Bryn Mawr Classical Review 2001.08.35
Miroslav Marcovich (ed.), Eustathius Macrembolites. De Hysmines et Hysminiae Amoribus Libri XI. München and Leipzig: K.G. Saur, 2001. Pp. xxiv, 158. ISBN 3-598-71232-4. DM 98.00.
Reviewed by Ingela Nilsson, Department of Classics, Göteborg University (email@example.com)
Word count: 2177 words
Along with the growing interest in the ancient novels, their Byzantine successors have begun to receive their due share of attention. Four novels survive from the twelfth century, the so-called Komnenian period: one in prose and three in verse. The verse-novels (of which one has been only fragmentarily preserved) have all appeared in modern editions,1 but the prose-novel, Eustathios (or Eumathios) Makrembolites' Hysmine and Hysminias (HH), has not been edited since Hilberg's edition of 1876.2 Reprints of all the four Komnenian novels with an Italian translation by Fabrizio Conca were published in 1994, a volume in which Hilberg's edition of HH appeared with some forty changes of the text made by Conca himself.3 Despite the relatively high quality of Hilberg's edition and--not the least--the availability of the novel which Conca's reprint and translation has provided, a new edition of HH was necessary. Even though Hilberg had 22 manuscripts at his disposal, we now know of as many as 43 manuscripts, of which the important thirteenth-century E was not known to Hilberg.4 It is thus indeed gratifying to see that the Bibliotheca Teubneriana (now K. G. Saur) has decided to present us with a new edition of Makrembolites' novel, edited by Miroslav Marcovich (M.).
The volume opens--as is the usual practice in the Teubner series--with a brief preface in Latin. Usually, however, this preface restricts itself to the dating of the text and editorial matters such as the manuscript tradition, and it is thus rather surprising that a number of the editor's personal opinions on the novel have been included in the first part of the preface (pp. vii-x). M. writes, for example, that in HH "versus Homeri, Hesiodi, Euripidis ... citantur sine vera ratione" and "repetitiones odiosae abundant" (pp. ix-x). These are subjective judgements that are both inappropriate in this context and unfortunate from a literary point of view, since it may be argued that precisely the aspects that M. condemns are part of a literary strategy or poetic intention.5
More seriously, Makrembolites is introduced by M. as an imitator of both Nikephoros Basilakes and Theodoros Prodromos, which means that HH should be dated to the reign of Manuel I Komnenos (1143-1180) (p. vii). This is indeed one of the possible--and plausible--datings of the novel, but by no means the only one suggested. It has been argued that Makrembolites imitated Basilakes and Prodromos, but it has also been suggested that he was a model for those very same authors. A number of different views on the relationship between Makrembolites and other twelfth-century authors--and accordingly on the dating of the novel--are, in fact, found in the works included in M.'s list of dissertationes selectae (pp. xv-xviii),6 but M. still presents Makrembolites' imitation of Basilakes and Prodromos as a matter of fact, and thus misleads a reader who is not yet familiar with the novel and the twelfth-century context. It should be added that one of the problems with the dating of the novel is the name of the author, which has been differently transmitted in the manuscripts. In modern scholarship he is therefore variously referred to as Eumathios or Eustathios Makrembolites, and we are not sure who he was or when he lived.7 This is yet another piece of information that, in my view, a reader of a new edition is entitled to find in the preface. M.'s assumption that the author of HH has drawn material from other twelfth-century authors also has consequences for his execution of the edition, and we will therefore return to this issue in a while.
The second part of the preface (pp. x-xiii) concerns editorial matters, presents the manuscripts and discusses previous editions. M.'s discussion of the manuscripts and the stemma is too brief and leaves the reader with a number of questions. It is true that a detailed description of the manuscripts appeared already twenty years ago (Cataldi Palau 1980, see above, n. 4) and one may thus argue that a thorough discussion here would be redundant. This is, however, the first modern edition of the text, and such a survey is accordingly required, especially by readers who are not already familiar with the text and its history. Furthermore, when the reader turns to Cataldi Palau's study, he finds that her stemma differs from that of M.;8 a few words from M. on that matter would have been appreciated. It is also unfortunate that M. cannot, even in this part of the preface, restrain himself from evaluating judgements on the novel, as he opens the section with the following remark: "at miro modo Macrembolitae fabula erotica cum Byzantii tum in Europa plures lectores quam Prodromi poema elegans attraxit" (p. x). At this point--if not before--the reader realises that the editor does not like the text he has edited.
