John Briscoe, Georg Korting: Varus’ Untergang. Textkritische Anmerkungen zu Florus 2,30,34b. in:

Gnomon, page 507 - 511

GNO, Volume 91 (2019), Issue 6, ISSN: 0017-1417, ISSN online: 0017-1417,

Browse Volumes and Issues: Gnomon

C.H.BECK, München
Bibliographic information
P. Fàbregas Salis: Luck (†), A textual commentary on Ovid 507 argued that 426–30 break the sequence of the transformation of Troy into Rome (424–5, 431). This is a very strong case against the authenticity of the lines, as L. admits. On the other hand, I do not find as compelling his discussion of 502, which Heinsius deleted. L.’s proposal to emend it (finxit me velle for finxit voluisse) is clever, but not a great improvement. But I totally agree on his defence of the less well attested version of 504 arguit immeritumque pater proiecit ab urbe (Heinsius’ text). Finally, let me focus on some entries of L.’s commentary that might look like a mere collection of parallels intended to justify the transmitted text. One might wonder why L. felt the need to do so. In some cases his instincts point out difficulties in the text that may need either to be explained away or corrected. I will produce one example (we could add 444, discussed above): 92–3 Tarrant’s text runs nil te nisi tristia saevo / vulnera dente iuvat. According to OLD it is the sole instance of vulnus (s.v. 1) in the specific sense of ‘wounded flesh’. L. compares tristia … vulnera with 7.849 vulnera saeva and 13.531 crudelia vulnera, which do not remove the difficulty. I thought of viscera or corpora, but they are ruled out by 88–9. G. Liberman (per litt.) brilliantly suggests funera (cf. OLD s.v. funus 2a) and supports it with am. 2.6.41. But the transmitted text might be defended with Petron. 121 vers. 120 concisaque vulnera mande (although Colladonius suggested funera or viscera; viscera is also a proposal by Liberman). Lastly, it is regrettable that the book came out with many misprints (many of them pointed out by Tarrant: see above) and without indexes. One also misses a comprehensive bibliography (that on 157–63 is too ‘select’, in my opinion). To sum up, we can only be grateful that L.’s commentary, though unfinished, did not remain unedited. For the acumen of L.’s textual discussions (see lines 426–30) guarantee that this book will be a crucial and indispensable tool for future editors of book 15, and it will certainly be a stimulus for further thought on the text. However, as I have tried to show, there are still many codices awaiting collation and not every conjecture ever made has been taken into account. Only when editors have at their disposal a complete catalogue of variant readings and conjectures will they be able to produce editions that might claim to be somewhat less provisional than our present ones. Meanwhile, as L. stated in his unforgettable article ‘Textual Criticism Today’,1 «the struggle goes on». Barcelona Pere Fàbregas Salis * Georg Korting: Varus’ Untergang. Textkritische Anmerkungen zu Florus 2,30,34b. Heidelberg: Propylaeum 2017 (online). 174 S. I should say at the outset that this work was sent to me online and that to the best of my knowledge a print version does not currently exist. It does, however, have a cover image, with a reproduction of Florus 2.30.33–34 from Bamberg, Staatsbibliothek Class. 31 (B), the second oldest of our manuscripts (s. x1; the oldest is Heidelberg, Pal. Lat. 894 [N; s. ix], one of the two principal manuscripts of the periochae of Livy) and I understand that a print version will follow: it is to be hoped that the many presentational errors in the electronic version (e.g. page numbers missing in ch. 1, hyphenated words in the middle of a line, italic for non-italic and vice-versa, the ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 1 AJPh 102, 1982, 164–94: 194. GNOMON 6/91/2019 J. Briscoe: Korting, Varus’ Untergang 508 beginning of a sentence without the rest, missing page numbers, Freinsheim’s name sometimes appearing thus, sometimes Latinised as Freinshemius) will be eliminated. Meanwhile, the work is available on open access at I am unaware of any previous publications by Korting (K.) (none are listed in the bibliography). Nevertheless, he possesses considerable learning and has accumulated a vast bibliography (see below); his knowledge of the editing and criticism of the text of Florus is particularly impressive (note the biographical material in ch. 4 on T. Faber and his daughter, Anna Dacier). Stylistic peculiarities which I have not previously encountered are ‘n. JN’ (nach Jesum natum?) instead of ‘n. Chr.’ to indicate dates AD (CE) and ‘Zeile’ to mean ‘chapter’. P. Quinctilius Varus, consul in 13 BC and related to the imperial family, was in command in Germany in AD 9 when he was attacked by Arminius, with the loss of three legions and his own life. (K. writes