Chapter VII: Caxton a Printer

The evidence as to where and from whom Caxton acquired his knowledge of the Art of Printing has been considered by nearly every bibliographer as being confined entirely to the information obtained from Caxton’s own Prologues and Epilogues, with the one addition of the well-known quatrain of Wynken de Worde, at the end of his “Bartholomæus de Proprietatibus Rerum.” The argument from technical peculiarities in the books themselves has hitherto been almost entirely overlooked, although a mass of the truest, because unintentional evidence may be found from the attentive study of these dumb witnesses.

Mr. Bradshaw, of Cambridge, has most truly observed, in his “Classified Index,” that the bibliographer should “make such an accurate and methodical study of the _types_ used and _habits of printing_ observable at different presses as to enable him to observe and be guided by these characteristics in settling the date of a book which bears no date upon the surface.”[5] But the great difficulty in the way of this systematic study is the impossibility of having the books side by side, for their rarity is so great that in no one existing library can they all be found.

The books printed in Caxton’s type No. 1, used only at Bruges, are five in number, although we can trace his direct connection with but two of them.

1. “The Recuyell of the Historyes of Troye,” with Prologues and
Epilogues.
2. “Le Recueil des Histoires de Troye.”
3. “The Game and Playe of the Chesse,” with Prologue by Caxton.
4. “Les Fais et Proesses du Chevalier Jason.”
5. “Meditacions sur les Sept Pseaulmes penitenciaulx.”

To these must be added one book printed at Bruges in type No. 2.

6. “Les Quatre Derrennieres Choses.”

Before analysing the evidence supplied by Caxton’s remarks and dates, it is necessary to explain how easily a mistake may be made, and an erroneous conclusion drawn, unless care be taken to remember the effect of the change of style upon the commencement of the year. In England, from the thirteenth century until 1752, the new year began on March 25th; while in Holland and Flanders it commenced on Easter Day. Neglect of this fact has led to many historical errors. Thus, one historian states that Charles I was beheaded on January 30th, 1648, whereas others assert that the event took place on the same day in 1649; one dates the flight of James II from his kingdom in February, 1688, whilst others date it in 1689. In these and many other instances one writer takes the old style of beginning the year, whilst others take the new style, each being right from his own standpoint. In a lately discovered tract printed by Caxton, and known as the “Sex Epistolae,” we have the text of several letters which passed between the Pope and the Doge of Venice, which will be more particularly described under “Books in type No. 4.” It is merely mentioned here as affording an apt illustration of the foregoing remarks. The letters commence on December 11th, 1482, and succeed one another in due order until the 7th of January, 1482, and the end of February, 1482. This was no blunder, for the old year continued until March 25th, which was New-Year’s Day, 1483. Returning now to the consideration of Caxton’s first lessons in the Art of Printing, we will examine each of the books attributed to him, commencing with

“THE RECUYELL.”--This occupies the foremost place, because Caxton himself tells us that with it he began his career as a printer. Its Prologues and Epilogues contain curious and interesting gossip from Caxton’s own pen, telling us how the Duchess of Burgundy, in whose service he then was, commanded him to complete the translation, which he had begun but not advanced with. He tells us that he began to translate the work at Bruges on March 1st, 1468, which, as the year in Flanders did not then commence till Easter, was really 1469, that he continued it at Ghent, and finished at Cologne on September 19th, 1471, thus making a period of two years and a half; that on its completion he presented it to the duchess, who rewarded him handsomely; that many persons desired copies of it, so that, finding the labour of writing too wearisome for him, and not expeditious enough for his friends, he had practised and learnt, at his great charge and expense, to ordain the book in print, to the end that every man might have them at once. As was natural to a person making practical acquaintance for the first time with the effects of typography, Caxton ends with noticing what in his eyes, accustomed to see one copy finished before another was begun, was the most wonderful feature of the new art, namely, that all the copies were begun upon one day, and were finished upon one day.

