University of Virginia Library

More on the 1532 Edition of Ariosto's Orlando Furioso
by
Conor Fahy

In October 1986, after I had returned the corrected proofs of my note on the definitive edition of Ariosto's Orlando furioso (Ferrara, Francesco Rosso, 1532), published in Studies in Bibliography, vol. 40,[1] I was able to examine two further copies of this edition, in the library of the Musée Condé, Château de Chantilly, Chantilly (Oise), near Paris.[2] The Musée Condé was established by Henri-Eugène-Philippe-Louis d'Orléans, duc d'Aumale (1822-1897), fourth son of king Louis-Philippe of France; the library consists essentially of the duke's private collection, formed in the second half of the nineteenth century, mainly in England, during a period of exile following the Revolution of 1848.

One of the new copies is printed on vellum. In the hand-written catalogue of printed books available for consultation in the library, which is based on notes made by the duke himself, it is stated that this copy was purchased in Berlin for the sum of 5,000 francs; in a note inserted in the volume, probably also compiled by the duke, the name of the bookseller is given as S. Calvary and the date of purchase 1877. On the title-page, and on the verso of the last leaf, there is a half-erased inscription, in which it is nevertheless still possible to read the words: ". . . Congregat. Oratorij Neapolis". This is enough to identify the copy as that which belonged to the Neapolitan intellectual Giuseppe Valletta (1636-1714), and which passed in 1727, with others of his books, to the Biblioteca Oratoriana of that city, where it figured in the old library catalogue, compiled before 1736, after which date it disappeared from view, and has not been identified since. With the four vellum copies included in the list appended to my note of 1987, we are now back at the figure of five surviving copies on vellum, originally claimed by Van Praet, but subsequently disputed by authorities on Ariosto, because of the disappearance of the Valletta copy.[3]

Even more interesting and important is the second Chantilly copy, on paper, the inspection of which has enabled me to tie up several of the loose ends of my previous note. In the printed catalogue of early editions in the Musée Condé, compiled by Léopold Delisle, the copy is stated to be on large paper, and is implicitly identified with the large-paper copy owned in the nineteenth century by the Italian bibliophile Gaetano Melzi.[4] This identification is undoubtedly incorrect. The Melzi large-paper copy is no. 10 in my list; its pedigree is impeccable (the present owner's uncle bought it from

Melzi's descendants), and it has a facsimile titlepage, as described by Melzi. Delisle was probably led astray by the fact, indicated in the duke's hand-written catalogue, that the Chantilly paper copy had formerly belonged to the English collector Frank Hall Standish, who bought some books from Melzi in 1821. These, however, were incunables and Aldine editions, which Melzi was disposing of to make room for the collection of editions of chivalric epics which he was planning to start; in any case, he did not buy the largepaper Furioso of 1532 until 1830 at the earliest.[5]

Delisle was right, however, in describing the Chantilly copy as a largepaper one. The dimensions of the bottom and outer margins are almost identical with those of copy no. 10, and substantially greater than those of the copy in sheets, now in Verona.[6] Further, like copies 9 and 10, it has the cancellans of sheet inner A, and the corrected state of every forme, sharing with copy no. 9 the sole exception of having the incorrect state of the second round of correction in the inner forme of sheet inner G, whereas copy no. 10 has the correct state of both rounds. In the excellent conditions of light which obtained during my visit, I was able to distinguish the watermark of the paper used, a fleur-de-lys, with a countermark of a cross surmounting the letters C D. During a subsequent visit to Italy, I re-examined copies 9 and 10, and several others, with particular reference to the paper, and now present the results of this further work, which corrects and augments some of the findings published last year, with implications for two of the three heads under which I considered the printing of this edition, namely, the question of the cancelled sheet, and the existence of perfect copies.[7]

Surviving paper copies of the 1532 Furioso can be divided into two groups, according to the paper used. The larger group, comprising sixteen copies (copies 1, 2, 3, 5, 6, 7, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22 in my list) can conveniently be characterized as the anchor watermark group. In point of fact, the single most common paper used in this group (30 sheets, or almost 50 per cent, in the Verona copy, which, being still in sheets, permits easy and accurate examination of the paper) has no watermark at all, but almost all the other sheets have some version of the anchor watermark inscribed in a circle, surmounted by a further element, such as a star or a cross.[8] None of these papers has a countermark. The dimensions of the sheets of the Verona copy are approximately 44 X 32 cm, that is to say, foolscap size, what the medieval Italian paper trade called reçute.[9] All sixteen copies contain a variable number of formes in the incorrect state, and all sixteen belong to Debenedetti's Type I, that is, they have the cancellandum of inner A.