Let us now proceed to the text. On the whole, the present text does not, at first sight, differ much from Hilberg's edition. This may seem surprising, considering the doubling of the number of available manuscripts since the last edition, but there are two obvious reasons. First, a large number of the "new" manuscripts date from the Renaissance, and accordingly do not contribute to the establishing of the text. Secondly, the learned Komnenian novels do not display different versions (as is sometimes the case with their successors, the Palaiologan romances), and the manuscripts do not differ dramatically. An important change in the present edition is, however, the critical apparatus. Although Hilberg's edition did contain an apparatus, it was somewhat confusing and difficult to read, which means that M.'s critical edition based on all the available manuscripts indeed is a great improvement.
Unfortunately, M.'s critical apparatus is sometimes treacherous. M. tends to display a confusing mixture of negative and positive apparatus,9 which in combination with an unorthodox use of square brackets often confuses the reader and makes him unsure about which reading is actually in the manuscripts.10 Also, references to previous editors are inconsistent. Although it is certainly the editor's privilege to include and exclude information according to his own judgement, it should be the reader's privilege to obtain from his reading of the critical apparatus as clear a picture as possible of manuscripts and previous editions, since earlier attempts to understand complicated passages--especially when the manuscripts differ widely--are valuable to the reader.11 Hilberg's choices of readings are not only often omitted from the apparatus, he is also at least once erroneously referred to. We read on p. 109.7 (9.1.2) that Hilberg has added a καὶ to the text, but when turning to the Hilberg edition we find that Hilberg has marked the text with an obelus (Hilberg 1876, 145), and after some more investigation we find that the suggestion originally came from Karl Plepelits.12
Plepelits is constantly referred to in M.'s critical apparatus as if he were an editor, whereas Conca is hardly mentioned at all (I have found only one reference on p. 86.20). On the other hand, Conca is included in the list of editions on p. xxi, with the comment "Hilbergii textus cum correctionibus Caroli Plepelits", but Plepelits appears only in the bibliography (p. xvii). For a reader who is not familiar with the text's history of editions and translations, this is all very puzzling. Nor is it entirely correct because all of Conca's critical notes do not derive from Plepelits. The interested reader may find Plepelits' critical notes on HH in Plepelits 1989, 79-81 as Vorbemerkung to his German translation, and Conca's in Conca 1994, 56-60 as nota critica to the reprint of Hilberg.13
Despite the reputation of the Komnenian novels as simple and uncomplicated texts, HH does not lack complicated passages. A new edition may be seen as an opportunity to find solutions to some of the textual problems, and M. has made a number of additions, omissions, and emendations which make the texts he presents different from that of Hilberg's. In some cases, M. has as the first editor of HH chosen to print a reading that is unanimously transmitted in the manuscripts but that previous editors have found problematic.14 This is a positive step towards an establishment of the original text, and, in my view, such changes could have been made also in a number of other cases.15 But there are also readings common to all manuscripts which still seem inexplicable unless you consider them as archetypal errors, such as the mysterious character Kallisthenes, who appears in HH 5.10.3 and stays until 6.1.3. Le Bas (1856) introduced the conjecture Kratisthenes, which indeed is consistent with the plot, but the unanimous transmission of the name Kallisthenes in all manuscripts is still a fact which deserves some further scholarly attention.16
A general problem of an edition without translation and commentary is the difficulty understanding how the editor has interpreted the text when choosing the readings he has printed in the text, and thus to appraise his choices. Any criticism may accordingly seem arbitrary, but I will still give an account of the aspects of M.'s edition that I find problematic and not in accordance with the text's nature.
