Response to Giambattista D’Alessio’s review of The Kallierges Pindar: A Study in Renaissance Greek Scholarship and Printing (2015)

My book is an interdisciplinary study, in which the first half deals with my discovery that as many as ten sheets of the 1515 Kallierges Pindar have been reset, most importantly at the beginning of the Pythian odes, where the first four quires (64 pages) have the Greek text in two textually different versions, which I call Variant a and Variant b. The second half of the book deals mainly with the perhaps even more important discovery of a unique and hitherto unknown document consisting of three densely set pages of Greek prose whose contents are sensational from a book historical point of view.

The major disagreement between D’Alessio and myself concerns the correct order of priority of the eight reset sheets of the Pythian odes. After a long and careful physical examination of the book by applying methods of analytical bibliography, I was finally able in 2015 to present the result of 25 years of research, based on the evidence of 227 copies from upwards of 80 libraries on two continents. The collected evidence indicated overwhelmingly that the variant which was ultimately called Variant a was printed before Variant b — hence the denominations. D’Alessio, however, is convinced that Variant b was printed first.

Our disagreement has led to a clash between two different methods of textual research. D’Alessio founds his opinion entirely on his own assessment of the merits of conflicting variant readings in the two settings. This is fine for an editor who is going to publish a new Greek edition, but is this a reliable tool also for deciding the question of priority?

In chapter 4 (212) I explain why I have made little use of “one specific line of argumentation that might seem relevant to the issue” [sc. that of priority] “namely, which of the two variant readings is textually preferable in each case and therefore likely to represent the second setting […]”. To show the fallacy of this method I compared the stands taken by a number of modern editors that showed that they were divided in their preferences in the choice of variant readings: “To a great extent, assessment of variant readings always was and always will be a matter of personal preference.” In other words, it is useless as evidence in a case of priority and that is why I have refrained from using it, contrary to D’Alessio, who makes it his preferred category of evidence.

There is another method, however, that can give reliable answers, if the questions and conditions are right. It is known as analytical bibliography and involves an examination of the physical properties of the text-bearer, and, by contrast, can yield 100% exact and reliable answers. This is the method I have applied most of the time throughout the book, as it is the perfect tool for answering my questions. It relies not on arbitrariness or personal opinion, but on hard physical evidence that can never be disputed or explained away and which is of an enormously higher evidential order than a discussion of the supposed merits of this or that variant reading.

Like the great majority in the world of classical scholarship D’Alessio obviously has never himself practised or come into contact with this method and so may be excused for not realizing its enormous potential for supplying relevant and trustworthy evidence. For the question of priority can only be answered, or at least answered satisfactorily, by the application of bibliographical method; this is exactly what I have done in my book. But seventy pages of hard core evidence (120-194) cannot be repeated here.

It is true that regarding the numerous arguments for the priority of Variant a that I have adduced, I have said that no one single argument alone is strong enough to settle the matter once and for all (336). However, in doing so I was referring to the kind of decisiveness with which the word is used in the sciences: I was simply anxious not to overstate my case. D’Alessio, however, seems to have misunderstood my caution as a confession that I myself have doubts about the priority of Variant a and about other matters that are expressed in an undogmatic way. This is not the case: many of the arguments based on bibliographical evidence are absolutely compelling, if we are to rely on sound logic.

There is no point in confronting his various arguments one by one as they will all disappear at once as soon as it has become evident that Variant a was set before Variant b. And such evidence there is. But since D’Alessio — who never tries to disprove any of my many bibliographical arguments — seems to be blind even to the most obvious facts, let me adduce one that gives conclusive evidence once and for all that Variant b was set after Variant a. And this is a new argument.

Bibliographical facts have made it evident that Kallierges began printing his Pindar on two presses: one had its signatures in minuscules, the other in majuscules. Kallierges was compositor for one of the presses, his foreman for the other, but the famous printer of course supervised the work at both presses. One press was going to print the Olympian odes and their scholia, the other, the Pythian, Nemean and Isthmian odes. This means that one press started with the first Olympian ode and the other, with the first Pythian ode. (The preliminaries were set and printed last as usual.)

The early printers went to extreme lengths to make their books as perfect as possible. And there was one thing that no printer, be he good or bad, would ever do: he would never change the layout of the text once he had made up his mind and started the typesetting.

The typesetting for the two presses started. Kallierges was the boss and he decided to use the layout we meet on the first page of the Olympian odes. Needless to say, his foreman at the other press was ordered to apply the same layout for the Pythian odes — and so he did, thereby creating what I have been referring to as Variant a; moreover, the layout of Variant a is exactly the same as that of the rest of the book, thereby proving its original status, as explained in the previous paragraph.

