Introduction (separate file)
List of Illustrations (separate file)

Chapter I
Chapter II
Chapter III
Chapter IV (separate file)
Chapter V (separate file)
Chapter VI (separate file)
Chapter VII (separate file)
Chapter VIII (separate file)
Chapter IX (separate file)

Index (separate file)

1

ON
WOOD ENGRAVING.

CHAPTER I.
ANTIQUITY OF ENGRAVING.

Engraving—the word explained—the art defined—distinction between engraving on copper and on wood—early practice of the art of impressing characters by means of stamps instanced in babylonian bricks; fragments of egyptian and etruscan earthenware; roman lamps, tiles, and amphoræ—the cauterium or brand—principle of stencilling known to the romans—royal signatures thus affixed—practice of stamping monograms on documents in the middle ages—notarial stamps— merchants’-marks—coins, seals, and sepulchral brasses—examination of mr. ottley’s opinions concerning the origin of the art of wood engraving in europe, and its early practice by two wonderful children, the cunio.

As few persons know, even amongst those who profess to be admirers of the art of Wood Engraving, by what means its effects, as seen in books and single impressions, are produced, and as a yet smaller number understand in what manner it specifically differs in its procedure from the art of engraving on copper or steel, it appears necessary, before entering into any historic detail of its progress, to premise a few observations explanatory of the word Engraving in its general acceptation, and more particularly descriptive of that branch of the art which several persons call Xylography; but which is as clearly expressed, and much more generally understood, by the term Wood Engraving.

The primary meaning of the verb “to engrave” is defined by Dr. Johnson, “to picture by incisions in any matter;” and he derives it from 2 the French “engraver.” The great lexicographer is not, however, quite correct in his derivation; for the French do not use the verb “engraver” in the sense of “to engrave,” but to signify a ship or a boat being embedded in sand or mud so that she cannot float. The French synonym of the English verb “to engrave,” is “graver;” and its root is to be found in the Greek γράφω (grapho, I cut), which, with its compound ἐπιγράφω, according to Martorelli, as cited by Von Murr,I.1 is always used by Homer to express cutting, incision, or wounding; but never to express writing by the superficial tracing of characters with a reed or pen. From the circumstance of laws, in the early ages of Grecian history, being cut or engraved on wood, the word γράφω came to be used in the sense of, “I sanction, or I pass a law;” and when, in the progress of society and the improvement of art, letters, instead of being cut on wood, were indented by means of a skewer-shaped instrument (stylus) on wax spread on tablets of wood or ivory, or written by means of a pen or reed on papyrus or on parchment, the word γράφω, which in its primitive meaning signified “to cut,” became expressive of writing generally.

From γράφω is derived the Latin scribo,I.2 “I write;” and it is worthy of observation, that “to scrive,”—most probably from scribo,—signifies, in our own language, to cut numerals or other characters on timber with a tool called a scrive: the word thus passing, as it were, through a circle of various meanings and in different languages, and at last returning to its original signification.

Under the general term Sculpture—the root of which is to be found in the Latin verb sculpo, “I cut”—have been classed copper-plate engraving, wood engraving, gem engraving, and carving, as well as the art of the statuary or figure-cutter in marble, to which art the word sculpture is now more strictly applied, each of those arts requiring in its process the act of cutting of one kind or other. In the German language, which seldom borrows its terms of art from other languages, the various modes of cutting in sculpture, in copper-plate engraving, and in engraving on wood, are indicated in the name expressive of the operator or artist. The sculptor is named a Bildhauer, from Bild, a statue, and hauen, to hew, indicating the operation of cutting with a mallet and chisel; the copper-plate engraver is called a Kupfer-stecher, from Kupfer, copper, and stechen, to dig or cut with the point; and the wood engraver is a Holzschneider, from Holz, wood, and schneiden, to cut with the edge.

It is to be observed, that though both the copper-plate engraver and 3 the wood engraver may be said to cut in a certain sense, as well as the sculptor and the carver, they have to execute their work reversed,—that is, contrary to the manner in which impressions from their plates or blocks are seen; and that in copying a painting or a drawing, it requires to be reversely transferred,—a disadvantage under which the sculptor and the carver do not labour, as they copy their models or subjects direct.

Engraving, as the word is at the present time popularly used, and considered in its relation to the pictorial art, may be defined to be—“The art of representing objects on metallic substances, or on wood, expressed by lines and points produced by means of corrosion, incision, or excision, for the purpose of their being impressed on paper by means of ink or other colouring matter.”

The impressions obtained from engraved plates of metal or from blocks of wood are commonly called engravings, and sometimes prints. Formerly the word cutsI.3 was applied indiscriminately to impressions, either from metal or wood; but at present it is more strictly confined to the productions of the wood engraver. Impressions from copper-plates only are properly called plates; though it is not unusual for persons who profess to review productions of art, to speak of a book containing, perhaps, a number of indifferent woodcuts, as “a work embellished with a profusion of the most charming plates on wood;” thus affording to every one who is in the least acquainted with the art at once a specimen of their taste and their knowledge.

Independent of the difference of the material on which copper-plate engraving and wood engraving are executed, the grand distinction between the two arts is, that the engraver on copper corrodes by means of aqua-fortis, or cuts out with the burin or dry-point, the lines, stipplings, and hatchings from which his impression is to be produced; while, on the contrary, the wood engraver effects his purpose by cutting away those parts which are to appear white or colourless, thus leaving the lines which produce the impression prominent.

In printing from a copper or steel plate, which is previously warmed by being placed above a charcoal fire, the ink or colouring matter is rubbed into the lines or incisions by means of a kind of ball formed of woollen cloth; and when the lines are thus sufficiently charged with ink, the surface of the plate is first wiped with a piece of rag, and is then further cleaned and smoothed by the fleshy part of the palm of the hand, slightly touched with whitening, being once or twice passed rather quickly and lightly over it. The plate thus prepared is covered with the paper intended to receive the engraving, and is subjected to the action of 4 the rolling or copper-plate printer’s press; and the impression is obtained by the paper being pressed into the inked incisions.

As the lines of an engraved block of wood are prominent or in relief, while those of a copper-plate are, as has been previously explained, intagliate or hollowed, the mode of taking an impression from the former is precisely the reverse of that which has just been described. The usual mode of taking impressions from an engraved block of wood is by means of the printing-press, either from the block separately, or wedged up in a chase with types. The block is inked by being beat with a roller on the surface, in the same manner as type; and the paper being turned over upon it from the tympan, it is then run in under the platen; which being acted on by the lever, presses the paper on to the raised lines of the block, and thus produces the impression. Impressions from wood are thus obtained by the on-pression of the paper against the raised or prominent lines; while impressions from copper-plates are obtained by the in-pression of the paper into hollowed ones. In consequence of this difference in the process, the inked lines impressed on paper from a copper-plate appear prominent when viewed direct; while the lines communicated from an engraved wood-block are indented in the front of the impression, and appear raised at the back.

  see text and caption

PRINTED FROM A WOOD-BLOCK.

  see text and caption

PRINTED FROM A COPPER-PLATE.

The above impressions—the one from a wood-block, and the other from an etched copper-plate—will perhaps render what has been already said, explanatory of the difference between copper-plate printing from hollowed lines, and surface printing by means of the common press from prominent lines, still more intelligible. The subject is a representation of the copper-plate or rolling press.

Both the preceding impressions are produced in the same manner by means of the common printing-press. One is from wood; the other, where the white lines are seen on a black ground, is from copper;—the hollowed lines, which in copper-plate printing yield the impression, 5 receiving no ink from the printer’s balls or rollers; while the surface, which in copper-plate printing is wiped clean after the lines are filled with ink, is perfectly covered with it. It is, therefore, evident, that if this etching were printed in the same manner as other copper-plates, the impression would be a fac-simile of the one from wood. It has been judged necessary to be thus minute in explaining the difference between copper-plate and wood engraving, as the difference in the mode of obtaining impressions does not appear to have been previously pointed out with sufficient precision.

As it does not come within the scope of the present work to inquire into the origin of sculpture generally, I shall not here venture to give an opinion whether the art was invented by Adam or his good angel Raziel, or whether it was introduced at a subsequent period by Tubal-Cain, Noah, Trismegistus, Zoroaster, or Moses. Those who feel interested in such remote speculations will find the “authorities” in the second chapter of Evelyn’s “Sculptura.”

Without, therefore, inquiring when or by whom the art of engraving for the purpose of producing impressions was invented, I shall endeavour to show that such an art, however rude, was known at a very early period; and that it continued to be practised in Europe, though to a very limited extent, from an age anterior to the birth of Christ, to the year 1400. In the fifteenth century, its principles appear to have been more generally applied;—first, to the simple cutting of figures on wood for the purpose of being impressed on paper; next, to cutting figures and explanatory text on the same block, and then entire pages of text without figures, till the “ARS GRAPHICA ET IMPRESSORIA” attained its perfection in the discovery of PRINTING by means of movable fusile types.I.4

At a very early period stamps of wood, having hieroglyphic characters engraved on them, were used in Egypt for the purpose of producing impressions on bricks, and on other articles made of clay. This fact, which might have been inferred from the ancient bricks and fragments of earthenware containing characters evidently communicated by means of a stamp, has been established by the discovery of several of those wooden stamps, of undoubted antiquity, in the tombs at Thebes, Meroe, and other places. The following cuts represent the face and the back of one of the most perfect of those stamps, which was found in a tomb at Thebes, and has recently been brought to this country by Edward William Lane, Esq.I.5

  see text

The original stamp is made of the same kind of wood as the 6 mummy chests, and has an arched handle at the back, cut out of the same piece of wood as the face. It is of an oblong figure, with the ends rounded off; five inches long, two inches and a quarter broad, and half an inch thick. The hieroglyphic characters on its face are rudely cut in intaglio, so that their impression on clay would be in relief; and if printed in the same manner as the preceding copy, would present the same appearance,—that is, the characters which are cut into the wood, would appear white on a black ground. The phonetic power of the hieroglyphics on the face of the stamp may be represented respectively by the letters, A, M, N, F, T, P, T, H, M; and the vowels being supplied, as in reading Hebrew without points, we have the words, “Amonophtep, Thmei-mai,”—“Amonoph, beloved of truth.”I.6 The name is supposed to be that of Amonoph or Amenoph the First, the second king of the eighteenth dynasty, who, according to the best authorities, was contemporary with Moses, and reigned in Egypt previous to the departure of the Israelites. There are two ancient Egyptian bricks in the British Museum on which the impression of a similar stamp is quite distinct; and there are also several articles of burnt clay, of an elongated conical figure, and about nine inches long, which have their broader extremities impressed with hieroglyphics in a similar manner. There is also in the same collection a wooden 7 stamp, of a larger size than that belonging to Mr. Lane, but not in so perfect a condition. Several ancient Etruscan terra-cottas and fragments of earthenware have been discovered, on which there are alphabetic characters, evidently impressed from a stamp, which was probably of wood. In the time of Pliny terra-cottas thus impressed were called Typi.

  see text

In the British Museum are several bricks which have been found on the site of ancient Babylon. They are larger than our bricks, and somewhat different in form, being about twelve inches square and three inches thick. They appear to have been made of a kind of muddy clay with which portions of chopped straw have been mixed to cause it to bind; and their general appearance and colour, which is like that of a common brick before it is burnt, plainly enough indicate that they have not been hardened by fire, but by exposure to the sun. About the middle of their broadest surface, they are impressed with certain characters which have evidently been indented when the brick was in a soft state. The characters are indented,—that is, they are such as would be produced by pressing a wood-block with raised lines upon a mass of soft clay; and were such a block printed on paper in the usual manner of wood-cuts, the impression 8 would be similar to the preceding one, which has been copied, on a reduced scale, from one of the bricks above noticed. The characters have been variously described as cuneiform or wedge-shaped, arrow-headed, javelin-headed, or nail-headed; but their meaning has not hitherto been deciphered.

Amphoræ, lamps, tiles, and various domestic utensils, formed of clay, and of Roman workmanship, are found impressed with letters, which in some cases are supposed to denote the potter’s name, and in others the contents of the vessel, or the name of the owner. On the tiles,—of which there are specimens in the British Museum,—the letters are commonly inscribed in a circle, and appear raised; thus showing that the stamp had been hollowed, or engraved in intaglio, in a manner similar to a wooden butter-print. In a book entitled “Ælia Lælia Crispis non nata resurgens,” by C. C. Malvasia, 4to. Bologna, 1683, are several engravings on wood of such tiles, found in the neighbourhood of Rome, and communicated to the author by Fabretti, who, in the seventh chapter of his own work,I.7 has given some account of the “figlinarum signa,”—the stamps of the ancient potters and tile-makers.

  LAR

The stamp from which the following cut has been copied is preserved in the British Museum. It is of brass, and the letters are in relief and reversed; so that if it were inked from a printer’s ball and stamped on paper, an impression would be produced precisely the same as that which is here given.

It would be difficult now to ascertain why this stamp should be marked with the word Lar, which signifies a household god, or the image of the supposed tutelary genius of a house; but, without much stretch of imagination, we may easily conceive how appropriate such an inscription would be impressed on an amphora or large wine-vessel, sealed and set apart on the birth of an heir, and to be kept sacred—inviolate as the household gods—till the young Roman assumed the “toga virilis,” or arrived at years of maturity. That vessels containing wine were kept for many years, we learn from Horace and Petronius;I.8

9

——Prome reconditum,

Lyde, strenua, Cæcubum,

Munitæque adhibe vim sapientiæ.

Inclinare meridiem

Sentis: ac veluti stet volucris dies,

Parcis deripere horreo

Cessantem Bibuli Consulis amphoram.

