Autograph Manuscripts Containing Eustathius’ Commentary on the Odyssey”, Mnemosyne 65, 2012, 445–461; “Editing Byzantine Scholarly Texts in Authorized Manuscripts”, in The Arts of Editing Medieval Greek and Latin: A Casebook, Toronto 2016, 72–95; Eustathios of Thessalonike: Commentary on Homer’s Odyssey (n. 10) 2*–9*, 35*–58*.


I am grateful for financial support from Riksbankens Jubileumsfond through a Pro Futura Scientia XIII fellowship, and for the excellent working conditions provided by the Swedish Collegium for Advanced Study in Uppsala and the Swiss Center for Affective Sciences in Geneva.


Geneva, 12 February 2022

Much of the work on this project was completed at Bilkent University in Ankara, where I have been a Visiting Professor since 2017. Other work was carried out at the Fondation Hardt in Geneva and at the University of Bari. I gratefully acknowledge support for my research in 2021–2022 carried out under an agreement for the provision of grants from the federal budget of the Russian Federation in the form of subsidies No. 075-15-2021-571, project “Digital commentaries to classical texts: Greek comedy” (IWL RAS, Moscow, Russia). The final set of proofs was read in the library of the American School of Classical Studies at Athens.


Athens, 7 February 2022