The periods of time here mentioned by Caxton require notice. He began to translate on March 1st, 1469, but soon relinquished his self-imposed task, after writing no more than five or six quires (or sections of four or five sheets each). After the lapse of two years, in March, 1471, he resumed the translation, and in the following September he presented the duchess with the completed work. Now, six months would have been a very likely time for the translation and a fair copy thereof to take; but it would have been impossible to have accomplished the printing also in that space of time, especially as the whole translation was finished before the first sheet was printed, as will be hereafter shown. We may also notice, that the duration of Caxton’s visit to Cologne must have been very short, as his absence from Bruges lasted no more than six months.

“LE RECUEIL” has but one date, and that evidently refers to the literary compilation alone, and affords no clue whatever to the year of printing. Indeed, the numerous copies still extant in manuscript prove that the work enjoyed considerable popularity before it came under the hands of the printer. The date of the printing of this book has been fixed, by several writers, between 1464 and 1467, from the consideration that Le Fèvre, the compiler, is spoken of in the prologue as chaplain to the Duke of Burgundy, and in such a manner as to signify that the duke was then living. But in the English version there is a material difference: Le Fèvre is not styled there as in the French, “Chappellain de montres redoubte seigneur Monseigneur le Duc Phillipe de Bourgoingne,” but “chapelayn vnto the ryght noble glorious and mighty prynce, _in his tyme_, Phelip duc of Bourgoyne.” Philip, therefore, was alive when “Le Recueil” was printed, but dead when “The Recuyell” went to press. The duke died in 1467; and it is therefore inferred that “Le Recueil” must date between 1464 and 1467, while “The Recuyell” must be later than 1467. That this should be considered as proving anything more than that the original French was compiled during the lifetime of Philip, and that when Caxton translated the same the duke was dead, seems unaccountable. All the copies of “Le Recueil,” both manuscript and printed, followed the wording of the original, and the printer would no more think of altering it in 1476, the probable date of imprint, than the transcriber would in copying the same twenty-five years later. The National Library at Paris has a manuscript of this very book written after 1500, but reproducing exactly the clause which, in the printed edition, is considered to be a proof of its having been executed prior to 1467. Caxton altered the prologue of Le Fèvre to suit his own time, because he was translating; but, in printing from the manuscript of another (assuming his connection with “Le Recueil”), he would have been in opposition to the practice of his age had he altered the original. His translation was in its turn printed and reprinted, word for word, long after it was out of date.

There is, therefore, no reason whatever for asserting that “Le Recueil,” written in 1464, was printed before “The Recuyell,” translated in 1474, and sent to press about the same date. In fact, the whole tone of the epilogue to Book III of “The Recuyell,” leads unquestionably to the conclusion that _that_ was the very first occasion on which Caxton had busied himself with typography. He would never have said, “I have learned to ordain _this book_ in printe at my great charge and expense,” if he had already printed one or two others. M. Bernard assumes that Caxton had nothing to do with the printing of “Le Recueil,” and that it was executed before he turned his attention to the new art. This opinion, however, has not a single fact to support it.

“THE CHESS BOOK” affords but little evidence of value, its prologue being, for the most part, merely a translation of that written by Jehan de Vignay for the French original. It offers, indeed, one date; but that is open to question in its application. “Fynysshid the last day of marche, 1474,” are the concluding words of the epilogue. But what was finished, the translation, or the printing? From the context it was probably the translation, although the printing was not many months later. This date also must be advanced a year; for, as already noticed, the new year did not commence, in Flanders, till Easter Day, which fell, in that year, on April 10th; so that March 31st, 1474, was, according to the modern reckoning, March 31st, 1475.

The prologue to the second edition throws a little light on the history of the first. Caxton there says, in reference to his connection with the book: “... an excellent doctor of divinity ... made a book of the Chess moralised, which, _at such time as I was resident in Bruges_, came into my hands.... And to the end that some which have not seen it, nor understand french nor latin, I deliberated in myself to translate into our maternal tongue; and when I had so achieved the said translation, _I did do set in imprinte_, a certain number of them which anon were depesshed and sold.” He here appears to mean that upon the completion of the translation he employed some one else to print it:--“I did do set in imprinte.” “Did do,” according to the idiom of those days, was commonly used for doing a thing through the medium of another. The phrase was borrowed from the French--“plain pouoir de prendre et faire prendre les larrons,” is the wording of an ordinance dated in the fifteenth century. “He did do be said to the messenger,” for “he caused to be said,” is found on folio 22 of the “History of Jason.” “The Emperor did do make a gate of marble” occurs in the second edition of the “Chess Book,” fol. 85. Similar examples abound, so that we may fairly conclude that Caxton did not himself print the first edition of the “Chess Book,” but that both the translation and the printing were executed in Bruges.