Carmin. lib. III. xxviii.

“Quickly produce, Lyde, the hoarded Cæcuban, and make an attack upon wisdom, ever on her guard. You perceive the noontide is on its decline; and yet, as if the fleeting day stood still, you delay to bring out of the store-house the loitering cask, (that bears its date) from the Consul Bibulus.”—Smart’s Translation.

Mr. Ottley, in his “Inquiry into the Origin and Early History of Engraving,” pages 57 and 58, makes a distinction between impression where the characters impressed are produced by “a change of form”—meaning where they are either indented in the substance impressed, or raised upon it in relief—and impression where the characters are produced by colour; and requires evidence that the ancients ever used stamps “charged with ink or some other tint, for the purpose of stamping paper, parchment, or other substances, little or not at all capable of indentation.”

It certainly would be very difficult, if not impossible, to produce a piece of paper, parchment, or cloth of the age of the Romans impressed with letters in ink or other colouring matter; but the existence of such stamps as the preceding,—and there are others in the British Museum of the same kind, containing more letters and of a smaller size,—renders it very probable that they were used for the purpose of marking cloth, paper, and similar substances, with ink, as well as for being impressed in wax or clay.

Von Murr, in an article in his Journal, on the Art of Wood Engraving, gives a copy from a similar bronze stamp, in Praun’s Museum, with the inscription “Galliani,” which he considers as most distinctly proving that the Romans had nearly arrived at the arts of wood engraving and book printing. He adds: “Letters cut on wood they certainly had, and very likely grotesques and figures also, the hint of which their artists might readily obtain from the coloured stuffs which were frequently presented by Indian ambassadors to the emperors.”I.9

At page 90 of Singer’s “Researches into the History of Playing-Cards” are impressions copied from stamps similar to the preceding; 10 which stamps the author considers as affording “examples of such a near approach to the art of printing as first practised, that it is truly extraordinary there is no remaining evidence of its having been exercised by them;—unless we suppose that they were acquainted with it, and did not choose to adopt it from reasons of state policy.” It is just as extraordinary that the Greek who employed the expansive force of steam in the Ælopile to blow the fire did not invent Newcomen’s engine;—unless, indeed, we suppose that the construction of such an engine was perfectly known at Syracuse, but that the government there did not choose to adopt it from motives of “state policy.” It was not, however, a reason of “state policy” which caused the Roman cavalry to ride without stirrups, or the windows of the palace of Augustus to remain unglazed.

The following impressions are also copied from two other brass stamps, preserved in the collection of Roman antiquities in the British Museum.

  OVIRILLIO, FLSCLADIOU

As the letters in the originals are hollowed or cut into the metal, they would, if impressed on clay or soft wax, appear raised or in relief; and if inked and impressed on paper or on white cloth, they would present the same appearance that they do here—white on a black ground. Not being able to explain the letters on these stamps, further than that the first may be the dative case of a proper name Ovirillius, and indicate that property so marked belonged to such a person, I leave them, as Francis Moore, physician, leaves the hieroglyphic in his Almanack,—“to time and the curious to construe.”

11

Lambinet, in his “Recherches sur l’Origine de l’Imprimerie,” gives an account of two stone stamps of the form of small tablets, the letters of which were cut in intaglio and reverse, similar to the two of which impressions are above given. They were found in 1808, near the village of Nais, in the department of the Meuse; and as the letters, being in reverse, could not be made out, the owner of the tablets sent them to the Celtic Society of Paris, where M. Dulaure, to whose examination they were submitted, was of opinion that they were a kind of matrices or hollow stamps, intended to be applied to soft substances or such as were in a state of fusion. He thought they were stamps for vessels containing medical compositions; and if his reading of one of the inscriptions be correct, the practice of stamping the name of a quack and the nature of his remedy, in relief on the side of an ointment-pot or a bottle, is of high antiquity. The letters

Q. JUN. TAURI. ANODY.

NUM. AD OMN. LIPP.

M. Dulaure explains thus: Quinti Junii Tauridi anodynum ad omnes lippas;I.10 an inscription which is almost literally rendered by the title of a specific still known in the neighbourhood of Newcastle-on-Tyne, “Dr. Dud’s lotion, good for sore eyes.”

Besides such stamps as have already been described, the ancients used brands, both figured and lettered, with which, when heated, they marked their horses, sheep, and cattle, as well as criminals, captives, and refractory or runaway slaves.

The Athenians, according to Suidas, marked their Samian captives with the figure of an owl; while Athenians captured by the Samians were marked with the figure of a galley, and by the Syracusans with the figure of a horse. The husbandman at his leisure time, as we are informed by Virgil, in the first book of the Georgics,

“Aut pecori signa, aut numeros impressit acervis;”

and from the third book we learn that the operation was performed by branding:

“Continuoque notas et nomina gentis inurunt.”I.11


12

Such brands as those above noticed, commonly known by the name of cauteria or stigmata, were also used for similar purposes during the middle ages; and the practice, which has not been very long obsolete, of burning homicides in the hand, and vagabonds and “sturdy beggars” on the breast, face, or shoulder, affords an example of the employment of the brand in the criminal jurisprudence of our own country. By the 1st Edward VI. cap. 3, it was enacted, that whosoever, man or woman, not being lame or impotent, nor so aged or diseased that he or she could not work, should be convicted of loitering or idle wandering by the highway-side, or in the streets, like a servant wanting a master, or a beggar, he or she was to be marked with a hot iron on the breast with the letter V [for Vagabond], and adjudged to the person bringing him or her before a justice to be his slave for two years; and if such adjudged slave should run away, he or she, upon being taken and convicted, was to be marked on the forehead, or on the ball of the cheek, with the letter S [for Slave], and adjudged to be the said master’s slave for ever. By the 1st of James I. cap. 7, it was also enacted, that such as were to be deemed “rogues, vagabonds, and sturdy beggars” by the 39th of Elizabeth, cap. 4, being convicted at the sessions and found to be incorrigible, were to be branded in the left shoulder with a hot iron, of the breadth of an English shilling, marked with a great Roman R [for Rogue]; such branding upon the shoulder to be so thoroughly burned and set upon the skin and flesh, that the said letter R should be seen and remain for a perpetual mark upon such rogue during the remainder of his life.I.12

From a passage in Quintilian we learn that the Romans were acquainted with the method of tracing letters, by means of a piece of thin wood in which the characters were pierced or cut through, on a principle similar to that on which the present art of stencilling is founded. He is speaking of teaching boys to write, and the passage referred to may be thus translated: “When the boy shall have entered upon joining-hand, it will be useful for him to have a copy-head of wood in which the letters are well cut, that through its furrows, as it were, he may trace the characters with his style. He will not thus be liable to make slips as on the wax [alone], for he will be confined by the boundary of the letters, and neither will he be able to deviate from his text. By thus more rapidly and frequently following a definite outline, his hand will become set, without his requiring any assistance from the master to guide it.”I.13

13

A thin stencil-plate of copper, having the following letters cut out of it,

DN CONSTAN

TIO AVG SEM

PER VICTORI

was received, together with some rare coins, from Italy by Tristan, author of “Commentaires Historiques, Paris, 1657,” who gave a copy of it at page 68 of the third volume of that work. The letters thus formed, “ex nulla materia,”I.14 might be traced on paper by means of a pen, or with a small brush, charged with body-colour, as stencillers slap-dash rooms through their pasteboard patterns, or dipped in ink in the same manner as many shopkeepers now, through similar thin copper-plates, mark the prices of their wares, or their own name and address on the paper in which such wares are wrapped.

  see text

In the sixth century it appears, from Procopius, that the Emperor Justin I. made use of a tablet of wood pierced or cut in a similar manner, through which he traced in red ink, the imperial colour, his signature, consisting of the first four letters of his name. It is also stated that Theodoric, King of the Ostrogoths, the contemporary of Justin, used after the same manner to sign the first four letters of his name through a plate of gold;I.15 and in Peringskiold’s edition of the Life of Theodoric, the annexed is given as the monogramI.16 of that monarch. The authenticity of this account has, however, been questioned, as Cochlæus, who died in 1552, cites no ancient authority for the fact.

14

  see text

It has been asserted by Mabillon, (Diplom. lib. ii. cap. 10,) that Charlemagne first introduced the practice of signing documents with a monogram, either traced with a pen by means of a thin tablet of gold, ivory, or wood, or impressed with an inked stamp, having the characters in relief, in a manner similar to that in which letters are stamped at the Post-office.I.17 Ducange, however, states that this mode of signing documents is of greater antiquity, and he gives a copy of the monogram of the Pope Adrian I. who was elected to the see of Rome in 774, and died in 795. The annexed monogram of Charlemagne has been copied from Peringskiold, “Annotationes in Vitam Theodorici,” p. 584; it is also given in Ducange’s Glossary, and in the “Nouveau Traité de Diplomatique.”

The monogram, either stencilled or stamped, consisted of a combination of the letters of the person’s name, a fanciful character, or the figure of a cross,I.18 accompanied with a peculiar kind of flourish, called by French writers on diplomatics parafe or ruche. This mode of signing appears to have been common in most nations of Europe during the ninth, tenth, and eleventh centuries; and it was practised by nobles and the higher orders of the clergy, as well as by kings. It continued to be used by the kings of France to the time of Philip III. and by the Spanish monarchs to a much later period. It also appears to have been adopted by some of the Saxon kings of England; and the authors of the “Nouveau Traité de Diplomatique” say that they had seen similar marks produced by a stamp of William the Conqueror, when Duke of Normandy. We have had a recent instance of the use of the stampilla, as it is called by diplomatists, in affixing the royal signature. During the illness of George IV. in 1830, a silver stamp, containing a fac-simile of the king’s sign-manual, was executed by Wyon, which was stamped on documents requiring the royal signature, by commissioners, in his Majesty’s presence. A similar stamp was used during the last illness of Henry VIII. for the purpose of affixing the royal signature. The king’s warrant empowering commissioners to use the stamp may be seen in Rymer’s Fœdera, vol. xv. p. 101, anno 1546. It is believed that the 15 warrant which sent the poet Surrey to the scaffold was signed with this stamp, and not with Henry’s own hand.

In Sempère’s “History of the Cortes of Spain,” several examples are given of the use of fanciful monograms in that country at an early period, and which were probably introduced by its Gothic invaders. That such marks were stamped is almost certain; for the first, which is that of Gundisalvo Tellez, affixed to a charter of the date of 840, is the same as the “sign” which was affixed by his widow, Flamula, when she granted certain property to the abbot and monks of Cardeña for the good of her deceased husband’s soul. The second, which is of the date of 886, was used both by the abbot Ovecus, and Peter his nephew; and the third was used by all the four children of one Ordoño, as their “sign” to a charter of donation executed in 1018. The fourth mark is a Runic cypher, copied from an ancient Icelandic manuscript, and given by Peringskiold in his “Annotations on the Life of Theodoric:” it is not given here as being from a stencil or a stamp, but that it may be compared with the apparently Gothic monograms used in Spain.

  see text

“In their inscriptions, and in the rubrics of their books,” says a writer in the Edinburgh ReviewI.19 “the Spanish Goths, like the Romans of the Lower Empire, were fond of using combined capitals—of monogrammatising. This mode of writing is now common in Spain, on the sign-boards and on the shop-fronts, where it has retained its place in defiance of the canons of the council [of Leon], The Goths, however, retained a truly Gothic custom in their writings. The Spanish Goth sometimes subscribed his name; or he drew a monogram like the Roman emperors, or the sign of the cross like the Saxon; but not unfrequently he affixed strange and fanciful marks to the deed or charter, bearing a close resemblance to the Runic or magical knots of which so many have been engraved by Peringskiold, and other northern antiquaries.”

To the tenth or the eleventh century are also to be referred certain small silver coins—“something between counters and money,” as is observed by Pinkerton—which are impressed, on one side only, with a kind of Runic monogram. They are formed of very thin pieces of 16 silver; and it has been supposed that the impression was produced from wooden dies. They are known to collectors as “nummi bracteati”—tinsel money; and Pinkerton, mistaking the Runic character for the Christian cross, says that “most of them are ecclesiastic.” He is perhaps nearer the truth when he adds that they “belong to the tenth century, and are commonly found in Germany, and the northern kingdoms of Sweden and Denmark.”I.20 The four following copies from the original coins in the Brennerian collection are given by Peringskiold, in his “Annotations on the Life of Theodoric,” previously referred to. The characters on the three first he reads as the letters EIR, OIR, and AIR, respectively, and considers them to be intended to represent the name of Eric the Victorious. The characters on the fourth he reads as EIM, and applies them to Emund Annosus, the nephew of Eric the Victorious, who succeeded to the Sueo-Gothic throne in 1051; about which time, through the influence of the monks, the ancient Runic characters were exchanged for Roman.

  see text

  see text and caption

NICOLAUS FERENTERIUS, 1236

The notaries of succeeding times, who on their admission were required to use a distinctive sign or notarial mark in witnessing an instrument, continued occasionally to employ the stencil in affixing their “sign;” although their use of the stamp for that purpose appears to have been more general. In some of those marks or stamps the name of the notary does not appear, and in others a small space is left in order that it might afterwards be inserted with a pen. The annexed monogram was the official mark of an Italian notary, Nicolaus Ferenterius, who lived in 1236.I.21

The three following cuts represent impressions of German notarial stamps. The first is that of Jacobus Arnaldus, 1345; the second that of Johannes Meynersen, 1435; and the third that of Johannes Calvis, 1521.I.22

17

  see text and caption

JACOBUS ARNALDUS, 1345.

see text and caption

JOHANNES MEYNERSEN, 1435.

see text and caption

JOHANNES CALVIS, 1521.