M. has a predilection towards additions, both those made by previous editors and his own contributions. Additions are always hazardous, especially when they are not really required by the text,17 or even seemingly made without consideration to the text's general character.18 A redundancy of additions may give the impression that the editor is establishing not the text that the author intended to write but the text that the editor would have wanted him to write. Makrembolites' style is short, metaphorical, and poetic, sometimes even enigmatic, and M.'s additions rarely attempt to reflect its peculiarity.19 The same tendency is unfortunately indicated also by some of M.'s omissions and conjectures, many of which show little understanding of the text's particular character.20 Others do not seem reasonable, either from a linguistic or from a narrative point of view,21 and others are even grammatically incorrect.22 These interferences with the text all contribute to a sense of the editor's unfamiliarity with the novel, which is implied also by some of his choices of readings,23 his punctuation,24 and his use of cross-references to other passages of HH, which I often find difficult to understand.25
A central aspect of literary studies of Byzantine texts is the relation of the Byzantine authors to ancient and/or Byzantine literature. When it comes to editions, this concerns both the apparatus fontium and, in the case of allusions and quotations, the actual text. I have already mentioned above M.'s assumption that Makrembolites imitated the rhetorician Nikephoros Basilakes. The result is an abundance of references to Basilakes in the apparatus fontium, but also, which is more serious, the use of Basilakes as an authority for establishing Makrembolites' text (even in cases where all the manuscripts of HH transmit the same reading).26 As long as the relationship between Makrembolites and Basilakes has not been firmly settled, I find this procedure misleading.
Similar interferences with the text are made by means of conjectures (most often made by previous editors, and accepted by M.) based on ancient authors, above all the tragedians.27 This kind of manipulation of the text through the adaptation of "correct" quotations displays an antiquated approach towards textual criticism and Byzantine texts. Just like their ancient predecessors, Byzantine authors quoted from memory, with the result that quotations often vary from the original. Corrections of words, word-order, or metre therefore damage rather than benefit a text.28 The task of a modern editor should be to remove such "corrections" and attempt to re-establish the original text.
My general impression is that M. has printed the text that he thinks that Makrembolites ought to have written and has provided it with an apparatus fontium that mirrors his own view of Makrembolites' relation to Achilles Tatius, Theodoros Prodromos, and Nikephoros Basilakes. The stress laid in the first part of the preface on Makrembolites' imitation of Tatius' Leukippe and Kleitophon and Prodromos' Rhodanthe and Dosikles 29 continues throughout the apparatus fontium. M. includes all kinds of parallels, including novelistic topoi. The result is sometimes astonishing, as when we read that Makrembolites simultaneously imitates the novels of Longus, Tatius, Prodromos, Eugenianos, and Manasses (pp. 18.9-20.5). The traditional view of Makrembolites as a mechanical imitator is thus--again--on display. Of course, M. cannot be censured for holding that opinion, but it should not have been allowed to influence his edition to such a large degree.
The interest in the Byzantine novels, and in Byzantine literature on the whole, will most probably continue to grow. Many of the texts are--contrary to their reputation as simple in both language, structure, and plot--complex and sometimes linguistically difficult texts, which demand a lot on the part of the reader. Appropriate tools are needed: editions, commentaries, and translations. It is nice to see Hysmine and Hysminias appear in a modern edition. It is, however, rather disappointing that it appears without commentary or translation, and--above all--with such a prejudiced and antiquated preface, and an apparatus fontium which to a new reader may give a false impression both of the text itself and of the current research situation.
1. Theodoros Prodromos' Rhodanthe and Dosikles was edited by M. Marcovich (Stuttgart 1991) and Niketas Eugenianos' Drosilla and Charikles by F. Conca (Amsterdam 1990). The fragments of Konstantinos Manasses' Aristandros and Kallithea appeared in two editions in the same year, by E. Tsolakis (Thessalonike 1967) and O. Mazal (Vienna 1967).
2. Eustathii Macrembolitae Protonobilissimi De Hysmines et Hysminiae amoribus libri XI. Recensuit Isidorus Hilberg. Vienna 1876.
3. Il romanzo bizantino del XII secolo. Teodoro Prodromo - Niceta Eugeniano - Eustazio Macrembolita - Constantino Manasse. A cura di F. Conca. Turin 1994.
4. On the mss of HH, see A. Cataldi-Palau, "La tradition manuscrite d'Eustathe Makrembolitès", Revue d'histoire des textes 10 (1980) 75-113. On the ms E (Cod. Baroccianus 131), see also N. G. Wilson, "Barocci 131 described", JÖB 27 (1978) 157-179.