_Type No. 1._

_From “The Chess-book,” 1st Edition._]

The other books, namely, the French “Jason,” the “Meditacions,” and the “Quatre Derrennieres Choses,” contain the bare text without remark or date of any kind, being, as bibliographers say, _sine ullâ notâ_.

The whole of the literary evidence therefore may be briefly summed up thus: “The Recuyell” was translated in 1471, and printed some time after; the “Chess Book” was printed after 1474, and probably in the latter half of 1475; and “Le Recueil” was compiled in 1464, but, like the other four, affords no evidence of date of the printing, which was probably about 1476.

We will now examine the testimony afforded by a comparison of the technical peculiarities of these six books. In collating “The Recuyell,” the make-up of the sections, at the beginning of the volume, is worth noting. It was the practice of Caxton, as of other printers, to commence the printing of his books with the text, any preface which might be requisite, being added afterwards in a separate section, with a different kind of signature. When, however, the whole of the manuscript, prologue as well as text, was complete before it came into the printer’s hands, there was no occasion for any such arrangement. This appears to have been the case with regard to “The Recuyell,” where nothing has been added at the beginning, as the first section of five sheets includes all the introductory matter, as well as a portion of the text. Now the first page, which bears the date of the conclusion of the translation, being on the same sheet as a portion of the text, it is evident that the whole volume must have been in manuscript before any part was set up in type. We may infer, indeed, from his own description of the effect that so much writing had upon him, that Caxton issued several manuscript copies before he thought of using the printing-press. The copy presented to the duchess was undoubtedly manuscript; or else how could Caxton have chronicled in the printed work her acceptance of the book and his reward for the present? And this again leads to the supposition that the portion of the epilogue relating to the printing was added by Caxton to his original manuscript when he determined to print it.

For precisely similar reasons, Caxton’s prologue to the “Chess Book,” which was a translation or adaptation of the original French, is also a portion of the first section of the volume. None of the other books under review having prologues, we will proceed to a comparison of some other typographical particulars.

The following table will show some of the technical features of each book, and some of what may be called the “habits” of the printer:--

┌─────────────────────────────┬──────────────────────────────────────────┐ │ │ Type No. │ │ │ │ No. of Sheets in a Section. │ │ │ │ │ No. of Lines in a Page. │ │ │ │ │ │ Measurement of a Page.-- │ │ │ │ │ │ Inches. │ │ │ │ │ │ │ Spacing of │ │ │ │ │ │ │ Lines │ │ No. Title. Size. │ │ │ │ │ │ Signature │ ├─────────────────────────────┼───┼───┼────┼────────┼────────┼───────────┤ │ 1 The Recuyell Fol. │ 1 │ 5 │ 31 │ 5 × 7⅜ │ uneven │ none │ │ 2 Le Recueil Fol. │ 1 │ 5 │ 31 │ 5 × 7⅜ │ uneven │ none │ │ 3 The Chess Book Fol. │ 1 │ 4 │ 31 │ 5 × 7⅜ │ uneven │ none │ │ 4 Les Fais du Jason Fol. │ 1 │ 4 │ 31 │ 5 × 7⅜ │ even │ none │ │ 5 Meditacions Fol. │ 1 │ 4 │ 31 │ 5 × 7⅜ │ even │ none │ │ 6 Les 4^{tre} │ │ │ │ │ │ │ │ derrennieres choses Fol. │ 2 │ 4 │ 28 │ 5 × 7⅜ │ uneven │ none │ └─────────────────────────────┴───┴───┴────┴────────┴────────┴───────────┘

From this table we perceive,--

First, That the first five books are printed with the same types, are all of the same size, and all without signatures; that all agree exactly in the size of the page; and that the even spacing of the lines in the “Meditacions” and the “Jason” proves that they were produced later than the others.

Secondly, That the five books in type No. 1 may be considered as the production of one printer.