Many of the merchants’-marks of our own country, which so frequently appear on stained glass windows, monumental brasses, and tombstones in the fourteenth, fifteenth, and sixteenth centuries, bear a considerable likeness to the ancient Runic monograms, from which it is not unlikely that they were originally derived. The English trader was accustomed to place his mark as his “sign” in his shop-front in the same manner as the Spaniard did his monogram: if he was a wool-stapler, he stamped it on his packs; or if a fish-curer, it was branded on the end of his casks. If he built himself a new house, his mark 18 was frequently placed between his initials over the principal door-way, or over the fireplace of the hall; if he made a gift to a church or a chapel, his mark was emblazoned on the windows beside the knight’s or the nobleman’s shield of arms; and when he died, his mark was cut upon his tomb. Of the following merchants’-marks, the first is that of Adam de Walsokne, who died in 1349; the second that of Edmund Pepyr, who died in 1483; those two marks are from their tombs in St. Margaret’s, Lynn; and the third is from a window in the same church.I.23

  see text

In Pierce Ploughman’s Creed, written after the death of Wickliffe, which happened in 1384, and consequently more modern than many of Chaucer’s poems, merchants’-marks are thus mentioned in the description of a window of a Dominican convent:

“Wide windows y-wrought, y-written full thick,

Shining with shapen shields, to shewen about,

With marks of merchants, y-meddled between,

Mo than twenty and two, twice y-numbered.I.24

Having thus endeavoured to prove by a continuous chain of evidence that the principle of producing impressions from raised lines was known, and practised, at a very early period; and that it was applied for the purpose of impressing letters and other characters on paper, though perhaps confined to signatures only, long previous to 1423,—which is the earliest date that has been discovered on a wood-cut, in the modern sense of the word, impressed on paper, and accompanied with explanatory words cut on the same block;I.25 and having shown that the principle of stencilling—the manner in which the above-named cut is 19 colouredI.26—was also known in the middle ages; it appears requisite, next to briefly notice the contemporary existence of the cognate arts of die-sinking, seal-cutting, and engraving on brass, and afterwards to examine the grounds of certain speculations on the introduction and early practice of wood-engraving and block-printing in Europe.

Concerning the first invention of stamping letters and figures upon coins, and the name of the inventor, it is fruitless to inquire, as the origin of the art is lost in the remoteness of antiquity. “Leaving these uncertainties,” says Pinkerton, in his Essay on Medals, “we know from respectable authorities that the first money coined in Greece was that struck in the island of Ægina, by Phidon king of Argos. His reign is fixed by the Arundelian marbles to an era correspondent to the 885th year before Christ; but whether he derived this art from Lydia or any other source we are not told.” About three hundred years before the birth of Christ, the art of coining, so far as relates to the beauty of the heads impressed, appears to have attained its perfection in Greece;—we may indeed say its perfection generally, for the specimens which were then produced in that country remain unsurpassed by modern art. Under the Roman emperors the art never seems to have attained so high a degree of perfection as it did in Greece; though several of the coins of Hadrian, probably executed by Greek artists, display great beauty of design and execution. The art of coining, with the rest of the ornamental arts, declined with the empire; and, on its final subversion in Italy, the coins of its rulers were scarcely superior to those which were subsequently minted in England, Germany, and France, during the darkest period of the middle ages.

The art of coining money, however rude in design and imperfect in its mode of stamping the impression, which was by repeated blows with a hammer, was practised from the twelfth to the sixteenth century in a greater number of places than at present; for many of the more powerful bishops and nobles assumed or extorted the right of coining money as well as the king; and in our own country the archbishops of Canterbury and York, and the bishop of Durham, exercised the right of coinage till the Reformation; and local mints for coining the king’s money were occasionally fixed at Norwich, Chester, York, St. Edmundsbury, Newcastle-on-Tyne, and other places. Independent of those establishments for the coining of money, almost every abbey struck its own jettons or 20 counters; which were thin pieces of copper, commonly impressed with a pious legend, and used in casting up accounts, but which the general introduction of the numerals now in use, and an improved system of arithmetic, have rendered unnecessary. As such mints were at least as numerous in France and Germany as in our own country, Scheffer, the partner of Faust, when he conceived the idea of casting letters from matrices formed by punches, would have little difficulty in finding a workman to assist him in carrying his plans into execution. “The art of impressing legends on coins,” says Astle in his Account of the Origin and Progress of writing, “is nothing more than the art of printing on medals.” That the art of casting letters in relief, though not separately, and most likely from a mould of sand, was known to the Romans, is evident from the names of the emperors Domitian and Hadrian on some pigs of lead in the British Museum; and that it was practised during the middle and succeeding ages, we have ample testimony from the inscriptions on our ancient bells.I.27

In the century immediately preceding 1423, the date of the wood-cut of St. Christopher, the use of seals, for the purpose of authenticating documents by their impression on wax, was general throughout Europe; kings, nobles, bishops, abbots, and all who “came of gentle blood,” with corporations, lay and clerical, all had seals. They were mostly of brass, for the art of engraving on precious stones does not appear to have been at that time revived, with the letters and device cut or cast in hollow—en creux—on the face of the seal, in order that the impression might appear raised. The workmanship of many of those seals, and more especially of some of the conventional ones, where figures of saints and a view of the abbey are introduced, displays no mean degree of skill. Looking on such specimens of the graver’s art, and bearing in mind the character of many of the drawings which are to be seen in the missals and other manuscripts of the fourteenth century and of the early part of the fifteenth, we need no longer be surprised that the cuts of the earliest block-books should be so well executed.

The art of engraving on copper and other metals, though not with the intention of taking impressions on paper, is of great antiquity. In the late Mr. Salt’s collection of Egyptian antiquities there was a small axe, probably a model, the head of which was formed of sheet-copper, and was tied, or rather bandaged, to the helve with slips of cloth. There were certain characters engraved upon the head in such a manner that if it were inked and submitted to the action of the rolling-press, impressions would be obtained as from a modern copper-plate. The axe, with other 21 models of a carpenter’s tools, also of copper, was found in a tomb in Egypt, where it must have been deposited at a very early period. That the ancient Greeks and Romans were accustomed to engrave on copper and other metals in a similar manner, is evident from engraved pateræ and other ornamental works executed by people of those nations. Though no ancient writer makes mention of the art of engraving being employed for the purpose of producing impressions on paper, yet it has been conjectured by De Pauw, from a passage in Pliny,I.28 that such an art was invented by Varro for the purpose of multiplying the portraits of eminent men. “No Greek,” says De Pauw, speaking of engraving, “has the least right to claim this invention, which belongs exclusively to Varro, as is expressed by Pliny in no equivocal terms, when he calls this method inventum Varronis. Engraved plates were employed which gave the profile and the principal traits of the figures, to which the appropriate colours and the shadows were afterwards added with the pencil. A woman, originally of Cyzica, but then settled in Italy, excelled all others in the talent of illumining such kind of prints, which were inserted by Varro in a large work of his entitled ‘Imagines’ or ‘Hebdomades,’ which was enriched with seven hundred portraits of distinguished men, copied from their statues and busts. The necessity of exactly repeating each portrait or figure in every copy of the work suggested the idea of multiplying them without much cost, and thus gave birth to an art till then unknown.”I.29 The grounds, however, of this conjecture are extremely slight, and will not without additional support sustain the superstructure which De Pauw—an “ingenious” guesser, but a superficial inquirer—has so plausibly raised. A prop for this theory has been sought for by men of greater research than the original propounder, but hitherto without success.

About the year 1300 we have evidence of monumental brasses, with large figures engraved on them, being fixed on tombs in this country; and it is not unlikely that they were known both here and on the 22 Continent at an earlier period. The best specimens known in this country are such as were in all probability executed previous to 1400. In the succeeding century the figures and ornamental work generally appear to be designed in a worse taste and more carelessly executed; and in the age of Queen Elizabeth the art, such as it was, appears to have reached the lowest point of degradation, the monumental brasses of that reign being generally the worst which are to be met with.

The figures on several of the more ancient brasses are well drawn, and the folds of the drapery in the dresses of the females are, as a painter would say, “well cast;” and the faces occasionally display a considerable degree of correct and elevated expression. Many of the figures are of the size of life, marked with a hold outline well ploughed into the brass, and having the features, armour, and drapery indicated by single lines of greater or less strength as might be required. Attempts at shading are also occasionally to be met with; the effect being produced by means of lines obliquely crossing each other in the manner of cross-hatchings. Whether impressions were ever taken or not from such early brasses by the artists who executed them, it is perhaps now impossible to ascertain; but that they might do so is beyond a doubt, for it is now a common practice, and two immense volumes of impressions taken from monumental brasses, for the late Craven Ord, Esq., are preserved in the print-room of the British Museum.

One of the finest monumental brasses known in this country is that of Robert Braunche and his two wives, in St. Margaret’s Church, Lynn, where it appears to have been placed about the year 1364. Braunche, and his two wives, one on each side of him, are represented standing, of the size of life. Above the figures are representations of five small niches surmounted by canopies in the florid Gothic style. In the centre niche is the figure of the Deity holding apparently the infant Christ in his arms. In each of the niches adjoining the centre one is an angel swinging a censer; and in the exterior niches are angels playing on musical instruments. At the sides are figures of saints, and at the foot there is a representation of a feast, where persons are seen seated at table, others playing on musical instruments, while a figure kneeling presents a peacock. The length of this brass is eight feet eleven inches, and its breadth five feet two inches. It is supposed to have been executed in Flanders, with which country at that period the town of Lynn was closely connected in the way of trade.I.30

It has frequently been asserted that the art of wood engraving in Europe was derived from the Chinese; by whom, it is also said, that the 23 art was practised in the reign of the renowned emperor Wu-Wang, who flourished 1120 years before the birth of Christ. As both these statements seem to rest on equal authorities, I attach to each an equal degree of credibility; that is, by believing neither. As Mr. Ottley has expressed an opinion in favour of the Chinese origin of the art,—though without adopting the tale of its being practised in the reign of Wu-Wang, which he shows has been taken by the wrong end,—I shall here take the liberty of examining the tenability of his arguments.

At page 8, in the first chapter of his work, Mr. Ottley cautiously says that the “art of printing from engraved blocks of wood appears to be of very high antiquity amongst the Chinese;” and at page 9, after citing Du Halde, as informing us that the art of printing was not discovered until about fifty years before the Christian era, he rather inconsistently observes: “So says Father Du Halde, whose authority I give without any comment, as the defence of Chinese chronology makes no part of the present undertaking.” Unless Mr. Ottley is satisfied of the correctness of the chronology, he can by no means cite Du Halde’s account as evidence of the very high antiquity of printing in China; which in every other part of his book he speaks of as a well-established fact, and yet refers to no other authority than Du Halde, who relies on the correctness of that Chinese chronology with the defence of which Mr. Ottley will have nothing to do.

It is also worthy of remark, that in the same chapter he corrects two writers, Papillon and Jansen, for erroneously applying a passage in Du Halde as proving that the art of printing was known in the reign of Wu-Wang,—he who flourished Ante Christum 1120; whereas the said passage was not alleged “by Du Halde to prove the antiquity of printing amongst the Chinese, but solely in reference to their ink.” The passage, as translated by Mr. Ottley, is as follows: “As the stone Me” (a word signifying ink in the Chinese language), “which is used to blacken the engraved characters, can never become white; so a heart blackened by vices will always retain its blackness.” The engraved characters were not inked, it appears, for the purpose of taking impressions, as Messrs. Papillon and Jansen have erroneously inferred. “It is possible,” according to Mr. Ottley, “that the ink might be used by the Chinese at a very early period to blacken, and thereby render more easily legible, the characters of engraved inscriptions.”I.31 The possibility of this may be granted certainly; but at the same time we must admit that it is equally possible that the engraved characters were blackened with ink for the purpose of being printed, if they were of wood; or that, if 24 cut in copper or other metal, they were filled with a black composition which would harden or set in the lines,—as an ingenious inquirer might infer from ink being represented by the stone ME; and thus it is possible that something very like “niello,” or the filling of letters on brass doorplates with black wax, was known to the Chinese in the reign of Wu-Wang, who flourished in the year before our Lord, 1120. The one conjecture is as good as the other, and both good for nothing, until we have better assurance than is afforded by Du Halde, that engraved characters blackened with ink—for whatever purpose—were known by the Chinese in the reign of Wu-Wang.I.32

Although so little is positively known of the ancient history of “the great out-lying empire of China,” as it is called by Sir William Jones, yet it has been most confidently referred to as affording authentic evidence of the high degree of the civilization and knowledge of the Chinese at a period when Europe was dark with the gloom of barbarism and ignorance. Their early history has been generally found, when opportunity has been afforded of impartially examining it, to be a mere tissue of absurd legends; compared to which, the history of the settlement of King Brute in Britain is authentic. With astronomy as a science they are scarcely acquainted; and their specimens of the fine arts display little more than representations of objects executed not unfrequently with minute accuracy, but without a knowledge of the most simple elements of correct design, and without the slightest pretensions to art, according to our standard.