5. This was argued already in the 1970s by M. Alexiou, "A Critical Reappraisal of Eustathios Makrembolites' Hysmine and Hysminias", BMGS 3 (1977) 23-43. See now also I. Nilsson, Erotic Pathos, Rhetorical Pleasure. Narrative Technique and Mimesis in Eumathios Makrembolites' Hysmine & Hysminias, Uppsala 2001.
6. E.g. the works of Cupane and Poljakova, MacAlister 1991, Magdalino 1992, and Plepelits 1989. For a survey of the different datings of HH, see Nilsson 2001, 16-19.
7. In fact, an article on precisely this aspect of the novel, i.e. the full name of the author and the dating, was written by the dedicatee of M.'s edition, Herbert Hunger: "Die Makremboliten auf byzantinischen Bleisiegeln und in sonstigen Belegen", Studies in Byzantine Sigillography 5 (1998) 1-28. Hunger's proposed biography identifies the author of HH with an imperial dignitary by the name Eumathios Makrembolites, and it has been assumed that this person--if he is indeed the author--wrote the novel at a young age while he was still a notary, ca. 1130-1135; see P. A. Agapitos, "Poets and Painters: Theodoros Prodromos' Dedicatory verses of his Novel to an Anonymous Caesar", JOB 50 (2000) 173-185, esp. 184-185, and Nilsson 2001, 18-19.
8. Firstly, M.'s stemma does not indicate the relation between the archetype and the manuscript group alpha; this may, however, be due to a misprint which has by accident left out a line in the stemma (p. xiv). Secondly, M. has removed Cataldi Palau's hyparchetype delta (cf. Cataldi Palau 1980, 113).
9. There are numerous examples, but see e.g. 1.6.2 (p. 5.11-12).
10. M. uses square brackets to indicate words or parts of the text which he wishes to omit, but which are also omitted in some of the mss. Traditionally, square brackets are in such cases considered unnecessary, or even wrong. There are a number of examples, of which one appears in the above example (n. 9).
11. Since minor variations that appear in Hilberg are quite often included in the apparatus, the reader is led to expect all variations to appear. That is however not the case, since references to Hilberg's text are often left out, e.g. on p. 87.21 (7.14.1), where Hilberg's word order is not mentioned. Also, Hilberg's apparatus (Hilberg 1876, 118) implies variants of the mss, which are not mentioned by M. Another example may be found on p. 151.15 (11.22.3), where neither Hercher's nor Hilberg's readings are recorded. Yet another puzzling case is the omission of Hilberg's reading on p. 98.18 (7.7.5). To judge from M.'s apparatus, the variants that appear in the manuscripts are ἀφαιρεῖ and ἀφαιρεῖται. Hilberg's reading (Hilberg 1876, 132) is, however, ὑφαιρεῖ, and his apparatus does not record any emendation. This may, of course, be a mistake of Hilberg's, but it would have been appropriate for M. to comment upon it. These kinds of inconsistencies all contribute to the reader's sense of insecurity and confusion.
12. K. Plepelits, Eustathios Makrembolites: Hysmine und Hysminias. Eingeleitet, Übersetzt und erläutert, Stuttgart 1989, 80.
13. A final remark on the general appearance of the edition: the text does not contain many misprints (I have found only four in all). I will mention here just one thing that may save the reader some pondering: in the critical apparatus on p. 73, I assume that we should read "19 ἡ μὲν γὰρ δὴ--20 ἐξήγειρεν om. E" and "20 τὸν θρῆνον--21 θρηνούση om. G".
14. M. dismisses, e.g., Le Bas' emendation of HH 4.14.3 (p. 42.18) κεκρυμμένα πτηνά and prints the κεκρυμμένοι πένητες of the mss (as did Conca 1994); cf. Plepelits 1989, 51-53, 80. See also 7.8.2 (p. 83.18) ἐκείνου and διπλοῦ. I am less convinced by 2.7.4 (p. 17.19), where M. has chosen the ὦ μειράκιον of the mss instead of the widely accepted emendation τὸ μειράκιον, since I find the change of addressee which such a reading implies puzzling.