According to D’Alessio, however, the text of Variant b was set and printed before that of Variant a. That would require that Kallierges would have to have had a total mental breakdown the first day on the job and ordered his foreman to apply a strikingly different layout for the Pythian odes, one with negatively indented left-margin emphasis, headings in majuscules instead of minuscules, etc. It would also require that Kallierges did not recover from his unfortunate mental breakdown until after his foreman had set and printed the fourth quire (i.e., after 64 pages), when Kallierges suddenly became his normal self again and ordered his foreman from now on to apply the same layout as the one he himself used for the Olympian odes.

Needless to say none of this ever happened, and could not have happened, and, honestly, I do not think that D’Alessio would still claim that it did.

Yet someone set the four quires of Variant b. When was it done and why? These questions are fully answered in my book and need not be repeated here in detail, but, as so often, D’Alessio hides vital information from his readers, for example, that also paper evidence proves beyond a shadow of a doubt that the resetting was done as an extra job after all the regular quires had been printed. This fact is mentioned a number of times in my book (192, 194, 234, 259-61, 275, 395, 402); hence, D’Alessio knows this very well, as well as the fact that this is still another incontrovertible piece of evidence to prove that Variant b was printed after Variant a.

Why was Variant b, a special edition, designed to stand out typographically? A full answer to this question can be found in chapter 9. If D’Alessio had read that chapter, he would not have suggested that Variant b was printed first.

Furthermore, D’Alessio thinks that the copy with the dedication to Musurus was “printed before the preliminaries were reset (and, again, I see no obstacle in the ‘analytical evidence generated by the physical examination’).”

My physical examination of the layout of the preliminaries of the “Musurus” copy shows, again without the slightest doubt, that they were reset and that the original was the ordinary preliminaries present in all known copies (316-322), not the other way round as D’Alessio would have it. Moreover, every resetting shows without a shadow of a doubt that the purpose of the resetting was to make room for the long dedication while at the same time keeping as much as possible of the original preliminary material: none of the resettings was contextual. To manage this Kallierges had to make more than thirty textual alterations, both large and small, all of them in the form of amputation. Yet, D’Alessio sees “no obstacle in the ‘analytical evidence generated by the physical examination’.” Personally, I could think of more than thirty obstacles. As usual, D’Alessio gives no alternative reason for these alterations. (The only piece of ‘evidence’ adduced by him is the usual one: “It seems more natural to me…”).

In the text for note 1, D’Alessio appears to have spotted a mistake on my part and takes the opportunity to put things right: the 1515 Pindar is not the first Greek book printed in Rome as I have claimed, but “one of the first Greek books printed in Rome.” Instead, with a reference to A. Rollo, D’Alessio informs us that the first Greek book printed in Rome was Operetta bellissima da imparare la lingua greca composta per Paulo Enea, printed in 1510. (This is a short book of 22 pages only.)

However, this is not a Greek book. A Greek book should be completely in Greek, or at least the major part of it: the 1515 Pindar is completely in Greek except for the privilege on the title page, which is in Latin. The Operetta bellissima, which is not even a Greek book — it is an Italian book! — has a fair amount of Greek in it, but the major part of the text is not in Greek but in Latin and Italian. Moreover, a Greek book would have the title in Greek (the 1515 Pindar does); the Operetta bellissima does not, because it is not a Greek book. In short: the Kallierges Pindar is still the first Greek book printed in Rome — pace D’Alessio and Rollo.

That D’Alessio is not comfortable with arguments of a bibliographical nature is evidenced also by the confusion that reigns in his bibliographical descriptions. Speaking of the first quire of copy no. 134 which is a ternion, he describes it as a quire consisting of three double leaves (i.e., three pairs of conjugate leaves) “corresponding to six sheets and twelve pages” (2, last paragraph). This does not make sense: twelve pages is correct, but six sheets in quarto correspond to 48 pages. He goes on to explain that “all copies bar one” begin with a binion (i.e., two pairs of conjugate leaves) “corresponding to four sheets, and eight pages.” However, four sheets equal 32 pages. This confusion continues on the next page. “After a lacuna of … two sheets, two further sheets are preserved…”. This is a misunderstanding: there is no lacuna of two sheets (16 pages), nor are 16 further pages preserved. I suspect that D’Alessio meant to write “leaves” instead of “sheets” in all these instances.

However, these are not D’Alessio’s only confusing statements. On the second page we find the following quotation: “Ultimately Fogelmark himself assesses the evidence and the arguments for the priority of Variant b as in some cases ‘bordering on decisiveness’, but finds ‘none of them strong enough to settle the matter once and for all’ (336).”

A Freudian slip?

The Editors of BMCR refused to publish my response…