The smaller group consists of three copies, copies 9 and 10 in my list and the paper copy at Chantilly. All three have the same paper, with the fleur-de-lys watermark and the cross and letter countermark, and this paper is used for every sheet. The Bologna copy (no. 9) has presumably been cropped; even so, the dimensions of its outer and bottom margins are slightly larger than the corresponding dimensions of the Verona copy. The outer and bottom margins of copy no. 10 and the Chantilly copy are consistently

larger by several millimetres than the Verona copy measurements. Clearly, the fleur-de-lys paper is larger than the anchor papers, though it scarcely seems big enough to belong to the next traditional Italian size, that of mezane, approximately 51.5 by 34.5 cm.[10] All three copies have no formes in the incorrect state, except that, as has already been said, copy no. 9 and the Chantilly copy do not have the second round of correction, comprising one authorial variant, in the inner forme of sheet inner G. Finally, all three copies belong to Debenedetti's Type II, that is, they have the cancellans of sheet inner A.

In addition to these two groups of paper copies, there is also a group of vellum copies, nos. 8, 11, 12 and 23 on my list, and the vellum copy at Chantilly. As I reported in my previous note (p. 76, n. 13), these copies were at first sometimes printed with the uncorrected state of the forme, but after gathering G they always have the correct state. The Chantilly copy, like copies 8, 12 and 23, has the cancellandum of sheet inner A; copy no. 11, the presentation copy for cardinal Ippolito II d'Este, son of the duke of Ferrara, thus remains the only surviving vellum copy to have the cancellans of this sheet.

Using the evidence provided by this grouping of surviving copies according to the material on which they were printed, it is possible to draw some further conclusions about the printing of the edition. These conclusions, of course, like those advanced in my previous note, are subject to the limitations of the inductive reasoning on which they are based, limitations about which, after D.F. McKenzie's famous article in this journal in 1969, no bibliographer can be in any doubt.[11] McKenzie's salutary illustration of the dangers of over-confidence does not, however, lessen the fact that hypothetical conclusions based on inductive reasoning, and consisting of a judicious weighing of the available evidence, are the stuff of which historical reconstructions are made, and there can be no forward movement in historical knowledge without them. The following observations are advanced in this spirit, accompanied by the unwritten proviso: "unless further evidence emerges to the contrary".

It seems that from the outset the printer of this edition, doubtless with the knowledge, indeed, probably on the instructions of, the author, distinguished in the printing of each forme between two sorts of paper copies, those, the majority (probably the vast majority), on unwatermarked or anchor-watermarked paper, and those, the minority (probably a small minority), on larger paper with the fleur-de-lys watermark, making sure that the latter never went through the press until after the correction of the forme, which, in this edition, as I argued in my note of last year, was a normal procedure. Only one hiccup seems to have occurred in this arrangement, in the inner forme of inner G, where the large-paper copies seem to have been actually on the press when printing was stopped to incorporate a second round of correction, in the form of a significant re-writing of a whole line (Canto XII, st. 85, l. 8, G6r a24 in the edition, from Sempre è in timore, e far contraria

via to Teme, e di far sempre contraria via). The disposition of copies in this variant—two of the large-paper copies, all the vellum copies and only two of the anchor paper copies with the incorrect state, and all the other copies, including fourteen anchor paper copies, with the correct state—shows that the large-paper copies did not necessarily go through the press at the very end of the press run of each forme. None the less, it is interesting that, from the start, more care was taken over the text which went into the large-paper copies than over that contained in the vellum copies.

This care over the text of the large-paper copies is further emphasized by considering the other main difference between copies of this edition, the presence, in the large-paper copies, and in one of the five surviving vellum copies, of the cancellans of inner A. This contains, as I argued in my note of last year, the definitive text of the 78 stanzas in question. I commented: "One has to admit, however, that if Type II [the cancellans] of sheet inner A was a cancel, intended to correct an oversight which had left some early pages of the text linguistically and stylistically disfigured, it was a very unsuccessful one. Of the twenty-four surviving copies listed in Appendix A only three contain the cancellans. I have no satisfactory explanation of why this should be so" (p. 82). The discovery of the Chantilly copy on paper now suggests the missing explanation. Unless and until a copy appears with the cancellans on anchor paper, we must assume that the cancel was never intended for such copies. While it seems to have been conceived of as applying to such of the vellum copies as had not already left the printing house, or the city, at the time of production (which must have been after all the rest of the edition had been printed), as far as paper copies were concerned it was a cancel for the large-paper group only. The vellum copies were clearly intended for Ariosto's masters and patrons; while considerable care was taken over the text, what mattered most here was the quality of the material. The anchor-paper copies, with their aleatory combination of corrected and uncorrected states, and the cancellandum of sheet inner A, were equally clearly meant for sale to the public. For whom were the large-paper copies designed? Adopting and expanding a suggestion of Debenedetti, I take this group of copies to have been planned ad usum auctoris et amicorum suorum, a readership expected to be fully capable of appreciating the nuances of style and language involved in the final corrections made to the text of Ariosto's work while it was printing. The qualitative difference in the paper used for the two groups of paper copies is not such as to exclude the possibility that the motive for using different paper for the smaller group was essentially practical, to aid the pressmen to recognize, by the size of the paper, the group of copies which they were on no account to put through the press, in the printing of each forme, until after there had been an opportunity for the correction of the type. It is tempting, or would have been to a previous generation of Renaissance scholars, to interpret the relegation of imperfect copies of the 1532 Furioso to the group intended for sale to the general public of readers as a sign of the aristocratic disdain of the court poet Ariosto, dedicated to "art for