15. 4.8.1 (p. 39.18) the reading of the mss is plausible if τόξον is understood metaphorically, i.e. flax is, in the form of a snare, the farmer's "bow" (i.e. weapon); 4.12.4 (p. 41.21) considering the context, the reading of the mss, σφάττει, is reasonable, and I do not understand what caused the emendation in the first place; 8.13.4 (p. 103.6-7) the accusatives of the mss are highly plausible; 6.15.3 (p. 77.16) Hercher's conjecture is unnecessary considering the text's repetitive structure, in the light of which the repetition of three superlatives has a stylistic function; 6.10.6 (p. 72.16-17) Hercher's conjecture is unnecessary since the datives of the mss may be read instrumentally; 7.12.2 (p. 86.20) M.'s transposition of words is puzzling, and the text is readable as it stands in the mss; 7.16.2 (p. 89.7) Hercher's conjecture is unnecessary, since the reading of the mss is plausible; 9.13.4 (p. 116.13) Hercher's conjecture disturbs the rhetorical pun of the passage, which consists in the repetition of γάμος with two different verbs ("not being able to endure even the sound of it we fled the marriage and bought the (i.e. our own) marriage"); 9.14.3 and 9.15.1 (p. 117.3 and 117.17) Plepelits' transpositions are unnecessary, since the author is not required to use exactly the same expression everywhere (cf. Plepelits 1989, 80-81). HH 3.4.3 (p. 26.24) is interesting: Teucher's addition -κις does not seem necessary, since the verb ἑκατονταχαίρω may have appeared in analogy with words such as ἑκατοντάφυλλος or ἑκατοντάχρους, in which case the reading of the mss is perfectly plausible (cf. LBG, which--surprisingly enough--has included Hilberg's conjecture ἑκατονταχαίροις) as a lemma). Yet another solution would be to read the passage as ἑκατὸν τὰ χαίρε).
16. My previous comments on this question (Nilsson 2001, 156, n. 424) are badly worded, and I also have to admit that I am beginning to doubt my prior conviction that Kallisthenes should be emended to Kratisthenes. In 7.3.1, Hysminias says to Hysmine "you are presumably not unaware of Kratisthenes, who sailed with me to your homeland of Aulikomis? He is my fellow countryman, my cousin, my alter ego", and Hysmine answers "I have attended on him, and mixed wine for him with this hand". Now, why would this conversation take place if Kratisthenes was indeed the person who served the wine in Eurykomis (5.10.3-6.1.3)? It is also likely that Makrembolites would have enjoyed the pun that the similar names Kratisthenes and Kallisthenes form, both in analogy with other names in the novel (Sosthenes-Sostratos) and as a hint at Leukippe and Kleitophon.
17. E.g. 2.5.1 (p. 14.22) Hercher's παρθένος is unnecessary, cf. 2.2.1 and 2.4.1; 5.9.1 (p. 56.24) Hilberg's παρά is not needed, since the meaning here is "you are welcome, Sosthenes, and I offer thanks to you, Zeus Xenios, for your reception of the herald"; 7.11.3 (p. 85.26) M.'s ἐπισχῆ is redundant; 8.3.2 (p. 95.27) M.'s ἔμενε is redundant; 8.7.5 (p. 98.18) M.'s δ̓ is redundant; 8.9.2 (p. 100.1) M.'s στρατῶ is unnecessary, since the text is plausible, especially with the accusative ἄλλο of RCM; 8.14.4 (p. 103.23) Le Bas' ἐκ is unnecessary; 8.16.2 (p. 105.1) M.'s κατ is not required.
18. E.g. 6.10.1 (p. 71.20) Hilberg's θύομεν is unnecessary and stylistically unsatisfactory, and the "parallels" that M. adduces are not comparable to this construction; 9.14.1 (p. 116.18) M.'s δοῦλαι is unnecessary and the reference to p. 117.2 is puzzling; 11.17.3 (p. 148.14) M.'s πλῆθος is redundant, cf. 1.2.2 (p. 2.2-3) πᾶν τὸ ταύτης. Cf. also the lacuna suspicion by M. in 3.5.2 (p. 96.26), where the text is in fact plausible in the sense that the song "saw the trireme adorned with a white sail" (i.e. the pirates sang while setting sail). An attempt to improve a passage by means of addition is also made in 10.20.2 (p. 121.17), where M. both includes the reading of KEJ ὅρον and adds πρὸς to the text; this may be correct, but I do not understand the meaning of ὅλον ὅρον, and have not been able to find the term in any other source. It seems, then, that the solution of previous editors was more reasonable, since the text is comprehensible as it reads in a number of mss.
