The way Florian Hurka begins his response — that he feels compelled to respond because of “the numerous factual errors and misunderstandings” in my recension — led me to expect that a demonstration would follow. But when I read Hurka’s response, it is sometimes hard for me to see that we are referring to the same book (Florian Hurka, Textkritische Studien zu Valerius Flaccus). For example, Hurka writes:
In 1989 Taylor studied the Humanist’s notes in his working text to Manilius to strengthen the view that Carrion collated faithfully. In my study I emphatically approve this opinion of Carrion’s credibility in view of the variant readings of D (p. 17, last line; Roggen seems to accuse me of the opposite …)
First of all, I think “accuse” is not the verb that should be used here: what I did was to sum up the author’s view regarding the crucial point, whether or not we have reason to trust Carrion in his description of his manuscript. After reading the chapter again, I still think I paraphrase correctly in my review: “H. finds that we should suspect Carrion of dishonesty in the presentation of readings from C …”. I even included a quotation, namely the first part of a very long sentence: “Dürften …”. The expression “seine übrigen Schwindeleien” is used (p. 17, n. 42). The content of the latter part of p. 17 simply cannot be summed up the way Hurka does in his response, leading me to wonder whether Hurka’s reference to p. 17 is wrong? But the same line of thought is expressed on p. 18:
Nicht nur mag der junge Carrion versucht gewesen sein, den entdeckten Zeugen als besonders wichtig erscheinen zu lassen, weshalb er fremde oder eigene Konjekturen für ihn in Anspruch nahm, sondern er trachtete möglicherweise zugleich danach, seine Qualitäten als Textkritiker unter Beweis zu stellen, weshalb er falsche Lesarten seines Kodex nur verzeichnete, um sie in souverän erscheinender Manier ablehnen zu können.
These are not reflections that are negated; what follows is Hurka’s principle that all variants that Carrion knew from other editions should be disregarded. The sentence is quoted in my review, paragraph 2 (“Aus metodischen Gründen …”). On the basis of what we have just read, it is now really surprising to read Hurka’s explanation of how the principle should be interpreted: “I am referring, of course, to the testimony provided by Carrion himself in his Scholia and Castigationes” (Response, paragraph 3). It is difficult to see how this might help if the problem is not that Carrion commits mistakes, but that he presents false readings on purpose. (Cf. the Response: “… falsely claims variants of the codex as his own emendations”.) But of course, it would have been very nice of Carrion (although not so intelligent) to note in what editions he had found his conjectures, so that his readers could have an easy way to detect the thefts. The way Hurka described the method in his book, this would have been valid, in my opinion, but difficult to carry out, and this was what I criticized (paragraph 2). The explanation in the response only makes things worse.
I assume that one of the numerous misunderstandings in my review is that I failed to understand this method. Among the factual errors I have found one, which I of course regret: I misunderstood Hurka’s expression “am Rand seiner Textausgabe” (p. 17, note 41) and used the expression “edition”. Hurka’s point seems to be that Carrion may well have collated conscientiously for his own, private use; this, however, does not prove that he was honest in a publication. So we are back to the question of Carrion’s honesty.
As a defence against my criticism that in his book he “takes the coniector doctus for granted, without arguing his view “, Hurka refers to pp. 22-36. I can only ask the reader to read the chapter, where my words are confirmed from the first line onwards. Where method is concerned, Hurka refers to pp. 16-38. This chapter is entitled “Der Codex Carrionis“, and is not dedicated to method in particular. For example, what sorts of arguments are valid in the discussion of genuine or interpolated verses? On pp. 87f an erroneous quam in L is used as an argument for the genuineness of verse 2.565a: “An einen Interpolator zu denken, dem ein solcher Fehler unterlaufen sein sollte, ist abwegig.” On the other hand, Hurka argues that “Die C-Lesart saevit amor muss auf Konjektur beruhen (sie ist inhaltlich unmoeglich)” (p. 23, on 2.398). The way of arguing is not clear, at least not to this reader, and I repeat: “In particular, one misses a chapter on method.”