from Paderborn, near the Teutobergensis saltus, where the events took place.) The transmitted text of Florus makes the slaughter occur inside the Roman camp, while Dio (56.18–22.2) places it in open country. Scholars have unanimously followed Dio, many taking a poor view of Florus as a whole, though Freinsheim thought that Florus was abbreviating the same source as that used by Dio. K. seeks to reconcile Florus with Dio by emending the transmitted text, altering o securitas to securi ita se and ex inprouiso to exin pro auio. One does not expect to find a whole (virtual) book to be devoted to justification of two related conjectures, though in fact the argument ends on p. 92; the rest of the work consists of appendixes; the longest (pp. 93–132) is a list of all instances of citare, with both Latin text and translation (the latter not K.’s own, hence in various languages), the remainder almost entirely bibliography. K. has devoted a vast amount of work to the problem and the details of his argument deserve serious consideration: I therefore discuss it seriatim. The problem is explained in ch. 1 (pp. 5–8), while ch. 2 (pp. 9–13) briefly sets out what is known about the author (K. alludes to Neuhausen’s view that the original text, covering the period from Romulus to the death of Augustus, was composed in AD 14 or 15, the rest being added in the age of Trajan or Hadrian; this explains the cryptic «um 15 n. JN oder 2. Jh.» on p. 5), briefly mentions manuscripts and editions, and cites views that have been held about Florus’ style and reliability. There follows (ch. 3; pp. 14–15) a tabulation of 2.30.32–4, with sub-divisions of each chapter (32 a–c, 33 a–b, 34 a–e; this explains «34b» in the title of the work), each sub-division having a coding (I. 1. A, I. 1. B etc.), indicating the structure of the passage. In ch. 4 (pp. 16–23) K. sets out three objections to the usual interpretation of the transmitted reading. Firstly, it takes cum ille … citaret as temporal (he does not, of course, base his argument on the subjunctive, which is regular with temporal cum in imperial Latin), but the structure of the passage requires it to be causal, parallel to at illi … ut primum (30.32) and cum interim … ut ne (30.33): not so; 30.34 is the climax of the narrative and there is no reason why Florus should not have said that the Germans attacked at the very moment that Varus was summoning defendants (see below) to his tribunal. Secondly, Florus is very fond of exclamations but when he uses o with a noun alone (1.22.30 o pudor, GNOMON 6/91/2019 J. Briscoe: Korting, Varus’ Untergang 509 2.12.2 o nefas), he is expressing his own judgement about an action; in other cases, the person, action or behaviour is accompanied by an adjective, pronoun or other addition. All true, but it does not follow that Florus did not write o securitas: we have three instances of o + noun alone; the nature of two of them cannot be used to cast doubt on the third. The last argument, for K. clearly the most important, occupying the rest of ch. 4 and the whole of chs. 5 and 6 (pp. 19–42; see also the appendix mentioned above), is that citaret lacks an object. He says, correctly, that citare is normally accompanied by an accusative (or a nominative if passive) and is not a verb which can be either transitive or intransitive. He then discusses all passages, from Plautus to the mulomedicina Chironis, where citare lacks an object and concludes that none of them are parallel to Florus 2.30.34. Ellipse of all sorts, including the object, is common in Latin and even if there were no instances with citare, it would not follow that Florus could not have written the transmitted text: in fact, there is no difficulty in understanding ‘defendants’ (not masses of Germans, as K.’s reductio ad absurdum [p. 41] postulates); one may compare the use of in senatum uocare and in consilium aduocare at Livy 2.55.10, 9.2.15, 23.32.3, 36.21.7. Normally, if I had argued that a transmitted text is unexceptionable, I would not feel obliged to discuss conjectures. Since, however, the conjectures are the purpose of the book, it would be unreasonable not to do so. As indicated above, K. (pp. 43–46, trailed on pp. 28 and 32) wants to change o securitas to securi ita se, making Florus say that by his previous treatment of the Germans, related by Florus at 2.30.31, he was in fact summoning himself to judgement: securi, by metonymy, means ‘execution’. Paleographically, the conjecture is not implausible: securi ita se could well have been corrupted to securitas (the addition of o would have been a subsequent change, since by itself securitas would lack syntax); as K. says, Gruter cited ad secures et from an early edition (in fact, as Titze believed [see pp. 47–8] Cologne 1474, one of the six incunables of Florus possessed by the John Rylands Library