:: :: University of Virginia Library

Among his publications for the year 1569,[1] Aldus Manutius the Younger issued the first editions of Annibale Caro's Rime and the same author's Italian rendering of the Due orationi of St. Gregory of Nazianzus.[2] Both works were (posthumously) edited by Giovanni Battista Caro, a nephew of the author (who had died on 21 November 1566).[3] The editor also supplied each edition with a dedicatory preface, dated 1 May 1568, that to the Rime being addressed to Alexander Farnese, Duke of Parma, the other to the latter's uncle, the Cardinal Alexander Farnese. Each work received a privilege for twenty years, which was granted by the Venetian Signoria ("in Pregadi") on 19 July 1568. In commenting on the Orations,[4] Renouard has remarked: "Ce volume est presque toujours relié avec le Rime." The "almost always" may be a slight exaggeration, but it is unquestionably true that the two works are often found bound together.

To his description of the Rime,[5] Renouard adds the comment: "Au verso de la page 103 est un errata de treize lignes, qui n'est pas imprimé dans tous les exemplaires. Je l'ai vu aussi imprimé à part, sur un feuillet blanc, ayant au bas: 'Humilissimo et Devotissimo Servitore, Gio. Battista Caro.' C'étoit peut-être l'exemplaire de dédicace." In the case of the Due orationi,[6]

220

Renouard observes: "à la fin un [feuillet] d'errata qui, ainsi que le précédent dans les Rime, n'est pas dans tous les exemplaires, et laisse blanc ce dernier feuillet." This information, though correct in itself, does not supply a full account of what happened in the printing office.

Apparently the sequence of events was this: Aldus Manutius Jr. intended to print the two works by Annibale Caro in such fashion that they could be sold either together or singly. After the machining of the two editions had been completed, the editor (or, possibly, the printer) became aware of certain misprints and decided to ameliorate these slips by providing a list of errata. The printer proceeded to apply this remedy in three different ways. Since the printing of both works had been completed, Aldus was able to insert the errata in those copies still in sheets (unbound) by printing the list on the blank verso of signature O4 of the Rime [7] and on the similar page (T6 recto) of the Due orationi [8] on a second trip through the press.[9] Both these lists have the simple heading "Errori de la stampa" and neither is signed with the name of the editor.[10] For the copies already disposed of or bound up, he printed a single leaf of errata for each work (or perhaps a broadside sheet which he then cut in half), which could be inserted into the separately-bound works.[11] These leaves have the headings "NELLE RIME. || Errori de la stampa." and "NELLE ORATIONI. || Errori de la stampa." respectively. Lastly, for those copies which he still planned to sell as a two-volume unit, Manutius issued a single sheet, on the outer forme of which he printed both sets of errata, with the fuller headings as given

221

above.[12] All three sets of errata, in both instances, are printed from different settings of type. Many copies lack the errata lists in any form.[13]

The problems that now arise are: What is the form of the "ideal copy," according to the principles as set forth by Professor Fredson Bowers,[14] and how will the collational formulae run for those copies where a leaf, or a sheet, has been inserted? But before we turn to these matters of central importance, it may be advisable to establish that all three printings, in both cases, undoubtedly represent different settings of type.

It may be remarked, first of all, that this fact is perfectly self-evident to anyone who has the three sets of errata before him, being quite apparent from the differing alignment and spacings in the text. However, it is very difficult to set forth such details in print. It is much simpler to note evidences for the different settings in the case of the errata for the Orations than it is for the Rhymes, due to the very short list found in the latter work. Taking the Orations of St. Gregory first, one may note that the first line of errata reads "a faccie 1 a uersi 6" in the sheet and in the text printed on T6. This is correct, but in the single leaf, the text incorrectly reads "a faccie 3 a uersi 6." The last words of line 6 are "come se pensasse" in both inserted versions, but "come se pensassero" in the errata printed on T6. In the last line, the leaf and the printed-in text have "figlie," whereas the sheet provides "filie." This data should sufficiently identify the three settings of errata in the Due orationi.

As for the Rime, we have already seen that the text in both the sheet and the leaf of errata is headed "NELLE RIME";[15] this, together with the signature of the nephew Caro, is unnecessary for (and is consequently omitted from) the errata when printed on O4v. After the signature of Caro, the leaf has a semi-colon, where the sheet provides a full-stop. The other differences are purely visual, but the points here given will serve to distinguish the three settings.

It is patently evident, of course, that the inclusion of the errata was decided upon after all the sheets of the Rime and of the Due orationi had been printed off. Therefore, those copies with the errata printed in signatures O and T represent the latest state of these editions — and those without this feature, an earlier one. On the other hand, the inserted leaves could be bound in at various places, though they are normally found after

222

P4 of the Rime [16] and after * 4 of the Due orationi.[17] This possibility seems to make an invariable collational formula a matter of speculation rather than of certainty.[18] Finally, the sheet which supplies the errata for both volumes presents (so far as the writer can see) a major problem for the descriptive bibliographer. In my copy, the sheet is bound between the two works, the errata for the Rime facing the last printed page of that work (P4v), while the corrections for the Due orationi face the title-page of this book. Should the sheet be listed at the end of the collational formula for the first work — or at the beginning of that of the second? Or does it properly belong to neither — being, actually, a separate (broadside) printing in its own right?

Manifestly, since the two leaves are indubitably conjugates, they cannot be (mentally) cut asunder; it is, thus, impossible to assign one leaf to each book. And it would seem to make no sense at all arbitrarily to append the errata leaves for both volumes to the collational formula of either one of them. Lastly, if the sheet is considered to be a separate publication, then the "ideal copy" ought not to include this "extraneous" publication. The ideal copy would then be one without the errata, which would leave us with a prime absurdity, since the errata was created in order to perfect the text. Having unsuccessfully pondered this "pons asinorum" for more than four years, I will be happy to let some one else supply the solution for this quandary.[19]

Notes

[1]

I should like to take this opportunity to thank Professor Fredson Bowers for his kindness in reading an earlier draft of this paper and for giving me the benefit of his invaluable criticism and suggestions.