Who, then, was this printer? When we attentively examine the shape of the letters in type No. 1, we notice a remarkable similarity between it and that of the writing of many Bruges manuscripts of the same period, which would induce us, at first sight, to attribute the design of the type to some artist of that city.

M. Bernard, whose opinion is of great weight, where his nationality is not concerned, traces the pattern of type No. 1 directly to Colard Mansion of Bruges. Speaking of a manuscript in the National Library at Paris, written by Colard Mansion’s own hand, he says, “This book is written in old bâtarde, and in exactly the same character as the types of ‘Le Recueil des histoires de Troyes;’” yet he attributes the cutting of the types to a French artist, and the printing to a German, Ulric Zel. The paper he also claims for a French mill, on account of the _fleurs de lis_, and the Gothic ~p~ with the quatrefoil, ignoring the fact that these are common Flemish watermarks of the fifteenth century, and found in abundance in the books from the Bruges and Westminster presses.

That any of these books in type No. 1 were printed by Ulric Zel, or any other Cologne printer, I cannot for a moment believe. It is possible, of course, that Zel, if employed to do so, could have designed and cut types of the gros-bâtarde pattern, although, as a fact, he never used such types himself; but all the Cologne printers of that period had their own peculiarities and habits, which were not at all those of the Bruges printers. Zel, from an early period, printed two pages at a time, as may be easily verified where a crooked page occurs; for the other page printed on the same side of the sheet will in every case be found crooked also. Now, “The Recuyell” was certainly printed page by page, as were likewise all the books from Mansion’s press. And Caxton, when printing his quarto books, cut the paper up and still printed but one page at a time. This accounts for the entire rejection by Mansion,[6] and the sparing use by Caxton of the quarto size for their productions, as it necessitated twice as much press-work as the larger size. But stronger evidence is to be found in the fact that Zel, after 1467, always spaced out the lines of his books to an even length, and would have taught any one learning the art from him to do the same; yet this improvement was not adopted by either Mansion or Caxton until several years later. Whoever may have been the instructor of Mansion and Caxton, and whatever may have been the origin of their typography, the opinion that either of them, after learning the art in an advanced school such as that of Cologne, would have adopted in their first productions, without any necessity for so doing, primitive customs which they had never been taught, and returned in after years by slow degrees to the rules of their original tuition, has only to be plainly stated to render it untenable.

The printer of all these works was undoubtedly Colard Mansion, who had just before established his press at Bruges--who cast the types on his own model for Caxton, and instructed him in the art while printing _with_ and _for_ him “The Recuyell” and the “Chess Book”--who _certainly_ printed “Les Quatre Derrennieres Choses”--who supplied Caxton with the material for the establishment of a press in England--who, about the time of Caxton’s departure, used the same type for “Le Recueil”--and who, at a still later period, printed alone the “Jason” and the “Meditacions.”

We will now examine “Les Quatre Derrennieres Choses,” of which the only copy known is in the Old Royal collection in the British Museum. Like all Colard Mansion’s books, and unlike any one of Caxton’s, it is in French. It is printed in type No. 2, the type of the “Dictes” of 1477, and all the early books which issued from the Westminster press. Then the peculiar appearance of the red ink at once attracts attention. The two colours have been evidently printed at the same pull of the press, as was Colard Mansion’s practice. Here the same process of wiping the black ink off lines purposely isolated, and then re-inking them with red, has been resorted to; and here, too, as in the acknowledged productions of the Bruges press, the same defects have been produced; the red ink having a tarnished appearance from the subjacent remains of the black, and the black lines nearest the red having received a red edging, which, however interesting as a connecting link between two celebrated printers, by no means increases their typographical beauty. Now, as no Cologne printer is known to have resorted to this unique method of working in colours, I feel no hesitation in ascribing “Les Quatre Derrennieres Choses” either to Colard Mansion or to Caxton working under his tuition; and as this peculiarity is nowhere found in Caxton’s productions of the Westminster press, the former would seem the more likely conjecture.

The connection thus established between the types used by Caxton in his first attempts in England and those used by Colard Mansion is still further strengthened by the fact that the form of the &c., peculiar to type No. 1, is in several instances, by an evident mixing of the founts, used instead of the proper sort belonging to type No. 2. This furnishes positive proof that the two founts were _under one roof_, whether at Cologne or Bruges, or elsewhere. Whoever printed the five books in type No. 1 most certainly owned type No. 2 also.