in Manchester), taken, as K. shows, from a now lost manuscript in Duisburg (later in Bonn). Were securi ita se the paradosis, I would not argue that Florus could not have expressed himself thus, but even if K.’s arguments for regarding o securitas as corrupt were compelling, securi ita se is not something that should be attributed to him by conjecture. (The only other previous conjecture is T. Faber’s illos for ille; see pp. 19–22). Three of the four remaining chapters (ch. 8 further arguments and remarks concerning the conjecture, ch. 9 other instances, both in Florus and elsewhere in Latin literature of the metaphorical use of citare and tribunal, ch. 11 possible objections) serve to expand K.’s case and I refrain from commenting on them. In ch. 10 (pp. 72–4), however, he presents another conjecture. K. objects to ex inprouiso following inprouidum, saying that such a repetition is «äußerst ungewöhnlich». First of all, Latin writers are tolerant of repetitions, of various kinds, which are offensive to modern taste; in this case, moreover, inprouidum refers to Varus’ lack of foresight, ex inprouiso to the suddenness of the German attack. K. bases his conjecture on the reading of B (he had access to online images; unfortunately, the directions he gives in n. 546 [the footnotes are numbered consecutively through the whole work, totalling 1588] proved unavailing). B, together with Jordanes, represents the A class of manuscripts and GNOMON 6/91/2019 J. Briscoe: Korting, Varus’ Untergang 510 has usually been regarded as the best witness to the text (see Marshall in L.D. Reynolds [ed.] ‘Texts and Transmission’ [Oxford, 1983], 165, apparently unknown to K.); on pp. 43–44, however, K. agrees with those who have stressed the number of errors it contains (Jal thought that the scribe had a very imperfect knowledge of Latin). When the C class of manuscripts presents a text which is both Latin and sense, one cannot safely use a manifestly corrupt reading in B as a guide to the truth. I should add that what K. says about the abbreviations in B suggests that his knowledge of abbreviations in Carolingian manuscripts is somewhat hazy (though Cappelli’s ‘Lexicon abbreviaturarum’ appears in his bibliography). B, according to K. (I do not understand his report of Havas’ note «inproaviso ([a] post i. s. l.)»), has ex inproauiso, the second i being superlinear between u and s. K. proposes exinde (he seems not to know that exin and exinde are alternative forms) pro auio ‘then with the remote area at their rear’ (or ‘using the remote area as protection’ or ‘to their advantage’). Suffice it to say that rhet. Her. 4.29, an invented example, is the only instance of the neuter singular of auius used substantivally (cf. TLL ii. 1447. 82 ff). I append some comments on matters of detail. – p. 6 n. 18: K. says «Hervorhebung von mir», but nothing in the note is emphasised. – p. 11: Malcovati, who wrote in Italian, is cited in German, presumably K.’s own translation, not the only occurrence of this practice. – p. 24: the examples K. gives from Livy (1.47.8, 27.24.5; he does not give the references here), like many others, are not instances of citare used «als terminus technicus im Bereich des Rechtes». – p. 46: K. refers to Paris, BnF Lat. 5690 as «Palat. Lat. 5690»; also containing Dictys of Crete, this is the famous manuscript of Livy, annotated by Landolfo Colonna and Petrarch, a primary witness for the fourth decade and one of the sources for the Spirensian tradition in books 26–30; see now M. Ciccuto, G. Crevatin, E. Fenzi, ‘Reliquiarum seruator’ (Pisa, 2012). – p. 59: K. fails to give the reference (2.13.27) for the passage he quotes. – p. 61: For Ennius ann. 488V = 461Sk see my commentary on Livy books 41–45 (Oxford, 2012), 36. –p. 66 line 5: for Glabione read Glabrione. The final appendix (pp. 135–174) begins with a list of «Allgemeine Hilfsmittel und Sammlungen», with a few exceptions consisting of abbreviations for websites providing access to online material (K. is clearly a digital expert and, like many younger scholars, who are rarely to be seen in libraries, consults as much as possible electronically). He says «Statt der URL wähle ich den link-Weg zu einer Webseite», by which he means the words which can be typed into a browser, not ‘link’ in the sense in which it is normally used in English. There follows (pp. 137–138) a list of abbreviations of «Hilfsmittel und Sammlungen zu den Altertumswissenschaften». Lewis and Short’s Latin dictionary, normally abbreviated to ‘L-S’ bizarrely appears as «FAL» (Freund, Andrews [K. says «Andrew»], Lewis); only Kühner’s original Latin grammar of 1878–1879 is listed: the standard edition, of course, is Kühner-Stegmann, revised by Thierfelder (1955). The rest of the bibliography is divided into six sections: 3, 4 and 5 contain editions of, commentaries