One of the two Mahometan travellers who visited China in the ninth century, expressly states that the Chinese were unacquainted with the sciences; and as neither of them takes any notice of printing, the mariner’s compass, or gunpowder, it seems but reasonable to conclude that the Chinese were unacquainted with those inventions at that period.I.33

Mr. Ottley, at pages 51 and 52 of his work, gives a brief account of 25 the early commerce of Venice with the East, for the purpose of showing in what manner a knowledge of the art of printing in China might be obtained by the Venetians. He says: “They succeeded, likewise, in establishing a direct traffic with Persia, Tartary, China, and Japan; sending, for that purpose, several of their most respectable citizens, and largely providing them with every requisite.” He cites an Italian author for this account, but he observes a prudent silence as to the period when the Venetians first established a direct traffic with China and Japan; though there is little doubt that Bettinelli, the authority referred to, alludes to the expedition of the two brothers Niccolo and Maffeo Polo, and of Marco Polo, the son of Niccolo, who in 1271 or 1272 left Venice on an expedition to the court of the Tartar emperor Kublai-Khan, which had been previously visited by the two brothers at some period between 1254 and 1269.I.34 After having visited Tartary and China, the two brothers and Marco returned to Venice in 1295. Mr. Ottley, however, does not refer to the travels of the Polos for the purpose of showing that Marco, who at a subsequent period wrote an account of his travels, might introduce a knowledge of the Chinese art of printing into Europe: he cites them that his readers may suppose that a direct intercourse between Venice and China had been established long before; and that the art of engraving wood-blocks, and taking impressions from them, had been thus derived from the latter country, and had been practised in Venice long before the return of the travellers in 1295.

It is necessary here to observe that the invention of the mariner’s compass, and of gunpowder and cannon, have been ascribed to the Chinese as well as the invention of wood engraving and block-printing; and it has been conjectured that very probably Marco Polo communicated to his countrymen, and through them to the rest of Europe, a knowledge of those arts. Marco Polo, however, does not in the account which he wrote of his travels once allude to gunpowder, cannon, or to the art of printing as being known in China;I.35 nor does he once mention the compass as being used on board of the Chinese vessel in which he sailed from the coast of China to the Persian Gulf. “Nothing is more common,” 26 says a writer in the Quarterly Review, “than to find it repeated from book to book, that gunpowder and the mariner’s compass were first brought from China by Marco Polo, though there can be very little doubt that both were known in Europe some time before his return.”—“That Marco Polo,” says the same writer, “would have mentioned the mariner’s compass, if it had been in use in China, we think highly probable; and his silence respecting gunpowder may be considered as at least a negative proof that this also was unknown to the Chinese in the time of Kublai-Khan.”I.36 In a manner widely different from this does Mr. Ottley reason, respecting the cause of Marco Polo not having mentioned printing as an art practised by the Chinese. He accounts for the traveller’s silence as follows: “Marco Polo, it may be said, did not notice this art [of engraving on wood and block-printing] in the account which he left us of the marvels he had witnessed in China. The answer to this objection is obvious: it was no marvel; it had no novelty to recommend it; it was practised, as we have seen, at Ravenna, in 1285, and had perhaps been practised a century earlier in Venice. His mention of it, therefore, was not called for, and he preferred instructing his countrymen in matters with which they were not hitherto acquainted.” This “obvious” answer, rather unfortunately, will equally apply to the question, “Why did not Marco Polo mention cannon as being used by the Chinese, who, as we are informed, had discovered such formidable engines of war long before the period of his visit?”

That the art of engraving wood-blocks and of taking impressions from them was introduced into Europe from China, I can see no sufficient reason to believe. Looking at the frequent practice in Europe, from the twelfth to the fifteenth century, of impressing inked stamps on paper, I can perceive nothing in the earliest specimens of wood engraving but the same principles applied on a larger scale. When I am once satisfied that a man had built a small boat, I feel no surprise on learning that his grandson had built a larger; and made in it a longer voyage than his ancestor ever ventured on, who merely used his slight skiff to ferry himself across a river.

In the first volume of Papillon’s “Traité de la Gravure en Bois,” there is an account of certain old wood engravings which he professes to have seen, and which, according to their engraved explanatory title, were executed by two notable young people, Alexander Alberic Cunio, knight, and Isabella Cunio, his twin sister, and finished by them when they were only sixteen years old, at the time when Honorius IV. was pope; that is, at some period between the years 1285 and 1287. This 27 story has been adopted by Mr. Ottley, and by Zani, an Italian, who give it the benefit of their support. Mr. Singer, in his “Researches into the History of Playing Cards,” grants the truth-like appearance of Papillon’s tale; and the writer of the article “Wood-engraving” in the Encyclopedia Metropolitana considers it as authentic. It is, however, treated with contempt by Heineken, Huber, and Bartsch, whose knowledge of the origin and progress of engraving is at least equal to that of the four writers previously named.

The manner in which Papillon recovered his memoranda of the works of the Cunio is remarkable. In consequence of those curious notes being mislaid for upwards of thirty-five years, the sole record of the productions of those “ingenious and amiable twins” was very nearly lost to the world. The three sheets of letter-paper on which he had written an account of certain old volumes of wood engravings,—that containing the cuts executed by the Cunio being one of the number,—he had lost for upwards of thirty-five years. For long he had only a confused idea of those sheets, though he had often searched for them in vain, when he was writing his first essay on wood engraving, which was printed about 1737, but never published. At length he accidentally found them, on All-Saints’ Day, 1758, rolled up in a bundle of specimens of paper-hangings which had been executed by his father. The finding of those three sheets afforded him the greater pleasure, as from them he discovered, by means of a pope’s name, an epoch of engraving figures and letters on wood for the purpose of being printed, which was certainly much earlier than any at that period known in Europe, and at the same time a history relative to this subject equally curious and interesting. He says that he had so completely forgotten all this,—though he had so often recollected to search for his memoranda,—that he did not deign to take the least notice of it in his previously printed history of the art. The following is a faithful abstract of Papillon’s account of his discovery of those early specimens of wood engraving. The title-page, as given by him in French from Monsieur De Greder’s vivâ voce translation of the original,—which was “en mauvais Latin ou ancien Italien Gothique, avec beaucoup d’abréviations,”—is translated without abridgment, as are also his own descriptions of the cuts.

“When young, being engaged with my father in going almost every day to hang rooms with our papers, I was, some time in 1719 or 1720, at the village of Bagneux, near Mont Rouge, at a Monsieur De Greder’s, a Swiss captain, who had a pretty house there. After I had papered a small room for him, he ordered me to cover the shelves of his library with paper in imitation of mosaic. One day after dinner he surprised me reading a book, which occasioned him to show me some very old ones which he had borrowed of one of his friends, a Swiss officer,I.37 that 28 he might examine them at his leisure. We talked about the figures which they contained, and of the antiquity of wood engraving; and what follows is a description of those ancient books as I wrote it before him, and as he was so kind as to explain and dictate to me

“In a cartouchI.38 or frontispiece,—of fanciful and Gothic ornaments, though pleasing enough,—nine inches wide, and six inches high, having at the top the arms, doubtless, of Cunio, the following words are coarsely engraved on the same block, in bad Latin, or ancient Gothic Italian with many abbreviations.

“‘The chivalrous deeds, in figures, of the great and magnanimous Macedonian king, the courageous and valiant Alexander, dedicated, presented, and humbly offered to the most holy father, Pope Honorius IV. the glory and stay of the Church, and to our illustrious and generous father and mother, by us Alexander Alberic Cunio, knight, and Isabella Cunio, twin brother and sister; first reduced, imagined, and attempted to be executed in relief with a little knife, on blocks of wood, joined and smoothed by this learned and beloved sister, continued and finished together at Ravenna, after eight pictures of our designing, painted six times the size here represented; cut, explained in verse, and thus marked on paper to multiply the number, and to enable us to present them as a token of friendship and affection to our relations and friends. This was done and finished, the age of each being only sixteen years complete.’”

After having given the translation of the title-page, Papillon thus continues the narrative in his own person: “This cartouch [or ornamented title-page] is surrounded by a coarse line, the tenth of an inch broad, forming a square. A few slight lines, which are irregularly executed and without precision, form the shading of the ornaments. The impression, in the same manner as the rest of the cuts, has been taken in Indian blue, rather pale, and in distemper, apparently by the hand being passed frequently over the paper laid upon the block, as card-makers are accustomed to impress their addresses and the envelopes of their cards. The hollow parts of the block, not being sufficiently cut away in several places, and having received the ink, have smeared the paper, which is rather brown; a circumstance which has caused the following words to be written in the margin underneath, that the fault might be remedied. 29 They are in Gothic Italian, which M. de Greder had considerable difficulty in making out, and certainly written by the hand either of the Chevalier Cunio or his sister, on this first proof—evidently from a block—such as are here translated.”

“‘It is necessary to cut away the ground of the blocks more, that the paper may not touch it in taking impressions.’”

“Following this frontispiece, and of the same size, are the subjects of the eight pictures, engraved on wood, surrounded by a similar line forming a square, and also with the shadows formed of slight lines. At the foot of each of those engravings, between the border-line and another, about a finger’s breadth distant, are four Latin verses engraved on the block, poetically explaining the subject, the title of which is placed at the head. In all, the impression is similar to that of the frontispiece, and rather grey or cloudy, as if the paper had not been moistened. The figures, tolerably designed, though in a semi-gothic taste, are well enough characterized and draped; and we may perceive from them that the arts of design were then beginning gradually to resume their vigour in Italy. At the feet of the principal figures their names are engraved, such as Alexander, Philip, Darius, Campaspe, and others.”

Subject 1.—Alexander mounted on Bucephalus, which he has tamed. On a stone are these words: Isabel. Cunio pinx. & scalp.

Subject 2.—Passage of the Granicus. Near the trunk of a tree these words are engraved: Alex. Alb. Cunio Equ. pinx. Isabel Cunio scalp.

Subject 3.—Alexander cutting the Gordian knot. On the pedestal of a column are these words: Alexan. Albe. Cunio Equ. pinx. & scalp. This block is not so well engraved as the two preceding.”

Subject 4.—Alexander in the tent of Darius. This subject is one of the best composed and engraved of the whole set. Upon the end of a piece of cloth are these words: Isabel. Cunio pinxit & scalp.

Subject 5.—Alexander generously presents his mistress Campaspe to Apelles who was painting her. The figure of this beauty is very agreeable. The painter seems transported with joy at his good fortune. On the floor, on a kind of antique tablet, are these words: Alex. Alb. Cunio Eques, pinx. & scalp.

Subject 6.—The famous battle of Arbela. Upon a small hillock are these words: Alex. Alb. Equ. & Isabel. pictor. and scalp. For composition, design, and engraving, this subject is also one of the best.”

Subject 7.—Porus, vanquished, is brought before Alexander. This subject is so much the more beautiful and remarkable, as it is composed nearly in the same manner as that of the famous Le Brun; it would seem that he had copied this print. Both Alexander and Porus have a grand 30 and magnanimous air. On a stone near a bush are engraved these words: Isabel. Cunio pinx. & scalp.

Subject 8 and last.—The glory and grand triumph of Alexander on entering Babylon. This piece, which is well enough composed, has been executed, as well as the sixth, by the brother and sister conjointly, as is testified by these characters engraved at the bottom of a wall: Alex. Alb. Equ. et Isabel. Cunio, pictor. & scalp. At the top of this impression, a piece about three inches long and one inch broad has been torn off.”

However singular the above account of the works of those “amiable twins” may seem, no less surprising is the history of their birth, parentage, and education; which, taken in conjunction with the early development of their talents as displayed in such an art, in the choice of such a subject, and at such a period, is scarcely to be surpassed in interest by any narrative which gives piquancy to the pages of the Wonderful Magazine.

Upon the blank leaf adjoining the last engraving were the following words, badly written in old Swiss characters, and scarcely legible in consequence of their having been written with pale ink. “Of course Papillon could not read Swiss,” says Mr. Ottley, “M. de Greder, therefore, translated them for him into French.”—“This precious volume was given to my grandfather Jan. Jacq. Turine, a native of Berne, by the illustrious Count Cunio, chief magistrate of Imola, who honoured him with his generous friendship. Above all my books I prize this the highest on account of the quarter from whence it came into our family, and on account of the knowledge, the valour, the beauty, and the noble and generous desire which those amiable twins Cunio had to gratify their relations and friends. Here ensues their singular and curious history as I have heard it many a time from my venerable father, and which I have caused to be more correctly written than I could do it myself.”

Though Papillon’s long-lost manuscript, containing the whole account of the works of the Cunio and notices of other old books of engravings, consisted of only three sheets of letter-paper, yet the history alone of the learned, beautiful, and amiable twins, which Turine the grandson caused to be written out as he had heard it from his father, occupies in Papillon’s book four long octavo pages of thirty-eight lines each. To assume that his long-lost manuscript consisted of brief notes which he afterwards wrote out at length from memory, would at once destroy any validity that his account might be supposed to possess; for he states that he had lost those papers for upwards of thirty five years, and had entirely forgotten their contents.

Without troubling myself to transcribe the whole of this choice morsel of French Romance concerning the history of the “amiable 31 twins” Cunio,—the surprising beauty, talents, and accomplishments of the maiden,—the early death of herself and her lover,—the heroism of the youthful knight, Alexander Alberic Cunio, displayed when only fourteen years old,—I shall give a brief abstract of some of the passages which seem most important to the present inquiry.I.39

From this narrative,—which Papillon informs us was written in a much better hand, though also in Swiss characters, and with much blacker ink than Turine the grandson’s own memorandum,—we obtain the following particulars: The Count de Cunio, father of the twins, was married to their mother, a noble maiden of Verona and a relation of Pope Honorius IV. without the knowledge of their parents, who, on discovering what had happened, caused the marriage to be annulled, and the priest by whom it was celebrated to be banished. The divorced wife, dreading the anger of her own father, sought an asylum with one of her aunts, under whose roof she was brought to bed of twins. Though the elder Cunio had compelled his son to espouse another wife, he yet allowed him to educate the twins, who were most affectionately received and cherished by their father’s new wife. The children made astonishing progress in the sciences, more especially the girl Isabella, who at thirteen years of age was regarded as a prodigy; for she understood, and wrote with correctness, the Latin language; she composed excellent verses, understood geometry, was acquainted with music, could play on several instruments, and had begun to design and to paint with correctness, taste, and delicacy. Her brother Alberic, of a beauty as ravishing as his sister’s, and one of the most charming youths in Italy, at the age of fourteen could manage the great horse, and understood the practice of arms and all other exercises befitting a young man of quality. He also understood Latin, and could paint well.