Against all this, however, has to be placed the direct assertion of Wynken de Worde, who, in the proheme to his undated edition of “Bartholomæus de Proprietatibus Rerum,” gives the following rhyme:--

“And also of your charyte call to remembraunce
The soule of William Caxton first prynter of this boke
In laten tonge at Coleyn hyself to auaũce
That euery well disposyd man may thereon loke.”

The phraseology of this verse is very ambiguous. Are we to understand that the _editio princeps_ of “Bartholomæus” proceeded from Caxton’s press, or that he only printed the first Cologne edition? that he issued a _translation_ of his own, which is the only way in which the production of the work could advance him in the Latin tongue? or, that he printed in Latin to advance his own interests? The last seems the most probable reading. But though the words will bear many constructions, they are evidently intended to mean that Caxton printed “Bartholomæus” at Cologne. Now this seems to be merely a careless statement of Wynken de Worde; for if Caxton did really print “Bartholomæus” in that city, it must have been with his own types and presses, as the workmanship of his early volumes proves that he had no connection with the Cologne printers, whose practices were entirely different. The time necessary for the production of so extensive a work would have been considerable; therefore, as Caxton’s stay at Cologne on the occasion of his finishing the translation of “Le Recueil” was but short, the printing of this apocryphal “Bartholomæus” would have been at a subsequent visit, of which there is no record. No edition has yet been discovered which can, by any stretch of the imagination, be attributed to Caxton, although there is more than one old undated edition belonging to the German school of printing. Accuracy of information was in those days not much studied, and to a general carelessness about names and dates Wynken de Worde added a negligence peculiarly his own. We may excuse him for using Caxton’s device in several books which by their dates and types are known to have been printed by himself, as well as for putting Caxton’s name as printer to the edition of the “Golden Legend,” printed in 1493, two years after his master’s death. Such inaccuracies were thought but little of at that time. But how can we account for the blundering alteration in the 1495 edition of the “Polycronicon,” where Wynken de Worde, making himself the speaker in Caxton’s prologue, four years after his master’s death, promises to carry the history down to 1485; or for the still greater error in the “Dictes” of 1528, in which, while adopting Caxton’s epilogue, but substituting his own for Caxton’s name, he makes all the transactions there related happen between Earl Rivers (who had been fifty years in his grave) and himself? Wynken de Worde’s blunders in statements are well matched by his blunders in workmanship, of which, however, we will quote but two. In Caxton’s edition of the “Stans Puer ad Mensam,” the third and fourth pages of the poem were accidentally transposed; yet Wynken de Worde, notwithstanding the break of sequence, blindly reprints the error! Again, in his edition of “The Horse, the Shepe, and the Ghoos,” he actually omits a whole page without discovering his mistake! Other examples might easily be quoted, but enough has been adduced to show that Wynken de Worde was by no means careful in his statements.[7]

We must remember that Wynken de Worde, moreover, was too young to have had any personal knowledge of Caxton’s early efforts, and that the vast importance of the art to the entire world, and the interest attaching to its origin, were ideas which would find no place in the mind of a fifteenth-century printer. We must not, therefore, regard De Worde’s statement as deliberately made for the purpose of telling posterity something about Caxton. Lewis, Caxton’s first biographer, was very sceptical concerning this Cologne edition of “Bartholomæus.” “Its having a Latin title,” he says, “might possibly deceive De Worde, and make him think it was printed in Latin. However this may be, it does not appear that any edition of it, printed by Caxton or any one else, either in Latin or English, that year, is now in being.”

Perhaps De Worde, who reprinted “The Recuyell,” had some vague recollection of Caxton having stated that he had been at Cologne, and carelessly adopted the idea as giving point and rhyme to his verses.