on and translations of, respectively, Florus, other ancient authors and Greek and Latin sources for the German attack on Varus; section 6 consists of literature on Florus, other ancient authors and Latin in general, sec- GNOMON 6/91/2019 J. Briscoe: Korting, Varus’ Untergang 511 tion 7 of literature on Varus’ defeat, section 8 of other literary and historical literature. In section 4 (p. 154) there appears a translation of the whole of Livy by W.M. Roberts (1905) and n. 1462 says «Wikisource»; it is cited on seven occasions, six in the list of instances of citare. This is in fact the translation of Livy in the ‘Everyman’ series, published in six volumes between 1912 and 1924 (information from the British Library catalogue; I possess the first volume, containing books 1–5); it is unclear why, if he wanted an English translation, K. has not used the Loeb, as he does elsewhere. Manchester John Briscoe * Caterina Mordeglia: Animali sui banchi di scuola. Le favole dello pseudo-Dositeo (ms. Paris, BnF, lat. 6503). Firenze: Sismel – Edizioni del Galluzzo 2017. XIV, 150 S. 8 Taf. (Micrologus Library. 86.) 34 €. Attribués au grammairien de la fin du IVe s. Dosithée, auteur d’une Ars grammatica, dotée en partie d’une traduction grecque, les Hermeneumata Pseudodositheana (HP) sont un manuel scolaire bilingue grec-latin d’origine incertaine. Durant le Moyen Âge, cet ouvrage composite a rencontré un grand succès et a été diffusé dans toute l’Europe. Il nous est parvenu dans neuf rédactions, dont il est difficile de reconstituer les liens qui les unissent, comme c’est presque toujours le cas pour les œuvres de caractère scolaire. Bien que le recueil se soit modifié au cours du temps, on peut toutefois y reconnaître une structure quadripartite: (1) glossaires alphabétiques reprenant surtout des termes de nature verbale et secondairement adjectivale, (2) glossaires thématiques, (3) des exercices de conversation sous la forme de dialogues entre un maître et son élève, (4) des exercices de lecture à travers une anthologie de textes. Ce sont probablement les sections lexicographiques qui sont les plus anciennes. Les parties 3 et 4 ont sans doute été ajoutées plus tard. Les colloquia, dont Eleanor Dickey a donné une édition remarquable,1 constituent la partie la plus originale du recueil, non seulement pour la langue utilisée – l’Umgangssprache – tant au niveau lexical que syntaxique, mais aussi pour la description des pratiques didactiques. Les textes reflètent les lectures qui avaient cours dans le système éducatif impérial, à savoir les auctoritates littéraires traditionnelles, comme Homère, Ésope, Babrios, Phèdre et les réélaborations en prose et/ou en vers, mais aussi des œuvres de nature juridique (Tractatus de manumissionibus), gnomique (préceptes delphiques, Responsa sapientium, Hadriani sententiae) ou mythologique (Hygini genealogia). Sur l’époque et le lieu de composition des HP, il est difficile de donner des précisions. Les progrès des connaissances relatives à la pratique pédagogique à la fin de l’Empire romain ainsi que les papyri bilingues gréco-latins permettent de situer la production des HP entre la moitié du IIIe s. et le IVe s. dans un contexte où la connaissance du latin, promu comme langue officielle des provinces hellénophones, constitue une garantie de promotion sociale. La date de 207 mention- ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 1 Dickey, E., ‘The Colloquia of the Hermeneumata Pseudodositheana. I. Colloquia Monacensia-Einsidlensia, Leidense-Stephani, and Stephani’, Cambridge 2012; ‘II. Colloquium Harleianum, Colloquium Montepessulanum, Colloquium Celtis, and Fragments’, Cambridge 2015. GNOMON 6/91/2019

Chapter Preview



As a critical journal for all classical studies, the GNOMON fosters the links between the distinct classical disciplines. It has thus an exceptional position among the classical review journals and allows familiarization with research and publications in neighbouring disciplines. The reviews address an international readership from all fields in classical studies. The GNOMON publishes reviews in German, English, French, Italian and Latin.

The GNOMON is published in eight issues a year.


Der GNOMON pflegt als kritische Zeitschrift für die gesamte Altertumswissenschaft die Verbindung zwischen den verschiedenen Disziplinen der Altertumswissenschaft. Er nimmt dadurch eine Sonderstellung unter den Rezensionsorganen ein und bietet die Möglichkeit, sich über wichtige Forschungen und Publikationen auch in den Nachbarbereichen des eigenen Faches zu orientieren. Die Rezensionen im GNOMON wenden sich an ein internationales Publikum, das aus allen Teilgebieten der Altertumswissenschaft kommt. Die Publikationssprachen im GNOMON sind: Deutsch, Englisch, Französisch, Italienisch und Lateinisch.

Der GNOMON erscheint acht Mal im Jahr.