Now, it must be constantly remembered that the entire weight of disproof lies with those who dispute the printed date. This is why it is simply amusing to read Blades’s sage words on the subject of this 1472 book with normal printed signatures. He is pledged to renounce the Oxford date, but he finds it awkward that there is an isolated book of 1472 in precisely the same category—with the same want of precedent, the same absence of imitators, the same forlorn appearance. Observe how he deals with it (p. [116] of the book above cited):—“This is a puzzling book, for it is at least two years earlier than any other book so signed. In this city, too. [i. e. Lübeck[[10]]] many works were issued with MS. signatures with a later date than this. It is dangerous to assert that a book is wrongly dated because you cannot make it fit into a bibliographical theory; but I feel inclined, from the general aspect of the book, to date it as 1482, rather than 1472.” And yet a very high authority on typography assures me that the book is undoubtedly of 1472! What then prevents the tentative and isolated experiment of Cologne from having a similar tentative and isolated forerunner, even at Oxford? We may remember too that in the infancy of printing it was common to detect errors as the book went through the press, and often the printer himself corrected an error with his pen, as in the colophon of the Aegidius (see p. [1]). Or a reader would do the same. But it is believed that in no copy of the Jerome is there any attempt to correct or even throw suspicion on the date. There is the date, plain and detailed, and it is allowable to wait for scientific proof before it is abandoned. A priori considerations have force, but they are liable to sudden overthrow.
Clearly the consideration of signatures alone cannot avail to disprove the date of the Jerome. But much more remains.
2. Signs of progress.
It is said that, if we consider the interval between 1468 and 1479, we shall reasonably expect definite signs of progress. On the contrary, the first three Oxford books are printed with the same type, with similar signatures, with the same sized page and the same number of lines in a column. “In fact,” says Blades in the Antiquary, vol. iii, no. 13, Jan. 1881, in an article on The First Printing Press at Oxford, “if a leaf of one was extracted and inserted in another it would, typographically, excite no remark.” Natura nihil facit per saltum, and we are accustomed to apply the idea of evolution and development to every art and trade. It is asserted also that there is no other case of the cessation of a press for over ten years. But cessation of printing for such a time is not unknown. No book was produced at Bamberg between 1462 and 1480, or at Caen between 1480 and 1500, or at Brussels between 1484 and 1500, or at Haarlem for some years after 1486, or at Saragossa after 1475 till 1485? Moreover the only early printing known at Tavistock is two books in 1525 and 1534. The same type and identical woodcuts are found in the two, with an interval of nine years. And where there is cessation, it is obvious that we may be content with fewer signs of advance when work is resumed at the same press with the same type, than if the activity had been continuous, or if the instruments were changed.
But this question of progress is a plain issue. Are there no signs of advance in the two later books compared with the earlier one?
The first book often has an unevenness at the right-hand edge of a column (in 28 pages out of 84). In the other two it is always perfectly even[[11]]. Again, the Jerome starts printing on sign. a 1, whereas the other two start with a blank leaf, the printing beginning on a 2. Again, in the Jerome there is a peculiar misuse of the capitals H and Q (see p. [241]), not found in the following books. And lastly, to omit smaller matters, there is the decided and important fact that whereas in the Jerome each page was printed separately, in the Aegidius and Aretinus two pages were printed at a time.
3. The Type.
Of the palmary arguments against the date, one still remains. The first Oxford type presents a remarkable similarity to that used by Gerard ten Raem de Bercka (see p. [242]), and his only dated book at present known is of 1478. There is certainly a real connexion between the two founts, but we know so extremely little of this printer that it is at present unsafe to base any conclusion on his work. The typographical genealogy of the early printers of the Netherlands and Germany has not yet been fully drawn out, and of the 1478 Modus Confitendi (Hain 11455), which is here in question, only two copies with the date are known, one in the John Rylands (Spencer) library at Manchester and one on the continent. On this point we shall doubtless know more in time, but at present we are bound to suspend our judgment.
4. Mistakes of date common.
There are two subsidiary considerations left. One is that mistakes of date in colophons are not uncommon. An edition of Aeneas Sylvius’s Epistolae (Cologne, printed by Koelhoff) is dated MCCCCLXVIII, which is stated to be an error for 1478, and an Opusculum de componendis versibus by Mataratius, printed at Venice, is also believed to be erroneously dated 1468 for 1478. Caxton’s edition of Gower’s Confessio Amantis is dated 1493 instead of 1483. I have noticed the following additional errors affecting dates before 1501:—720 for 1720, 1061 for 1601, 1099 for 1499, 1334 for 1734, 1400 for 1490 or 1500, 1444 for 1494, 1461 for 1471, 1461 for 1641, 1462 for 1472, 1472 for 1482.