The following anecdotes illustrate in a curious manner the typographical connection between Mansion and Caxton. A bookseller of Paris purchased an old volume for the moderate sum of one louis. Ignorant of its great value, he took it to M. de La Serna Santander, and asked him if he thought two louis too dear. “No,” replied the wary bibliographer, and gave him the money. That volume is now in the National Library at Paris, and contains, bound together in the _original_ boards, the “Quadrilogue,” printed by Mansion at Bruges, and the French “Jason,” printed in Caxton’s type No. 1. Something similar to this happened in 1853, when Mr. Winter Jones discovered in the Library of the British Museum, “Les Quatre Derrennieres Choses,” in Caxton’s type No. 2, bound up with the “Meditacions,” in type No. 1, and with contemporary handwriting running from the last page of one work to the first of the other, the volume being evidently in its original state, just as it was printed and bound at Bruges, in the little workshop of Colard Mansion over the church porch of St. Donatus.

Here, perhaps, I may be excused if I venture to build a brief history, founded, in the absence of sure foundation, in many parts on probability only, but which may nevertheless be welcome to some as an attempt to draw into a consistent narrative the scattered threads of Caxton’s career between 1471 and his establishment at Westminster.

Caxton, having finished and been rewarded for his trouble in translating “Le Recueil” for the Duchess of Burgundy, found his book in great request. The English nobles at Bruges wished to have copies of this the most favourite romance of the age, and Caxton found himself unable to supply the demand with sufficient rapidity. This brings us to the year 1472 or 1473. Colard Mansion, a skilful caligrapher, must have been known to Caxton, and may even have been employed by him to execute commissions. Mansion, who had obtained some knowledge of the art of printing, although certainly not from Cologne, had just begun his typographical labours at Bruges, and was ready to produce copies by means of the press, if supported by the necessary patronage and funds. Caxton found the money, and Mansion the requisite knowledge, by the aid of which appeared “The Recuyell,” the first book printed in the new type, and moreover the first book printed in the English language. This, probably, was not accomplished till 1474, and was succeeded, on Caxton’s part, in another year, by an issue of the “Chess Book,” which, as we are informed in a second edition, was “anone depesshed and solde.” Mansion, finding success attended the new adventure, printed the French “Recueil,” and, after Caxton’s return to England, the French “Jason” and the “Meditacions.” The three French works were doubtless published by Mansion alone, as Caxton is not known to have printed a single book in French, although perfectly acquainted with that language. Caxton, having thus printed at Bruges “The Recuyell” and the “Chess Book” with types either wholly or in part belonging to Mansion, now obtained a new fount of the pattern of the large bâtarde already in use by Mansion, but smaller in size, with the intention of practising the art in England. To test its capabilities, “Les Quatre Derrennieres Choses” was then produced under the immediate supervision of Mansion.

Early in 1476 Caxton appears to have taken leave of the city where he had resided for five and thirty years, and to have returned to his native land laden with a more precious freight than the most opulent merchant-adventurer ever dreamt of, and to endow his country with a blessing greater than any other which had ever been bestowed, save only the introduction of Christianity.

FOOTNOTES:

[5] A classified Index of the fifteenth-century books in the collection of M. J. de Meyer. 8vo. London, 1870.

[6] Van Praet, Brunet, and especially Campbell in his “Annales de la Typographie Néerlandaise,” err in describing “Le Purgatoire des mauvais Maris,” printed by Colard Mansion, as a “petit in-4°.” The copy described is cut a little more than usual, but the watermark which _is in the middle of the page_ proves the size to be folio, whereas had it been quarto the watermark must have been in the back and partly hidden by the binding.

[7] William Caxton, except in the occasional interchange of _i_ and _y_, which were at that period considered as equivalents, never altered the orthography of his name, a fact the more noticeable as the name certainly varied in pronunciation: but Wynken de Worde, although mentioning his master’s name but eight times, contrived to make the four variations of Caxton, Caxston, Caston, and Caxon. With regard to his own name Wynken de Worde appears to have tried how many variations he _could_ invent, of which the following list is not even complete:--

Wynken de Worde.
Wynden de Worde.
Wynkyn de Worde.
Wynkyn Theworde.
Wynkyn the Worde.
Wynkyn de Word.
VVinquin de VVorde.
Wynandus de Worde.
Wynandus de word.
winandus de worde.
Vunandus de worde.
Vuinandi de vuorde.
VVinand i VVordensi.
Winandi de Wordensis.