19. E.g. in 8.7.1 (p. 97.22-23), which is one of the truly problematic passages of the novel, M. has added three words to the text, apparently based on 8.9.3 (p. 100.11). In my view, this is not convincing, nor does it solve the problems of the passage. The question here is not whether the addition makes the text better, but how one can explain its absence from all the mss (provided that one is indeed trying to establish the original text). Since the mss do not imply any textual problem here, one should probably assume that φέρει here is used in a particular sense, namely "considers" or "holds", in which case one could indeed compare p. 100.11. Cf. also 11.17.2 (p. 148.10) χρυσὸν which is not actually M.'s addition, but a conjecture from a 16th-century ms (see Cataldi Palau 1980, p. 79). But if χρυσὸν was originally part of the text, how did it become excluded from all mss, later to appear in the 16th century? Since the text does not require any word here, it is more likely that χρυσὸν is an interpolation, and we must again ask ourselves what kind of text M. wishes to establish.
20. In 1.4.3 (p. 3.14-15) the words that M. wishes to omit are not likely to be a dittography but most likely a poetic repetition; in 5.7.1 (p. 55.24) the exclusion of παστάσι seems entirely arbitrary and is not consistent with the text's poetic repetitive pattern (cf. Nilsson 2001, 64-74); cf. M.'s expressed dislike of repetition in the preface, pp. ix-x.
21. E.g. 7.1.3 (p. 80.12) M.'s omission of οὐκ changes the meaning radically, but the point here is that Hysmine scolds Hysminias for his neglect of the divine signs: both Zeus and Eros hint at their love, but Hysminias has so far not come up with any plan in order to prevent their separation (in fact Kratisthenes has, but Hysmine does not yet know that); 8.2.3 (p. 95.16-17) something has probably happened to the text, but it is more likely that κατεπιβάντες καὶ should be omitted, since συνεπιβάντες ἀυτοί requires τούτοις in order to make sense; 8.9.1 (p. 99.24) there seems to be no reason to omit ἡμᾶς just on the basis of E; in 2.9.3 (p. 19.5), 9.17.3 (p. 120.3), and 10.2.2 (p. 124.17) M.'s conjecture κἂν is unnecessary, since καί may well be used; 7.9.5 (p. 84.19) M.'s conjecture γάμος is most probably influenced by previous attempts to emend the text, but there is no reason to change the τάφος which is practically unanimously transmitted by the mss, since it is part of the paradoxical "bride in Hades" motif which is frequently used in the following passages 7.10.4-5 and 7.11; 8.17.1 (p. 105.21-22) M.'s exclusion of τῆ γλώσση and his conjecture μεστῆ both disturbs the repetitive scheme of the passage and changes its meaning, which, in my view, is as follows: "lest Eros should, unbeknownst to me, use my tongue to corrupt my soul, or even the tongue itself, which I shall keep for the maiden Hysmine to be full of the passionate graces and the honey of Aphrodite which we devoured insatiably with our tongues alone".
22. E.g. 2.14.5 (p. 22.11) M.'s conjecture, the active ἀποδύσας, is grammatically incorrect, since middle voice is required (all the mss except one have ἀποδύση δὲ, which is a correct form); 4.3.8 (p. 53.12-13) Osann's ἐξυφανώμεθα is non-classical, whereas the majority of the mss transmit the correct form ἐξυφανούμεθα.