The troubles in Italy having caused the Count Cunio to take up arms, his son, young Alexander Alberic, accompanied him to the field to make his first campaign. Though not more than fourteen years old, he was entrusted with the command of a squadron of twenty-five horse, with which, as his first essay in war, he attacked and put to flight near two hundred of the enemy. His courage having carried him too far, he was surrounded by the fugitives, from whom, however, he fought himself clear without any further injury than a wound in his left arm. His father, who had hastened to his succour, found him returning with the enemy’s banner, which he had wrapped about his wound. Delighted at the valour displayed by his son, the Count Cunio knighted him on the spot. The young man then asked permission to visit his mother, which 32 was readily granted by the count, who was pleased to have this opportunity of testifying the love and esteem he still retained towards that noble and afflicted lady, who continued to reside with her aunt; of which he certainly would have given her more convincing proofs, now that his father was dead, by re-establishing their marriage and publicly espousing her, if he had not been in duty bound to cherish the wife whom he had been compelled to marry, and who had now borne him a large family.

After Alexander Alberic had visited his mother, he returned home, and shortly after began, together with his sister Isabella, to design and work upon the pictures of the achievements of Alexander. He then made a second campaign with his father, after which he continued to employ himself on the pictures in conjunction with Isabella, who attempted in reduce them and engrave them on wood. After the engravings were finished, and copies had been printed and given to Pope Honorius, and their relations and friends, Alexander Alberic proceeded again to join the army, accompanied by Pandulphio, a young nobleman, who was in love with the charming Isabella. This was his last campaign, for he was killed in the presence of his friend, who was dangerously wounded in defending him. He was slain when not more than nineteen; and his sister was so affected by his death that she resolved never to marry, and died when she was scarcely twenty. The death of this lovely and learned young lady was followed by that of her lover, who had fondly hoped that she would make him happy. The mother of those amiable twins was not long in following them to the grave, being unable to survive the loss of her children. The Countess de Cunio took seriously ill at the loss of Isabella, but fortunately recovered; and it was only the count’s grandeur of soul that saved him from falling sick also.

Some years after this, Count Cunio gave the copy of the achievements of Alexander, in its present binding, to the grandfather of the person who caused this account to be written. The binding, according to Papillon’s description of it, was, for the period, little less remarkable than the contents. “This ancient and Gothic binding,” as Papillon’s note is translated by Mr. Ottley. “is made of thin tablets of wood, covered with leather, and ornamented with flowered compartments, which appear simply stamped and marked with an iron a little warmed, without any gilding.” It is remarkable that this singular volume should afford not only specimens of wood engraving, earlier by upwards of a hundred and thirty years than any which are hitherto known, but that the binding, of the same period as the engravings, should also be such as is rarely, if ever, to be met with till upwards of one hundred and fifty years after the wonderful twins were dead.

33

As this volume is no longer to be found, as no mention is made of such a work by any old writer, and as another copy has not been discovered in any of the libraries of Italy, nor the least trace of one ever having been there, the evidence of its ever having existed rests solely on the account given of it by Papillon. Before saying a word respecting the credit to be attached to this witness, or the props with which Zani and Ottley endeavour to support his testimony, I shall attempt to show that the account affords internal evidence of its own falsehood.

Before noticing the description of the subjects, I shall state a few objections to the account of the twins as written out by order of the youngest Turine, the grandson of Jan. Jacq. Turine, who received the volume from Count Cunio himself, the father of the twins, a few years after their death, which could not well happen later than 1291; as Pope Honorius, to whom their work was dedicated when they were sixteen years old, died in 1287, and Isabella Cunio, who survived her brother, died when she was not more than twenty. Supposing that Count Cunio gave the volume to his friend, J. J. Turine, a native of Berne, in 1300, and that the grandson of the latter caused the history of the twins to be written out eighty years afterwards,—and we cannot fairly assume that it was written later, if indeed so late,—we have thus 1380 as the date of the account written “in old Swiss characters, in a better hand, and with much blacker ink,” than the owner’s own memorandum of the manner in which the volume came into his family, and his reasons for prizing it so highly. The probable date of the pretended Swiss history of the Cunio, Papillon’s advocates carefully keep out of sight; for what impartial person could believe that a Swiss of the fourteenth century could give utterance to the sentimental fustian which forms so considerable a portion of the account? Of the young knight Cunio he knows every movement; he is acquainted with his visit to his repudiated mother; he knows in which arm he was wounded; the number of men that he lost, when with only five-and-twenty he routed two hundred; the name of Isabella’s lover; the illness and happy recovery of Count Cunio’s wife, and can tell the cause why the count himself did not fall sick.

To any person who reflects on the doctrine of the church of Rome in the article of marriage, it certainly must appear strange that the parents of the Count Cunio and his first wife, the mother of the twins, should have had the power of dissolving the marriage and of banishing the priest by whom it was solemnized; and still more singular it is that the Count Cunio, whom we must suppose to have been a good Catholic, should speak, after his father’s death, of re-establishing his marriage with his first wife and of publicly espousing her; and that he should make such a communication to her through the medium of her son, who, 34 as well as his sister, must have been declared illegitimate by the very fact of their mother’s divorce. It is also strange that this piece of family history should come to the knowledge of the grandson of Jan. Jacq. Turine. The Count Cunio’s second marriage surely must have been canonically legal, if the first were not; and if so, it would not be a sense of duty alone to his second wife that would prevent him divorcing her and re-marrying the first. On such subjects the church was to be consulted; and to such playing fast-and-loose with the sacrament of marriage the church said “NO.” Taking these circumstances into consideration, I can come to no other conclusion than that, on this point, the writer of the history of the Cunio did not speak truth; and that the paper containing such history, even if it could be produced, is not genuine, as every other part of it which has the slightest bearing on the point at issue, is equally, if not more, improbable.

With respect to the cuts pretended to be executed by the twins themselves, I shall waive any objections which might be urged on the ground of it being unlikely that they should be executed by a boy and a girl so young. Supposing that the twins were as learned and accomplished as they are represented, still it would be a very surprising circumstance that, in the thirteenth century, they should have executed a series of wood engravings of the actions of Alexander the Great as an appropriate present to the pope; and that the composition of one of those subjects, No. 7, should so closely resemble one of Le Brun’s—an artist remarkable for the complication of his designs—that it would seem he had copied this very print. Something like the reverse of this is more probable; that the description of the pretended work of the Cunio was suggested by the designs of Le Brun.I.40 The execution of a set of designs, in the thirteenth century, illustrating the actions of Alexander in the manner described by Papillon, would be a rarity indeed even if not engraved on wood; but that a series of wood engravings, and not a saint in one of them, should be executed by a boy and a girl, and presented to a pope, in 1286, is scarcely short of miraculous. The twins must have been well read in Quintus Curtius. Though we are informed that both were skilled in the Latin language, yet it plainly appears on two occasions, when we might suppose that they would be least liable to trip, that their Latinity is questionable. The sixth and the eighth subjects, which were accomplished by their joint efforts, are 35 described as being marked: Alex. Alb. Equ. et Isabel, Cunio pictor. et scalp.

“Thus painters did not write their names at Co.”

Why do not the advocates of those early specimens of wood engraving in Italy point out to their readers that these two children were the first who ever affixed the words pinx. et scalp. to a woodcut? I challenge any believer in Papillon to point out a wood engraving on which the words pinxit and scalpsit, the first after the painter’s name, and the second after the engraver’s, appear previous to 1580. This apparent copying—and by a person ignorant of Latin too—of the formula of a later period, is of itself sufficient to excite a suspicion of forgery; and, coupled with the improbable circumstances above related, it irresistibly compels me to conclude that the whole account is a mere fiction.

With respect to the credibility of Papillon, the sole evidence upon which the history of the wonderful twins rests, I shall have occasion to say very few words. That he was credulous, and excessively vain of what he considered his discoveries in the history of wood engraving, is admitted by those who profess to believe him. He appears also from an early age to have been subject to mental hallucination; and in 1759, the year after he found his papers containing the account of the Cunio, he had a fit of decided insanity which rendered it necessary to convey him to a mad-house, where by copious bleeding he soon recovered his senses.I.41 To those interested in the controversy I leave to decide how far the unsupported testimony of such a person, and in such a case, ought to be relied on. How easily he might be deceived on a subject relating to the early history of his art, it is not difficult to comprehend; and even allowing him to be sincere in the belief of what he related, he was a person very likely to occasionally deceive both himself and others.I.42

Papillon’s insanity had been previously adverted to by Heineken; and this writer’s remarks have produced the following correction from Mr. Ottley: “Heineken takes some pains to show that poor Papillon was not in his right mind; and, amongst his other arguments, quotes a passage 36 from his book, t. i. p. 335, in which he says, ‘Par un accident et une fatalité commune à plusieurs graveurs, aussi bien qu’à moi, Le Fevre est devenu aliéné d’esprit:’ as if a little pleasantry of expression, such as the French writers, especially, have ever felt themselves at full liberty to indulge in, could really constitute fit grounds for a statute of lunacy.”I.43 Had Mr. Ottley, instead of confidently correcting Heineken when the latter had stated nothing but the fact, turned to the cited page of Papillon’s volume, he would there have found that Papillon was indulging in no “little pleasantry of expression,” but was seriously relating a melancholy fact of two brother artists losing their senses about the same time as himself; and had he ever read the supplement, or third volume, of Papillon’s work, he would have seen, at p. 39, the account which Papillon himself gives of his own insanity.

Having disposed of the story as told by Papillon, it remains now to notice “the learning and deep research” with which it has been supported by Zani, and some of the arguments which have been alleged in its favour by Mr. Ottley.

In the first place, Zani has discovered that a family of the name of Cunio, in which the name of Alberico more than once occurs, actually resided in the neighbourhood of Ravenna at the very period mentioned in the title-page to the cuts by the Cunio, and in the history written in old Swiss characters. Upon this, and other similar pieces of evidence, Mr. Ottley remarks as follows: “Now both these cities [Ravenna and Imola] are in the vicinity of Faenza, where the family, or a branch of it, is spoken of by writers of undoubted credit in the twelfth, the thirteenth, and the fourteenth centuries. These circumstances, therefore, far from furnishing any just motive of additional doubt, form together such a phalanx of corroborative evidence in support of the story, as, in my opinion, those who would impeach the truth of Papillon’s statement can never break through.” “Argal,” Rowley’s poems are genuine, because such a person as “Maistre William Canynge” lived at Bristol at the period when he is mentioned by the pseudo Rowley. Zani, however, unfortunately for his own argument, let us know that the names and residence of the family of the Cunio might be obtained from “Tonduzzi’s History of Faenza,” printed in 1675. Whether this book appeared in French, or not, previous to the publication of Papillon’s works, I have not been able to learn; but a Swiss captain, who could read “old Gothic Italian,” would certainly find little difficulty in picking a couple of names out of a modern Italian volume.

The reasoning faculties of Signor Zani appear to have been very imperfectly developed, for he cites the following as a case in point; and 37 Mr. Ottley, who gives it in his text, seems to concur in its applicability. He is noticing the objections which have been made to Papillon’s account, on the ground of no previous author mentioning the existence of such a work, and that no person subsequently had ever seen a copy. Zani’s argument, as given by Mr. Ottley,I.44 is as follows: “He, however, who should reason in this manner, might, upon the same grounds, deny the loss of many manuscripts, and even of printed books, which, according to the testimony of credible authors, have become a prey to the flames, or have perished during the anarchy of revolutions, or the distresses occasioned by wars. The learned part of my readers will not require examples. Nevertheless, let him who wants such conviction search throughout all the libraries of Europe for the work entitled ‘Meditationes Reverendissimi patris Domini Johannis de Turre-cremata,’ printed at Rome by Ulrich Hahn, in 1467, and he will presently be informed by the learned librarians, that of that edition there exists but one copy, which is preserved in the library of Nuremberg. This book is, therefore, unique.I.45 Now let us suppose that, by some accident, this book should perish; could our descendants on that account deny that it ever had existed?” And this is a corroborative argument in support of the truth of Papillon’s tale! The comment, however, is worthy of the text. It is to be observed that Ulrich Hahn’s edition of Turre-cremata appeared ten years after Faust and Scheffer’s Psalter, of the date 1457, was printed; and that the existence of several hundred volumes printed before 1467 proves that the art of printing was then practised to a considerable extent. That Ulrich Hahn was a printer at Rome in 1468 and subsequent years is proved by many copies of works which proceeded from his press; and the existence of the identical “unique” copy, referred to by Zani, is vouched for by upwards of fifty learned men who have seen it; and, what is more, mentioned the place where it was preserved, so that, if a person were sceptical, he might satisfy himself by the evidence of his own senses. But who, except Papillon, has ever seen the engravings of the Cunio, executed upwards of a hundred and thirty years prior to the earliest authentic specimen of the art, and who has ever mentioned the place where they were to be seen? Had any person of equal credibility with Papillon described a volume printed at Rome in 1285, the date of the pretended wood-cuts of the Cunio, the case would then have been in point, and the decision of every person in the slightest degree acquainted with the subject, and not rendered blind to simple truth by the vivid brightness of his own speculations, would be 38 inevitably the same; that is, the evidence in both cases would not be relied on.

