University of Virginia Library

The First Edition of Ficino's De Christiana Religione: A Problem in Bibliographical Description.
by
Curt F. Bühler [1]

The first edition of the De christiana religione by Marsilius Ficinus[2] was produced at Florence, some time within the month following 10 November 1476, by Nicolaus Laurentii,[3] a German printer who had emigrated to Italy from his native diocese of Breslau. The book itself is an unsigned quarto, and the collation given in the British Museum's monumental incunabula catalogue reads: [a b10 c-f8 g6+1 h-p8 q10 *2]. A manuscript entry in the Museum's own "working copy" of this catalogue[4] notes that "[g]1 was occupied by the same text as [g]2 in a different setting-up & was cancelled. The Phillipps copy (Sotheby sale cat. 25 Nov. 1946 no. 122) contained both leaves. Sold to Harvard Univ. Lib." Accordingly, it would seem proper to emend the collation of this copy so that the seventh quire would be described as [g]8-1.

A thorough consideration of the problem presented by this irregular quire poses such questions as to WHY the duplication of text came into being, WHAT would be the true make-up of an ideal copy of this printing, and HOW should the seventh quire be properly listed?

Turning, then, to the redundant text, a close comparison of the two settings shows that for fifty lines the two agree, line for line;[5] the only difference between the two occurs in lines 5-6 of the verso, where [g]1 divides "criſti∥ana" and [g]2 has "chriſtia∥na." Such agreement can hardly

be accidental![6] Of even greater significance, at least in the opinion of the writer, is the fact that both settings offer the identical misprints or errors: in line 4 (recto), both copies have "ſactio" where "ſanctio" is required; in line 12, "elegi" for "elegit"; in line 12 (verso), "denu(m)ptiat" for "denunciat"; and in line 14, both read "permictit" where "permittit" is wanted.[7] Of course, the archetype itself might have been corrupt at these several places—but it seems strange that no emendation was attempted for these rather obvious mistakes by the compositor of either setting. Then, too, neither setting is any real improvement on the other; on the verso, setting [g]1 has the slip "exrahi" for the "extrahi" on [g]2, while [g]2 incorrectly has "ſeruari" where "ſeruare" is preferable.[8] In any event, minor corrections of this sort could have been made in the forme without the necessity of resetting two entire pages of type. Significantly enough, the two rectos display only orthographic variations,[9] so that it cannot very well be argued that the printer reset the type in order to correct his text and then, by some mishap, used both settings in the printing.

Before analyzing our chief problem, it would be best to examine briefly Laurentii's printing practices at this time, which marked the very outset of his career as an independent printer.[10] His two earliest productions appear to be the work under discussion and an Italian version of this same text (HCR 7071; Stillwell Census F135), both being quartos. The evidence of the watermarks suggests that these books were printed by half-sheet imposition, for in some five quires of two copies,[11] watermarks are found in six of the eight leaves. Again, it is clear that Laurentii possessed at least two presses; the BMC (VI:625), for example, remarks that, in the Italian version, "a miscalculation of the length of ch. 28 has resulted in the greater part of 81b and the whole of 82 (end of quire [k]) being left blank. Possibly

the book was set up on three presses, ending with quires [e], [k] and [o] respectively." That quire [l] was probably printed before [k] seems reasonable enough, and two presses may, therefore, be postulated. However, a third press could hardly have started with quire [f], since this begins in the middle of the word "[absti]nentia." Finally, it seems certain that Laurentii was not printing from cast-off copy but that he was setting type seriatim by pages. Of the 27 quires in the two volumes which come into consideration, 10 begin in the middle of a sentence and 16 in the middle of a word.[12] Since he was following this practice (in effect, printing as in folio), it is likely that he was working "from inside out"; that is, in the order: [g]4-[g]5, [g]3-[g]6, [g]2-[g]7, and [g]1-[g]8

Only one explanation for the redundant text—and not a very good one at that—has occurred to the writer. For the printing of the two innermost sheets, there was, of course, no problem, but either with the third or with the outermost one, Laurentii may well have suddenly discovered that, by some miscalculation, he did not have enough pages to fill out the formes. Clearly, there was some excellent reason for his not being able to use two pages from the next quire as type-high supports—and one may consequently assume that quire [h] was then either printing or had already been printed off. Thus, one may again assume that Laurentii was working with at least two presses.

Now if the printer did not realize that he was short of type-pages till he was machining [g]2-[g]7, then it is self-evident that [g]2 must be the earlier setting. For some reason not clear to us, Laurentii apparently had no standing type [or blocks, furniture or quads] to make up the two pages necessary to fill the formes. His compositor was, therefore, obliged to supply him with two duplicate pages, which he set up quickly from the already printed copy. If, however, Laurentii was working with more than one press, he might have been machining the two sheets simultaneously, in which case it would be impossible to determine whether [g]1 or [g]2 was set first.

The purpose of these two reset pages was, of course, to act as "bearers" which were probably never meant to be inked but simply to serve as a support during the printing. Normally, they would have provided a blank leaf, subsequently to be cut out. Here, however, they were inked and printed, and thus escaped excision.[13] Examples of accidental inking and

printing of "bearer" type, and the consequent survival of such, are not unknown in the incunabula period.[14] In any case, one cannot assert with any degree of certainty which pages were set first or which leaf was supposed to have been excised.[15]

For the descriptive bibliographer, of course, the perplexing problem is what will here constitute an "ideal copy" and what index number shall be given to the seventh quire. According to Professor Fredson Bowers[16] (and all will agree with him on this), "an ideal copy is not a redundant copy." With this in mind, one may well argue that [g]8 would not be, in the least, acceptable. Equally, any bibliographer of the Greg-Bowers school would argue that an "index figure should represent the number of conjugate quired leaves in the original gathering, with all abnormalities accounted for by other means."[17] An uneven index-number is repugnant to this school of bibliography,[18] and for very excellent reasons too, since a sheet could not very well be machined with an odd number of leaves. Therefore [g]7, too, is inadmissible. One could, of course, write [g]8 (-g1) if one were positive that the publisher wished the first leaf to be cut out, or [g]8 (-g2) if it were the second which Laurentii might have wished to suppress.

Since the ideal copy is that which represents the "most perfect state of the book as the printer or publisher finally intended to issue it,"[19] one turns to the surviving copies to see if a clue as to what Laurentii's intention was might be discovered there. The make-up of the seventh quire in the sixteen copies known to the writer provides the following information:

What can be deduced from this? Perhaps the not unreasonable assumption that Laurentii may have been quite indifferent to [possibly even ignorant of] the true state of affairs. It is certainly unclear whether such excisions as there are were made in the shop or by the individual purchasers. That one or the other leaf should have been deleted is quite self-evident, but this does not get us very far. One may still wonder (1) whether the publisher [as distinct from the press-man or compositor] was aware of the repetition, (2) whether, being aware of it, he had any interest in what the purchaser did about it, or (3) whether he himself cut out one or the other of the offending leaves in those copies still unsold when he discovered the duplication, without caring very much which of the leaves he removed. In view of such uncertainties as these, how can the ideal copy be identified and what should be the index number of the seventh quire?

Notes