23. In 1.8.1 (p. 6.2) the longer reading breaks off the textual flow and is not consistent with Makrembolites' style. Such a choice seems to be made in line with M.'s general tendency to fill out the text, but also the opposite tendency may be seen, e.g. in 3.8.2 (p. 31.23), where τῶ is omitted, which disturbs the textual flow (cf. 4.2.3, p. 36.9). In 6.2.5 (p. 66.19-20) the reading in all other mss is plausible, so why choose that of R? In 4.6.2 (p. 82.22) the majority of the mss transmit the aorist subjunctive, which is also what one expects in such a construction. M. may have chosen the future optative because that is the reading of K, which is indeed a good ms, but in many other cases the readings of K are dismissed (for example, in the first two examples above). See also 7.16.2 (p. 89.6) where ὅλον seems to be part of the repetitive pattern, as does λαμπρόν in 8.20.1 (p. 107.8); 8.14.4 (p. 103.23-24) παρθενική is often in the novel used as an attribute to laurel, which makes the singular most plausible here, and the dative σοί in the next line is the most probable reading. In 11.5.3 (p. 141.13) the majority of the mss transmit μεταπλάττω, which is plausible albeit asyndetic.
24. See e.g. 8.18.2 (p. 106.6-7), cf. Hilberg and Conca. Note also 9.17.2 (p. 119.24), where ἀλλ̓, is most probably part of Hysmine's reply, ἀλλ̓, οὐ τοῦ νῦν etc.
25. E.g. in 4.13.5 (p. 42.7) where M.'s conjecture πληγήν is made with a reference to pp. 43.12 and 49.6, which makes no sense to me. Again, one wishes for a commentary.
26. See e.g. 5.19.3 (p. 64.7); 6.10.6 (p. 72.16-17); 9.12.3 (p. 115.12).
27. In 3.9.6 (p. 33.12), the form ἀμηχανώταται is based on seven late mss and Euripides, whereas all other mss and edd have ἀμηχανώτατοι. It is likely that the late mss have a corrected form in accordance with Euripides, and that the older mss represent the original text. In 6.14.7 (p. 76.15) Hilberg changed the θρασύτητα of the mss into the τραχυτῆτα of Aeschylus, and in 6.15.3 (p. 77.10) he changed the λόγος of the mss into the μύθος of Euripides. In 9.23.1 (p. 123.3) M. has added the δ̓ of Euripides, and note also Le Bas' conjecture ἐς for εἰς of the mss. Euripides' ἠδικημένη appears only in late mss. In 9.23.2 (p. 123.8) Osann's γάρ does not appear in any ms. There is also Hercher's conjecture from Homer in 3.1.1 (p. 23.1). There is no reason to suspect that the text originally had the dative; why would someone then change it into the accusative? Not all quotations are, however, corrected, see e.g. 8.12.5 (p. 102.6) or 10.6.5 (p. 127.8), which makes you wonder about M.'s principles in these cases. The form γλώττης in 1.13.2 (p. 10.6) is indeed confusing: since it stands out against the usual form in the novel (see the same line, γλῶσσαν) one may get the impression that it is part of the quotation from Hesiod, but it is not. Le Bas, Hercher, Hilberg, and Conca all have γλώσσης. If all the mss transmit γλώττης, which is the conclusion one must draw from M.'s apparatus, it is quite remarkable and should have been commented upon; either all the previous editors changed the text without mentioning it, or M. did.
28. A related matter is M.'s apparent wish to find and correct "hidden verses" in HH. E.g. in 8.14.4 (p. 104.3) the word order has been reversed (first by Le Bas) so that the sentence forms a correct verse. We must ask ourselves if this really is necessary. How, if the verse was "correct" in the original text, did the words change place in all the mss? And since we cannot attribute the verse to any poet, why should we read it as anything more than a maxim in prose? The same questions apply to 11.11.2 (p. 145.1), where we also ask ourselves how M.'s προσκαθιζάνει could possibly have resulted in the προκάθηται of KE? In 1.13.5 (p. 10.8) φειδωλῆς has been moved by Hercher in order to achieve a full verse. This is not necessary, since the quotation is full in all mss, even if not metrically correct. And again: how did the words change place in the first place if the word order was not part of the original text? Cf. also the remark in the apparatus on p. 86.5.
29. On Leukippe and Kleitophon, see p. vii (Makrembolites' novel, writes M. in the very first sentence, "Achillis Tatii Leucippen et Clitophontem tamquam umbra hominem sequitur"); on Rhodante and Dosikles, esp. pp. vii, n. 2 and pp. viii-ix.