“It is possible,” says Zani, “that at this moment I may be blinded by partiality to my own nation; but I would almost assert, that to deny the testimony of the French writer, would be like denying the existence of light on a fine sun-shiny day.” His mental optics must have been of a peculiar character, and it can be no longer doubtful that he

“Had lights where better eyes are blind,

As pigs are said to see the wind.”

Mr. Ottley’s own arguments in support of Papillon’s story are scarcely of a higher character than those which he has adopted from Zani. At page 40, in answer to an objection founded on the silence of all authorities, not merely respecting the particular work of the Cunio, but of the frequent practice of such an art, and the fact of no contemporary specimens being known, he writes as follows: “We cannot safely argue from the silence of contemporaneous authorities, that the art of engraving on wood was not practised in Europe in those early times; however, such silence may be an argument that it was not an art in high repute. Nor is our ignorance of such records a sufficient proof of their non-existence.” The proof of such a negative would be certainly difficult; but, according to this mode of argument, there is no modern invention which might not also be mentioned in “certain ancient undiscovered records.” In the general business of life, that rule of evidence is a good one which declares “de non-apparentibus et non-existentibus eadem est ratio;” and until it shall be a maxim in logic that “we ought readily to believe that to be true which we cannot prove to have been impossible,” Mr. Ottley’s solution of the difficulty does not seem likely to obtain general credence.

At page 41, speaking of the probability of wood-engraving, for the purpose of taking impressions, being practised at an earlier period than has been generally supposed, Mr. Ottley expresses himself as follows: “Nor is it any proof or strong argument against the antiquity of such a practice, that authentic specimens of wood-engraving of those early times are not now to be found. They were, it may be supposed, for the most part, detached pieces, whose merits, as works of art, were not such as to render their preservation at all probable. They were the toys of the day; and, after having served the temporary purpose for which they were manufactured, were, no doubt, swept away to make room for others of newer fashion.” He thus requires those who entertain an opinion contrary to his own to prove a negative; while he assumes the point in dispute as most clearly established in his own favour.

If such wood engravings—“the toys of the day”—had been known 39 in the thirteenth, or even the fourteenth century, is it not likely that some mention would be made of them in the writings of some one of the minstrels of the period to whom we are indebted for so many minute particulars illustrative of the state of society at the period referred to? Not the slightest allusion to anything of the kind has hitherto been noticed in their writings. Respecting such “toys” Boccaccio is silent, and our countryman Chaucer says not a word. Of wood-cuts not the least mention is made in Petrarch; and Richard de Bury, bishop of Durham, who lived in the reign of Edward III., in his curious Essay on the Love of Books, says not a syllable of wood-cuts, either as toys, or as illustrations of devotional or historical subjects. Upon this question, affirmed by Papillon, and maintained as true by Zani and Ottley, contemporary authorities are silent; and not one solitary fact bearing distinctly upon the point has been alleged in support of Papillon’s narrative.

  see text

I.1 C. G. Von Murr, in his Journal zur Kunstgeschichte, 2 Theil, S. 253, referring to Martorelli, De Regia Theca Calamaria.

I.2 If this etymology be correct, the English Scrivener and French Greffier may be related by descent as well as professionally; both words being thus referable to the same origin, the Greek γράφω. The modern Writer in the Scottish courts of law performs the duties both of Scrivener and Greffier, with whose name his own is synonymous.

I.3 Towards the close of the seventeenth century we find books “adorned with sculptures by a curious hand;” about 1730 we find them “ornamented with cuts;” at present they are “illustrated with engravings.”

I.4 Astle on the Origin and Progress of Writing, p. 215, 2nd edit.

I.5 Author of “An Account of the Manners and Customs of the Modern Egyptians, written in Egypt during the years 1833, ’34, and ’35.”

I.6 On a mummy in the royal collection at Paris, the six first characters of this stamp occur. Champollion reads them, “Amenoftep,” or “Amonaftep.” He supposes the name to be that of Amonoph the First; and says that it signifies “approuvé par Ammon.”—Précis du Système Hiéroglyphique. Planches et Explication, p. 20, No. 161.

I.7 Inscriptionum Explicatio, fol. Romæ, 1699.

I.8 “O nata mecum consule Manlio!” says Horace, addressing an amphora of wine as old as himself; and Petronius mentions some choice Falernian which had attained the ripe age of a hundred: “Statim allatæ sunt amphoræ vitreæ diligenter gypsatæ, quarum in cervicibus pittacia erant affixa, cum hoc titulo: Falernum Opimianum annorum centum.” Pittacia were small labels—schedulæ breves—attached to the necks of wine-vessels, and on which were marked the name and age of the wine.

I.9 Journal zur Kunstgeschichte, 2 Theil, S. 81. By grotesque—“Laubwerk”—ornamental foliage is here meant;—grot-esque, bower-work,—not caricatures.

I.10 M. Dulaure’s latinity is bad. “Lippas” certainly is not the word. His translation is, “Remède anodin de Quintus Junius Tauridus, pour tous les maux d’yeux.” Other stone stamps, supposed to have been used by oculists to mark the vessels containing their medicaments, were discovered and explained long before M. Dulaure published his interpretation. See “Walchii Antiquitates Medicæ Selectæ, Jenæ, 1772,” Num. 1 and 2, referred to by Von Murr.

I.11 Hermannus Hugo, De prima Origine Scribendi, cap. xix. De Notis Servilibus, et cap. xx. De Notis pecudum. A further account of the ancient stigmata, and of the manner in which slaves were marked, is to be found in Pignorius, De Servis.

I.12 History of the Poor Laws, 8vo. 1764, by Richard Burn, LL.D., who in his observations on such punishments says: “It is affecting to humanity to observe the various methods that have been invented for the punishment of vagrants; none of all which wrought the desired effect . . . . . . This part of our history looks like the history of the savages in America. Almost all severities have been exercised against vagrants, except scalping.”

I.13 “Quum puer jam ductus sequi cœperit, non inutile erit, litteras tabellæ quam optime insculpi, ut per illos, velut sulcos, ducatur stylus. Nam neque errabit, quemadmodum in ceris, continebitur enim utrimque marginibus, neque extra præscriptum poterit egredi; et celerius ac sæpius sequendo certa vestigia firmabit articulos, neque egebit adjutorio manum suam, manu superimposita, regentis.” Quintiliani Instit. Orator., lib. i. cap. I.

I.14 Prosper Marchand, at page 9 of his “Histoire de l’Imprimerie,” gives the following title of a book in 8vo. which was wholly, both text and figures, executed in this manner, percé au jour, in vellum: “Liber Passionis Domini Nostri Jesu Christi, cum figuris et characteribus ex nulla materia compositis.” He states that in 1640 it was in the collection of Albert Henry, Prince de Ligne, and quotes a description of it from Anton. Sanderi Bibliotheca Belgica Manuscripta, parte ii. p. 1.

I.15 “Rex Theodoricus inliteratus erat, et sic obruto sensu ut in decem annos regni sui quatuor literas subscriptionis edicti sui discere nullatenus potuisset. De qua re laminam auream jussit interrasilem tieri quatuor literas regis habentem, unde ut si subscribere voluisset, posita lamina super chartam, per eam pennam duceret et subscriptio ejus tantum videretur.”—Vita Theodorici Regis Ostrogothorum et Italiæ, autore Joanne Cochlæo; cum additamentis Joannis Peringskiold, 4to. Stockholmiæ, 1699, p. 199.

I.16 A monogram, properly, consists of all, or the principal letters of a name, combined in such a manner that the whole appear but as one character; a portion of one letter being understood to represent another, two being united to form a third, and so on.

I.17 Mabillon’s opinion is founded on the following passage in the Life of Charlemagne, by his secretary Eginhard: “Ut scilicet imperitiam hanc [scribendi] honesto ritu suppleret, monogrammatis usum loco proprii signi invexit.”

I.18 “Triplex cruces exarandi modus: 1. penna sive calamo; 2. lamina interrasili; 3. stampilla sive typo anaglyptico. Laminæ interrasiles ex auro aliove metallo, vel ex ebore etiam confectæ sunt, atque ita perforatæ, ut hiatus, pro re nata, crucium cet. speciem præ se ferrent, per quos velut sulcos, calamus sive penna ducebatur. Stampillæ vero ita sculptæ sunt, ut figuræ superficiem eminerent, quæ deinde atramento tinctæ sunt, chartæque impressæ.”—Gatterer, Elementa Artis Diplomaticæ, § 264, De Staurologia.

I.19 No. lxi. p. 108, where the preceding Gothic marks, with the explanation of them, are given.

I.20 Essay on Medals, pp. 144, 145. Edit. 1784.

I.21 It it given by Gatterer in his “Elementa Artis Diplomaticæ,” p. 166; [4to. Gottingæ, 1765;] who refers to Muratori, Antiquit. Italiæ Medii Ævi, t. vi. p. 9.

I.22 These stamps are copied from “D. E. Baringii Clavis Diplomatica,” 4to. Hanoveræ, 1754. There is a work expressly treating of the use of the Diplomatic Stamp—J. C. C. Oelrichs de Stampilla Diplomatica, folio, Wismariæ, 1762, which I have not been able to obtain a sight of.

I.23 The marks here given are copied from Mackarel’s History of King’s Lynn, 8vo. 1737. In the same book there are upwards of thirty more of a similar kind, from the middle of the fourteenth century to the latter end of the seventeenth. Perhaps no two counties in the kingdom afford so many examples of merchants’-marks and monumental brasses as Norfolk and Suffolk.

I.24Y-meddled is mixed; the marks of merchants are put in opposition to the ‘shapen shields,’ because merchants had no coats of arms.”—Specimens of the Early English Poets, by George Ellis, Esq. vol. i. p. 163. Edit. 1811.

I.25 “Till lately this was the earliest dated evidence of block printing known; but there has just been discovered at Malines, and now deposited at Brussels, a woodcut of similar character, but assumed to be Dutch or Flemish, dated MCCCCXVIII.; and though there seems no reason to doubt the genuineness of the cut, it is currently asserted that the date bears evidence of having been tampered with.”—Extract from Bohn’s Lecture on Printing.

I.26 The woodcut referred to is that of St. Christopher, discovered by Heineken, pasted within the cover of a book in the Monastery of Buxheim, near Memmingen, in Suabia. It is of a folio size, and is coloured by means of stencils; a practice which appears to have been adopted at an early part of the fifteenth century by the German Formschneiders and Briefmalers, literally, figure-cutters and cardpainters, to colour their cuts and their cards. The St. Christopher is now in Earl Spencer’s library. (See a reduced copy of it at p. 46).

I.27 The small and thick brass coins, struck by Grecian cities under the Roman emperors, and known to collectors as “colonial Greek,” appear to have been cast, and moulds for such a purpose have been discovered in our own country.

I.28 “That a strong passion for portraits formerly existed, is attested both by Atticus, the friend of Cicero, who wrote a work on this subject, and by M. Varro, who conceived the very liberal idea of inserting by some means or other, in his numerous volumes, the portraits of seven hundred individuals; as he could not bear the idea that all traces of their features should be lost, or that the lapse of centuries should get the better of mankind.”—Pliny’s Natural History, Book XXXV. chap. 2.—(Bohn’s Ed. vol. vi. p. 226. M. Deville is of opinion that these portraits were made in relief upon plates of metal, perhaps bronze, and coloured with minium, a red tint much esteemed by the Romans).

I.29 See De Pauw, Recherches Philosophiques sur les Grecs, t. ii. p. 100. The subject is discussed in Meusel’s “Neue Miscellaneen von artistischen Inhalts,” part xii. p. 380-387, in an article, “Sind wirklich die Römer die Erfinder der Kupferstecherkunst?—Were the Romans truly the inventors of copper-plate engraving?”—by A. Rode. Böttiger, one of the most learned and intelligent of all German writers on the fine arts, and Fea, the editor of Winkleman’s History of Art, do not admit De Pauw’s conjecture, but decide the question in the negative.

I.30 An excellent representation of this celebrated monument is given in Cotman’s “Engravings from the most remarkable Sepulchral Brasses in Norfolk,” folio, 1819 (republished with considerable additions in 2 vols. folio, 1839).

I.31 At page 7, Mr. Ottley, borrowing from Du Halde, has erroneously stated that the delicate nature of their paper would not permit the use of a press. He must have forgot, for he cannot but have known, that impressions on the finest India paper had been frequently taken from wood-blocks by means of the common printing-press many years previous to 1816, the date of the publication of his book. I have never seen Chinese paper that would bear printing by hand, which would not also bear the action of the press, if printed without being wet in the same manner as common paper.

I.32 It would appear that Chinese annalists themselves were not agreed as to the period when printing by the hand from wood-blocks was first practised in that country. “Nicholas Trigaltius, a member of our order,” writes Herman Hugo, “who has recently returned from China, gives the following information respecting printing, which he professes to have carefully extracted from the annals of the Chinese themselves. ‘Typography is of somewhat earlier date in China than in Europe, for it is certain that it was practised in that country about five centuries ago. Others assert that it was practised in China at a period prior to the Christian era.’”—Hermannus Hugo, De Prima Origine Scribendi, p. 211. Antwerpiæ, 1617.

I.33 The pretensions of the Chinese to excellence in science are ably exposed by the learned Abbé Renaudot in a disquisition “Sur les sciences des Chinois,” appended to his translation, from the Arabic, entitled “Anciennes Relations des Indes et de la Chine, de deux Voyageurs Mahométans, qui y allèrent dans le neuvième siècle.”—8vo. Paris, 1718.