[2]

The imprint in both works reads: "IN VENETIA. || Appresso ALDO MANVTIO. || M D LXIX." and the books are assigned to Aldus Jr. by the Short-Title Catalogue of Books Printed in Italy and of Italian Books Printed in other Countries from 1465 to 1600 now in the British Museum (1958), p. 882. In Antoine Augustin Renouard, Annales de l'imprimerie des Alde (Paris, 1834), p. 206, they are listed under Paul Manuce.

[3]

For a short biographical sketch, see the Enciclopedia Italiana (Rome, 1929-39), IX, 109-110.

[4]

Renouard, p. 206.

[5]

In the University of Pennsylvania copy *46-1585, page 75 is signed N3 instead of L2, while other copies (e. g., Yale, Paris) are signed L3. Apparently the printer wished to correct the wrong signature mark L3 but mistakenly altered it to N3 rather than to the correct L2.

[6]

The University of Pennsylvania copy 858 C22 R has page 59 misnumbered 58, and there is an indentation at line 16. Other copies (thus the two in the British Museum and the four in the Bibliothèque Nationale) have the correct pagination and no indentation, apparently the result of stop-press emendation. The misprint on this page noted in the errata, however, is not corrected in either state.

[7]

For example, in the copies: Bodleian Library, Toynbee 702; John Carter Brown Library; Princeton University Library.

[8]

Thus in the following copies: Bodleian Library, Auct. R. 1. 5. 18; Trinity College Cambridge, N. 3. 41; Harvard College Library; Bibliothèque Nationale, Rés. C. 1418 (3), Rés. C. 1482, and C. 1483.

[9]

This is made evident by the fact that, in the Rime, the errata printed on signature O4v [p. 104] includes the correction of a misprint on page 101 (O3 recto). Thus, both the mistake and the correction are printed in the outer forme of the same signature. If O3 had not already been printed, the printer could not have been aware of the misprint — or, if he had been aware of the error through page-proof or the like, he could have corrected the slip at the proper place without further comment. In the Orationi, the errata printed on T6 corrects a misprint on p. 137 (T1), the conjugate leaf. In short (and for the same reason as given above), the errata in the inner forme could only have been compiled after the outer forme had been printed, though normally the inner one was printed first. Water-marks appear on the two inner sheets, proving that the innermost pair form the odd half-sheet in this quire.

[10]

Copy *46-1586 (Pennsylvania) presents a problem. A single leaf, with the simple heading "Errori de la stampa," is inserted between T5 and T6. T1 and T6 are certainly conjugates, and T6 has an obvious offset from T5v, proving that the errata leaf is inserted and is not the original T6. This might suggest that there was a fourth set of errata, one printed as an insert but with the shorter heading. To the writer, however, it seems more probable that this insert was the original T6 of some other copy and that it was added to the present copy long after the offsetting had occurred.

[11]

Thus in copies Pennsylvania 858 C22 R (Caro and Gregory) and Bodleian Toynbee 1116 (Rime only).

[12]

The only such copy that I have seen is in my own library.

[13]

The errata is wanting in the following copies of the Rime: Trinity College Cambridge, N. 3. 40; British Museum, G 11117 and 85. c. 2; Pennsylvania *46-1585; Bodley, Auct. 2. R. III. 34; Bibliothèque Nationale, Rés. Yd 667; Harvard College Library (not bound with Gregory); Yale University and University of Illinois Libraries. These copies of the Due orationi lack the errata: British Museum, 1412. h. 17 and 85. c. 2; Yale and Illinois (both bound with Rime); New York Public Library; and Bibliothèque Nationale, C. 1481.

[14]

In his Principles of Bibliographical Description (1949).

[15]

In the Due orationi too, of course, the printed version differs from the inserted ones, which alone have the heading "NELLE ORATIONI."

[16]

It could also, for example, be inserted after O4. The collation for the Rime is: Quarto; *4 B-P4. Quire P contains only the "Tavola" for Caro's verse.

[17]

As we have seen, it is bound between T5 and T6 in one copy. The collation for the Due orationi is: Quarto; *4 B-S4 T6.

[18]

One may also debate as to whether the ideal copy is one with the errata printed in it or one with an inserted leaf (or sheet). If both forms are "ideal copies," then the collational formula would exist in four (possibly five) varying forms (singly, with and without extra leaf; together, with and without inserted sheet or leaves). In any case, the inclusion of the errata was not envisaged when publication was begun — and it is possible (though not very likely) that a considerable length of time elapsed before the errata was produced. However, so large a percentage of copies without errata is noted here that it seems equally unlikely that the addition of the errata was determined upon immediately after original publication of the two books. This possible time interval may raise the problem as to whether we are here dealing with states or issues.

[19]

Professor Bowers has suggested to me that this would present no problem, if one argues that the two works together form a collected edition, and that the sheet exists solely as a supplement to the two-work unit. In that case, the ideal formula would simply be a variant, consisting of the formulae for the two works with the sheet χ2 between them. Nevertheless, one must face the possibility that the survival of the double errata as conjugate leaves may be purely accidental and not by the will of the printer. Under these circumstances, we do not actually know what the printer had in mind — and how he wished the "ideal copy" to appear (if, indeed, he had the slightest interest in the matter). In any case, I cannot find any simple explanation as to why different settings were needed — or used — to supply the errata.