In his book Hurka presents an interesting discussion of the handwriting in Delta in comparison with Carrion’s description of his manuscript, and I can understand his frustration when I sum up my opinion with the words that it “does not convince”. Let me go through my line of thought here. Carrion writes in his Castigationes to 1, 338 that most books and manuscripts have Tolum, while he agrees with Sabellicus and codex Lugdunensis, which have Folum. Carrion thinks that the origin of this error was a similarity between the letters, so that the scribes ” non F, sed T, esse arbitrarentur. (nam in veteri meo codice, forma huius literae quam simillima est literae F) et Folum in Tolum permutarunt.” (quoted from Hurka, p. 20). As one sees, Carrion uses capital letters here, and, as Hurka observes, capital F and capital T are not particularly similar in fragmentum Valerianum. But I would like to argue that Carrion does not necessarily have capital letters in mind. Carrion comments on 5, 151: ” Macrones. Vet. cod. Magrones … idque ex affinitate litterarum C. et G.” (Hurka, note 63). As one sees, Carrion uses capital letters in his commentary, even though the letters he treats are lower case. The small f and the small t in the fragment are more similar in shape than the corresponding capital letters. But if we return to Carrion, I find it likely that he used the similarity between the (small) t and f the way philologists do: as an explanation of how the error might have arisen — at an earlier stage in the transmission. Obviously, since he explicitly says that his manuscript has Tolum (tolum), not Folum (folum), there was a noticeable difference between the two letters in C.
The rediscovery of Delta confirms Carrion’s report of his manuscript, through the common variant readings and the fact that it ends with the same verse. Regarding Delta, we know that the last verse is at the bottom of a verso page, and it seems likely that the manuscript originally had a continuation. When it comes to C, Carrion writes of 8.105: ” Hic est postremus versus vet. mei codicis manuscripti, reliqui temporum iniuria perierunt” (Hurka, note 50).
One of my main points is that a logical basis for the treatment of the question whether or not Carrion’s manuscript represents (represented) an independent branch of the textual tradition, is a certain trust in Carrion and his reports. If one concludes that there is good reason to suspect Carrion of “Schwindeleien”, the good variant readings that Carrion reports may be either genuine or results of conjectures made in any period: during the Middle Ages, in the humanist period (falsely claimed by Carrion as manuscript readings), or by Carrion himself (likewise). In that case, I don’t see how it can be possible to identify which is which.
Hurka’s intention, as pointed out in the introduction, is to try to prove that at least one of the variant readings is genuine, and on this basis establish C as an independent textual witness (p. 8). His crucial verse is 7.633. He argues that if the same person is responsible for this as for the other “Texteingriffe”, the coniector doctus did not write the verse, since he would have written a flawless and more beautiful verse. In my review I quoted the sentence and stated that I find it meaningless and even absurd. What Hurka has demonstrated is that coniector doctus (let us assume that such a person existed) did not write the verse. I think perhaps he meant to say that if no other interpolator wrote it either, it must have been written by Valerius Flaccus. In his response, Hurka claims regarding verse 7.633: “It goes without saying that the verse could have been added either before or after C was emended (p. 62)”. This is not stated explicitly on p. 62, but is, perhaps, what is meant by the reservation “Unter der Voraussetzung …” in the problematic sentence. But the issue is how Hurka gets from this point to his conclusion: “The sum of the arguments seems to support the thesis that C is independent (p. 63).” It is by no means clear why it is more likely that Valerius Flaccus himself wrote what Hurka characterizes as “einen … unschönen Vers” than that a late interpolator wrote it. I may be wrong, but it seems to me that the problem is connected with Hurka’s way of arguing, through reservations and conditional clauses.
Incidentally, it is not correct, as Hurka asserts in his response, that the variants in 6.3 are tuentem/videntem; here he repeats an error from the conclusion of his book (p. 124). The presentation on p. 62 is correct: tueri in C, videri in Gamma. If we may trust critical editions of Valerius Flaccus (and for that matter, Hurka’s book, note 142), tuentem is not a variant reported from C, but Heinsius’ conjecture for videntem in 7.503.