I.34 See the Travels of Marco Polo. (In Bohn’s Antiq. Library).

I.35 It has been conjectured that the following passages in the travels of Marco Polo might suggest the idea of block-printing, and consequently wood engraving: “Gradatim reliquos belli duces in digniorem ponit statum, donatque illis aurea et argentea vasa, tabulas, privilegia atque immunitatem. Et hæc quidem privilegia tabulis vel bracteis per sculpturas imprimuntur.” “Moneta magni Cham non fit de auro vel argento, aut alio metallo, sed corticem accipiunt medium ab arbore mori, et hunc consolidant, atque in particular varias et rotundas, magnas et parvas, scindunt, atque regale imprimunt signum.”—M. Pauli Veneti Itiner. lib. ii. capp. vii. & xxi. The mention of paper money impressed with the royal stamp also occurs in the Eastern History of Haython, an Armenian, whose work was written in 1307, in Latin, and has been printed several times, of which the last edition is by And. Müller, Colon. 1671, 4to.

I.36 An article on Marsden’s “Translation of the Travels of Marco Polo,” in the Quarterly Review, No. xli. May, 1819, from p. 191 to 195, contains some curious particulars respecting the early use of the mariner’s compass, and of gunpowder and cannon in Europe.

I.37 A Monsieur Spirchtvel, as Papillon informs us. Tom. i. p. 92.

I.38 Cartouch. “This word is used to denote those fantastic ornaments which were formerly introduced in decorating the wainscots of rooms; and frequently served the purpose of frames, surrounding inscriptions, small paintings, or other devices. These cartouches were much in vogue in the sixteenth and seventeenth centuries for the frontispieces of books of prints; and indeed Callot and Della Bella etched many entire sets of small subjects surrounded by similar ornaments. From the irregularity of their forms, the terms tablet shield, or panel, would be but ill expressive of their character.”—Ottley’s Inquiry, vol. i. p. 12.

I.39 Readers of French romances will find the tale of the Cunio at p. 89, tom. i. of Papillon’s “Traité de la Gravure en Bois,” or at p. 17, vol. i. of Mr. Ottley’s “History of Engraving.”

I.40 Of Le Brun’s five subjects illustrative of the actions of Alexander the Great, four of them are precisely the same as four of those said to be executed by the Cunio: 1. Alexander passing the Granicus; 2. the battle of Arbela; 3. the reception of Porus by Alexander; 4. Alexander’s triumphant entry into Babylon. There certainly has been some copying here; but it is more likely that Papillon or his informant had seen Le Brun’s paintings, than that Le Brun had seen the original wood engravings executed by the Cunio.

I.41 From the age of sixteen, cruel and secret annoyances interrupted his studies; shortly after his marriage, in 1723, his absent manner was a source of uneasiness to his wife; and in 1759 he fairly lost his senses. See Papillon, Traité de la Gravure en Bois, 8vo. 1766, Preface, p. xi.; & p. 335, tom. i. et Supplement, p. 39.

I.42 It is worthy of remark that Papillon, when questioned by Heineken, who called on him in Paris after the publication of his work, respecting the account of the Cunio, did not produce his three sheets of original memoranda. He might thus have afforded a proof of his own good faith, by producing the manuscript written by him in 1720 from the dictation of Captain de Greder.

I.43 Inquiry into the Early History of Engraving, vol. i. p. 23.

I.44 History of Engraving, vol. i. p. 28.

I.45 Three copies of this supposed unique book have long been known to bibliographers; one in the public library of Nuremberg, another in the Imperial library of Vienna, and the third in Lord Spenser’s library.

Errors in Chapter I

the loitering cask, (that bears its date) from
date, from
in the same passage, “Lyde” for expected “Lydus” is in Smart

and even allowing him to be sincere
eve nallowing

which have been alleged in its favour by Mr. Ottley.
Mr Ottley.

“It is possible,” says Zani,
say

Footnote I.39

the tale of the Cunio at p. 89, tom. i.
tom i.

40

CHAPTER II.
PROGRESS OF WOOD ENGRAVING.

Playing-cards printed from wood-blocks—early german wood-engravers at augsburg, nuremberg, and ulm—card-makers and wood-engravers in venice in 1441—figures of saints engraved on wood—the st. christopher, the annunciation, and the st. bridget in the collection of earl spencer, with other old wood-cuts described—block-books—the apocalypse, the history of the virgin, and the work called biblia pauperum—speculum salvationis—figured alphabet formerly belonging to sir george beaumont—ars memorandi, and other smaller block-books.

From the facts which have been produced in the preceding chapter, there cannot be a doubt that the principle on which wood engraving is founded,—that of taking impressions on paper or parchment, with ink, from prominent lines,—was known and practised in attesting documents in the thirteenth and fourteenth centuries. Towards the end of the fourteenth, or about the beginning of the fifteenth century, there is reason to believe that this principle was adopted by the German card-makers for the purpose of marking the outlines of the figures on their cards, which they afterwards coloured by means of a stencil.II.1

The period at which the game of cards was first known in Europe, as well as the people by whom they were invented, has been very learnedly, though not very satisfactorily discussed. Bullet has claimed the invention for the French, and Heineken for the Germans; while other writers have maintained that the game was known in Italy earlier than in any other part of Europe, and that it was introduced from the East.

From a passage discovered by M. Van Praet, in an old manuscript copy of the romance of Renard le Contrefait, it appears that cards were known in France about 1340, although Bullet was of opinion that they 41 were invented in that country about 1376. At whatever period the game was introduced, it appears to have been commonly known in France and Spain towards the latter part of the fourteenth century. John I., King of Castile, by an edict issued in 1387, prohibited the game of cards; and in 1397, the Provost of Paris, by an ordonnance, forbid all working people to play at tennis, bowls, dice, cards, or nine-pins, on working days. From a passage in the Chronicle of Petit-Jehan de Saintré, written previous to 1380, it would appear that the game of cards at that period was in disrepute. Saintré had been one of the pages of Charles V. of France; and on his being appointed, on account of his good conduct, to the situation of carver to the king, the squire who had charge of the pages, lectured some of them on the impropriety of their behaviour; such as playing at dice and cards, keeping bad company, and haunting taverns and cabarets, those not being the courses by which they might hope to arrive at the honourable post of “ecuyer tranchant,” to which their companion, Saintré, had been raised.

In an account-book of Charles Poupart, treasurer to Charles VI. of France, there is an entry, made about 1393, of “fifty-six sols of Paris, given to Jacquemin Gringonneur, painter, for three packs of cards, gilt and coloured, and of different sorts, for the diversion of his majesty.” From this passage the learned Jesuit Menestrier, who was not aware of cards being mentioned by any earlier writer, concluded that they were then invented by Gringonneur to amuse the king, who, in consequence of a coup de soleil, had been attacked with delirium, which had subsided into an almost continual depression of spirits. There, however, can be no doubt that cards were known in France at least fifty years before; though, from their being so seldom noticed previous to 1380, it appears likely that the game was but little played until after that period. Whether the figures on the cards supplied for the king’s amusement were drawn and coloured by the hand, or whether the outlines were impressed from wood-blocks, and coloured by means of a stencil, it is impossible to ascertain; though it has been conjectured that, from the smallness of the sum paid for them, they were of the latter description. That cards were cheap in 1397, however they might be manufactured, may be presumed from the fact of their being then in the hands of the working people.

To whatever nation the invention of cards is owing, it appears that the Germans were the first who practised card-making as a trade. In 1418 the name of a “Kartenmacher”—card-maker—occurs in the burgess-book of the city of Augsburg; and in an old rate-book of the city of Nuremburg, under the year 1433, we find “Ell. Kartenmacherin;” that is, Ell.—probably for Elizabeth—the card-maker. In the same book, under the year 1435, the name of “Eliz. Kartenmacherin,” probably 42 the same person, is to be found; and in 1438 there occurs the name “Margret Kartenmalerin”—Margaret the card-painter. It thus appears that the earliest card-makers who are mentioned as living at Nuremberg were females; and it is worthy of note that the Germans seem to have called cards “Karten” before they gave them the name of “Briefe.” Heineken, however, considers that they were first known in Germany by the latter name; for, as he claimed the invention for his countrymen, he was unwilling to admit that the name should be borrowed either from Italy or France. He has not, however, produced anything like proof in support of his opinion, which is contradicted by the negative evidence of history.II.2

The name Briefe, which the Germans give to cards, also signifies letters [epistolæ]. The meaning of the word, however, is rather more general than the French term lettres, or the Latin epistolæ which he gives as its synonyms, for it is also applied in the sense in which we sometimes use the word “paper.” For instance, “ein Brief Stecknadeln, ein Brief Tabak,” are literally translated by the words “a paper of pins, a paper of tobacco;” in which sense the word “Brief” would, in Latin, be more correctly rendered by the term charta than epistola. As it is in a similar sense—cognate with “paper,” as used in the two preceding examples—that “Briefe” is applied to cards, I am inclined to consider it as a translation of the Latin chartaæ, the Italian carte, or the French cartes, and hence to conclude that the invention of cards does not belong to the people of Germany, who appear to have received cards, both “name and thing,” from another nation, and after some time to have given them a name in their own language.

In the town-books of Nuremberg, the term Formschneider—figure-cutter,—the name appropriated to engravers on wood, first occurs in 1449;II.3 and as it is found in subsequent years mentioned in the same page with “Kartenmaler,” it seems reasonable to conclude that in 1449, and probably earlier, the business of the wood-engraver proper, and that of the card-maker, were distinct. The primary meaning of the word form or forma is almost precisely the same in most of the European languages. 43 It has erroneously been explained, in its relation to wood engraving, as signifying a mould, whereas it simply means a shape or figure. The model of wood which the carpenter makes for the metal-founder is properly a form, and from it the latter prepares his mould in the sand. The word form, however, in course of time declined from its primary signification, and came to be used as expressive both of a model and a mould. The term Formschneider, which was originally used to distinguish the professed engraver of figures from the mere engraver and colourer of cards, is still used in Germany to denote what we term a wood-engraver.

About the time that the term Formschneider first occurs we find Briefmalers mentioned, and at a later period Briefdruckers—card-printers; and, though there evidently was a distinction between the two professions, yet we find that between 1470 and 1500 the Briefmalers not only engraved figures occasionally, but also printed books. The Formschneiders and the Briefmalers, however, continued to form but one guild or fellowship till long after the art of wood-engraving had made rapid strides towards perfection, under the superintendence of such masters as Durer, Burgmair, and Holbein, in the same manner as the barbers and surgeons in our own country continued to form but one company, though the “chirurgeon had long ceased to trim beards and cut hair, and the barber had given up bleeding and purging to devote himself more exclusively to the ornamental branch of his original profession.” “Kartenmacher and Kartenmaler” says Von Murr, “or Briefmaler, as they were afterwards called [1473], were known in Germany eighty years previous to the invention of book-printing. The Kartenmacher was originally a Formschneider, though, after the practice of cutting figures of saints and of sacred subjects was introduced, a distinction began to be established between the two professions.”

The German card-makers of Augsburg, Nuremberg, and Ulm, it is stated, sent large quantities of cards into Italy; and it was probably against those foreign manufacturers that the fellowship of painters at Venice obtained an order in 1441 from the magistracy, declaring that no foreign manufactured cards, or printed coloured figures, should be brought into the city, under the penalty of forfeiting such articles, and of being fined xxx liv. xii soldi. This order was made in consequence of a petition presented by the Venetian painters, wherein they set forth that “the art and mystery of card-making and of printing figures, which were practised in Venice, had fallen into total decay through the great quantity of foreign playing-cards and coloured printed figures, which were brought into the city.”II.4 It is hence evident that the art both of the German 44 Kartenmacher and of the Formschneider was practised in Venice in 1441; and, as it is then mentioned as being in decay, it no doubt was practised there some time previously.

Heineken, in his “Neue Nachrichten,” gives an extract from a MS. chronicle of the city of Ulm, completed in 1474, to the following effect: “Playing-cards were sent barrelwise [that is, in small casks] into Italy, Sicily, and also over sea, and exchanged for spices and other wares. From this we may judge of the number of card-makers who resided here.” The preceding passage occurs in the index, under the head, “Business of card-making.” Heineken also gives the passage in his “Idée Générale,” p. 245; but from the French translation, which he there gives, it appears that he had misunderstood the word “leglenweiss”—barrelwise—which he renders “en ballots.” In his “Neue Nachrichten,” however, he inserts the explanation between parentheses, (“das ist, in kleinen Fässern”)—i. e. in small casks; which Mr. Singer renders “hogsheads,” and Mr. Ottley, though he gives the original in a note, “large bales.” The word “lägel,” a barrel, is obsolete in Germany, but its diminutive, “leglin,”—as if “lägelen”—is still used in Scotland for the name of the ewe-milker’s kit.

Some writers have been of opinion that the art of wood-engraving was derived from the practice of the ancient caligraphists and illuminators of manuscripts, who sometimes formed their large capital letters by means of a stencil or of a wooden stamp. That large capitals were formed in such a manner previous to the year 1400 there can be little doubt; and it has been thought that stencils and stamps were used not only for the formation of capital letters, but also for the impression of a whole volume. Ihre, in a dissertation on the Gospels of Ulphilas,II.5 which are supposed to be as old as the fifth century, has asserted that the silver letters of the text on a purple ground were impressed by means of heated iron stamps. This, however, is denied by the learned compilers of the “Nouveau Traité de Diplomatique,” who had seen other volumes of a similar kind, the silver letters of which were evidently formed with a pen. A modern Italian author, D. Vincenzo Requeno, has published a tractII.6 to prove that many supposed manuscripts from the tenth to the fourteenth century, instead of being written with a pen, were actually impressed by means of stamps. It is, however, extremely 45 probable that he is mistaken; for if his pretended discoveries were true, this art of stamping must have been very generally practised; and if so, it surely would have been mentioned by some contemporary writers. Signor Requeno’s examination, I am inclined to suspect, has not been sufficiently precise; for he seems to have been too willing to find what he sought. In almost every collection that he examined, a pair of fine compasses being the test which he employed, he discovered voluminous works on vellum, hitherto supposed to be manuscript, but which according to his measurement were certainly executed by means of a stamp.