art's sake", for the approval of the common reader, whom he did not expect fully to understand his airy fantasies, and so was not concerned if they were presented to them in an imperfect text. I wonder, however, whether it is not more consistent with the disenchanted realism which we have now come to accept as a profounder characteristic of the "King of Court Poets" to suggest that the division of paper copies of the 1532 Furioso, as set out above, is simply a recognition by Ariosto both of the human error to be expected in the complex operation of bringing his long poem to print, and of the irrepressible nature of his desire to emend his text up to and beyond the last possible moment.[12]

It is interesting to note that a somewhat similar division of copies occurred in the first printing, in 1528, of another Italian classic, Il libro del Cortegiano, by Baldassar Castiglione, though whether accompanied by significant textual differences nobody knows, since the edition in question, printed in Venice by the Aldine press, has not been subjected to bibliographical analysis. Castiglione, a Mantuan nobleman, was appointed Papal Nuncio to the emperor Charles V in 1524, and spent the rest of his life in Spain, dying there in 1529, so that he was not in Italy when his edition was published. In two letters to his steward he described the sort of edition he wanted, and gave instructions for the disposal of the copies. There were to be 1000 "ordinary" copies, plus thirty on "carta reale". He was to pay for 500 of the ordinary copies, plus all thirty of the second group; one hundred of the 500 ordinary copies belonging to him were to be retained for distribution to his friends, and the others sold in Italy through booksellers to recoup his expenses. The thirty copies on "carta reale" were presentation copies for Castiglione's patrons and his more distinguished friends, of whom he provided a list in his second letter (the king of France, the Pope, the duke and duchess of Mantua, etc.). In this letter he also enquired tentatively about a copy on vellum which his steward had asked the printer to prepare for him, if possible; modern scholars have suggested that this copy was intended for Charles V. While the vellum copy, if it ever existed, has not reappeared in modern times, copies of the 1528 Cortegiano on different sorts of paper certainly survive: in the British Library, the King's Library copy (pressmark: 31.g.9) is on larger and heavier paper than the Grenville copy (pressmark: G.2458), and is about twice as thick, though it has the same collational formula; it is presumably one of the thirty copies on "carta reale".[13]

Finally, I would like to correct an error in my note of last year, which is relevant to my contention that Ariosto was presented with proofs of each forme or sheet of the 1532 Furioso. In my note (p. 80, n. 19) I stated that some autograph fragments of the material added to the 1532 Furioso had survived, but not the copy sent to the printers. In fact, among these fragments, as Debenedetti realised, there are indeed some leaves which served as printer's copy, comprising almost the whole of one of the additional episodes, the story of Olympia. The signs of the casting-off of copy are clearly visible, including erroneous calculations later corrected.[14] There are numerous differences

between the text of these autograph pages and that printed in copies of the 1532 edition. Some of these, concerning abbreviations, capitalisation and punctuation, could be due to the compositor; but others involve linguistic and stylistic choices, similar to those displayed in many of the press-corrections in the volume, and in the cancellans of sheet inner A. The crucial cases are those relating to parts of the text carried on formes of the 1532 edition which are already known to have been press-corrected. As an example, I take the inner forme of sheet inner E. There are two press-variants in this forme, the uncorrected state, which I give first, being found in four anchor-type paper copies, against all the rest: E5v a34 Et fatto / E fatto; and E5v b13 vn tempo data / vn tempo io data. The autograph manuscript has the reading of the uncorrected state in both these cases; elsewhere in the parts of the text corresponding to the pages of this forme it has some twenty linguistic and stylistic variants with respect to all surviving copies of the 1532 edition (e.g. Canto IX, st. 50, l. 1; E5v a17 in the edition: (autograph) Il padre: li fratelli / (edition) Mio padre, e miei fratelli).[15] In these cases the change is clearly authorial, and since it is not the result of press-correction, it can only have been instigated during a previous process of proof-correction, which, if it was thought to be necessary for new material neatly set out in a clearly written manuscript, would a fortiori also be likely to be thought necessary for those parts of the text——the vast majority—where the exemplar was a copy of the 1521 edition heavily corrected.

Notes