It has been conjectured that the art of wood-engraving was employed on sacred subjects, such as the figures of saints and holy persons, before it was applied to the multiplication of those “books of Satan,” playing-cards. It however is not unlikely that it was first employed in the manufacture of cards; and that the monks, availing themselves of the same principle, shortly afterwards employed the art of wood-engraving for the purpose of circulating the figures of saints; thus endeavouring to supply a remedy for the evil, and extracting from the serpent a cure for his bite.

Wood-cuts of sacred subjects were known to the common people of Suabia, and the adjacent districts, by the name of Helgen or Helglein, a corruption of Heiligen, saints;—a word which in course of time they used to signify prints—estampes—generally.II.7 In France the same kind of cuts, probably stencil-coloured, were called “dominos,”—the affinity of which name with the German Helgen is obvious. The word “domino” was subsequently used as a name for coloured or marbled paper generally, and the makers of such paper, as well as the engravers and colourers of wood-cuts, were called “dominotiers.”II.8

As might, à priori, be concluded, supposing the Germans to have been the first who applied wood-engraving to card-making, the earliest wood-cuts have been discovered, and in the greatest abundance, in that district where we first hear of the business of a card-maker and a wood-engraver. From a convent, situated within fifty miles of the city of Augsberg, where, in 1418, the first mention of a Kartenmacher occurs, has been obtained the earliest wood-cut known,—the St. Christopher, now in the possession of Earl Spencer, with the date 1423. That this was the first cut of the kind we have no reason to suppose; but though others executed in a similar manner are known, to not one of them, upon anything like probable grounds, can a higher degree of antiquity be 46 assigned. From 1423, therefore, as from a known epoch, the practice of wood engraving, as applied to pictorial representations, may be dated.

  see text

The first person who published an account of this most interesting wood-cut was Heineken, who had inspected a greater number of old wood-cuts and block-books than any other person, and whose unwearied perseverance in searching after, and general accuracy in describing such early specimens of the art of wood-engraving, are beyond all praise. He found it pasted on the inside of the right-hand cover of a manuscript volume in the library of the convent of Buxheim, near Memmingen in Suabia. The manuscript, entitled Laus VirginisII.9 and finished in 1417, 47 was left to the convent by Anna, canoness of Buchaw, who was living in 1427; but who probably died previous to 1435. The above reduced copy conveys a pretty good idea of the composition and style of engraving of the original cut, which is of a folio size, being eleven and a quarter inches high, and eight inches and one-eighth wide.II.10

The original affords a specimen of the combined talents of the Formschneider or wood-engraver, and the Briefmaler or card-colourer. The engraved portions, such as are here represented, have been taken off in dark colouring matter similar to printers’ ink, after which the impression appears to have been coloured by means of a stencil. As the back of the cut cannot be seen, in consequence of its being pasted on the cover of the volume, it cannot be ascertained with any degree of certainty whether the impression has been taken by means of a press, or rubbed off from the block by means of a burnisher or rubber, in a manner similar to that in which wood-engravers of the present day take their proofs.

This cut is much better designed than the generality of those which we find in books typographically executed from 1462, the date of the Bamberg Fables, to 1493, when the often-cited Nuremberg Chronicle was printed. Amongst the many coarse cuts which “illustrate” the latter, and which are announced in the book itselfII.11 as having been “got up” under the superintendence of Michael Wolgemuth, Albert Durer’s master, and William Pleydenwurff, both “most skilful in the art of painting,” I cannot find a single subject which either for spirit or feeling can be compared to the St. Christopher. In fact, the figure of the saint, and that of the youthful Christ whom he bears on his shoulders, are, with the exception of the extremities, designed in such a style, that they would scarcely discredit Albert Durer himself.

To the left of the engraving the artist has introduced, with a noble disregard of perspective,II.12 what Bewick would have called a “bit of Nature.” In the foreground a figure is seen driving an ass loaded with 48 a sack towards a water-mill; while by a steep path a figure, perhaps intended for the miller, is seen carrying a full sack from the back-door of the mill towards a cottage. To the right is seen a hermit—known by the bell over the entrance of his dwelling—holding a large lantern to direct St. Christopher as he crosses the stream. The two verses at the foot of the cut,

Cristofori faciem die quacunque tueris,

Illa nempe die morte mala non morieris,

may be translated as follows:

Each day that thou the likeness of St. Christopher shalt see,

That day no frightful form of death shall make an end of thee.

They allude to a popular superstition, common at that period in all Catholic countries, which induced people to believe that the day on which they should see a figure or image of St. Christopher, they should not meet with a violent death, nor die without confession.II.13 To this popular superstition Erasmus alludes in his “Praise of Folly;” and it is not unlikely, that to his faith in this article of belief, the squire, in Chaucer’s “Canterbury Tales,” wore

“A Christofre on his brest, of silver shene.”

The date “Millesimo cccco xxo tercio”—1423—which is seen at the right-hand corner, at the foot of the impression, most undoubtedly designates the year in which the engraving was made.

The engraving, though coarse, is executed in a bold and free manner; and the folds of the drapery are marked in a style which would do credit to a proficient. The whole subject, though expressed by means of few lines, is not executed in the very simplest style of art. In the draperies a diminution and a thickening of the lines, where necessary to the effect, may be observed; and the shades are indicated by means of parallel lines both perpendicular, oblique, and curved, as may be seen in the saint’s robe and mantle. In many of the wood-cuts executed between 1462 and 1500, the figures are expressed, and the drapery indicated, by simple lines of one undeviating degree of thickness, without the slightest attempt at shading by means of parallel lines running in a direction different to those marking the folds of the drapery or the outlines of the figure. If mere rudeness of design, and simplicity in the mode of execution, were to be considered as the sole tests of antiquity in wood-engravings, upwards of a hundred, positively known to have been executed between 1470 and 1500, might be produced as affording intrinsic evidence of their having been executed at a period antecedent to the date of the St. Christopher.

49

In the Royal Library at Paris there is an impression of St. Christopher with the youthful Christ, which was supposed to be a duplicate of that in the possession of Earl Spencer. On comparing them, however, “it was quite evident,” says Dr. Dibdin, “at the first glance, as M. Du Chesne admitted, that they were impressions taken from different blocks. The question therefore was, after a good deal of pertinacious argument on both sides—which of the two impressions was the more ancient? Undoubtedly it was that of Lord Spencer.” At first Dr. Dibdin thought that the French impression was a copy of Earl Spencer’s, and that it might be as old as the year 1460; but, from a note added in the second edition of his tour, he seems to have received a new light. He there says: “The reasons upon which this conclusion [that the French cut was a copy of a later date] was founded, are stated at length in the preceding edition of this work: since which, I very strongly incline to the supposition that the Paris impression is a proof—of one of the cheats of De Murr.”II.14

On the inside of the first cover or “board” of the Laus Virginis, the volume which contains the St. Christopher, there is also pasted a wood engraving of the Annunciation, of a similar size to the above-named cut, and impressed on the same kind of paper. As they are both worked off in the same kind of dark-coloured ink, and as they evidently have been coloured in the same manner, by means of a stencil, there can be little doubt of their being executed about the same time. From the left-hand corner of the Annunciation the figure of the Almighty has been torn out. The Holy Ghost, who appears descending from the Father upon the Virgin in the material form of a dove, could not well be torn out without greatly disfiguring the cut. An idea may be formed of the original from the following reduced copy.

  see text

Respecting these cuts, which in all probability were engraved by some one of the Formschneiders of Augsburg, Ulm, or Nuremberg,II.15 50 P. Krismer, who was librarian of the convent of Buxheim, and who showed the volume in which they are pasted to Heineken, writes to Von Murr to the following effect: “It will not be superfluous if I here point out a mark, by which, in my opinion, old wood engravings may with certainty be distinguished from those of a later period. It is this: In the oldest wood-cuts only do we perceive that the engraver [Formschneider] has frequently omitted certain parts, leaving them to be afterwards filled up by the card-colourer [Briefmaler]. In the St. Christopher there is no such deficiency, although there is in the other cut which is pasted on the inside of the fore covering of the same volume, and which, I doubt not, was executed at the same time as the former. It represents the salutation of the Virgin by the angel Gabriel, or, as it is also called, the Annunciation; and, from the omission of the colours, the upper part 51 of the body of the kneeling Virgin appears naked, except where it is covered with her mantle. Her inner dress had been left to be added by the pencil of the card-colourer. In another wood-cut of the same kind, representing St. Jerome doing penance before a small crucifix placed on a hill, we see with surprise that the saint, together with the instruments of penance, which are lying near him, and a whole forest beside, are suspended in the air without anything to support them, as the whole of the ground had been left to be inserted with the pencil. Nothing of this kind is to be seen in more recent wood-cuts, when the art had made greater progress. What the early wood-engravers could not readily effect with the graver, they performed with the pencil,—for the most part in a very coarse and careless manner,—as they were at the same time both wood-engravers and card-colourers.”II.16

Besides the St. Christopher and the Annunciation, there is another old wood-cut in the collection of Earl Spencer which appears to belong to the same period, and which has in all probability been engraved by a German artist, as all who can read the German inscription above the figure would reasonably infer. Before making any remarks on this engraving, I shall first lay before the reader a reduced copy.

The figure writing is that of St. Bridget of Sweden, who was born in 1302 and died in 1373. From the representation of the Virgin with the infant Christ in her arms we may suppose that the artist intended to show the pious widow writing an account of her visions or revelations, in which she was often favoured with the blessed Virgin’s appearance. The pilgrim’s hat, staff, and scrip may allude to her pilgrimage to Jerusalem, which she was induced to make in consequence of a vision. The letters S. P. Q. R. in a shield, are no doubt intended to denote the place, Rome, where she saw the vision, and where she died. The lion, the arms of Sweden, and the crown at her feet, are most likely intended to denote that she was a princess of the blood royal of that kingdom. The words above the figure of the saint are a brief invocation in the German language, “O Brigita bit Got für uns!” “O Bridget, pray to God for us!” At the foot of the desk at which St. Bridget is writing are the letters M. I. Chrs., an abbreviation probably of Mater Jesu Christi, or if German, Mutter Iesus Christus.II.17

  see text

From the appearance of the back of this cut, as if it had been rubbed 52 smooth with a burnisher or rubber, there can be little doubt of the impression having been taken by means of friction. The colouring matter of the engraving is much lighter than in the St. Christopher and the Annunciation, and is like distemper or water-colour; while that of the latter cuts appears, as has been already observed, more like printer’s ink. It is coarsely coloured, and apparently by the hand, unassisted with the stencil. The face and hands are of a flesh colour. Her gown, as well as the pilgrim’s hat and scrip, are of a dark grey; her veil, which she wears hoodwise, is partly black and partly white; and the wimple which she wears round her neck is also white. The bench and desk, the pilgrim’s staff, the letters S. P. Q. R., the lion, the crown, and the nimbus 53 surrounding the head of St. Bridget and that of the Virgin, are yellow. The ground is green, and the whole cut is surrounded with a border of a shining mulberry or lake colour.

Mr. Ottley, having at the very outset of his Inquiry adopted Papillon’s story of the Cunio, is compelled, for consistency’s sake, in the subsequent portion of his work, when speaking of early wood engravings such as the above, to consider them, not as the earliest known specimens of the art, but merely as wood engravings such as were produced upwards of a hundred and thirty years after the amiable and accomplished Cunio, a mere boy and a girl, had in Italy produced a set of wood engravings, one of which was so well composed that Le Brun might be suspected of having borrowed from it the design of one of his most complicated pictures. In his desire, in support of his theory, to refer the oldest wood-cuts to Italy, Mr. Ottley asks: “What if these two prints [the St. Christopher and the Annunciation] should prove to be, not the productions of Germany, but rather of Venice, or of some district of the territory then under the dominion of that republic?”

His principal reasons for the preceding conjecture, are the ancient use of the word stampide—“printed”—in the Venetian decree against the introduction of foreign playing-cards in 1441; and the resemblance which the Annunciation bears to the style of the early Italian schools. Now, with respect to the first of these reasons, it is founded on the assumption that both those impressions have been obtained by means of a press of some kind or other,—a fact which remains yet to be proved; for until the backs of both shall have been examined, and the mark of the burnisher or rubber found wanting, no person’s mere opinion, however confidently declared, can be decisive of the question. It also remains to be proved that the word stampide, which occurs in the Venetian decree, was employed there to signify “printed with a press.” For it is certain that the low Latin word stampare, with its cognates in the different languages of Europe, was used at that period to denote impression generally. But even supposing that “stampide” signifies “printed” in the modern acceptation of the word, and that the two impressions in question were obtained by means of a press; the argument in favour of their being Italian would gain nothing, unless we assume that the foreign printed cards and figures, which were forbid to be imported into Venice, were produced either within the territory of that state or in Italy; for the word stampide—“printed,” is applied to them as well as those manufactured within the city. Now we know that the German card-makers used to send great quantities of cards to Venice about the period when the decree was made, while we have no evidence of any Italian cities manufacturing cards for exportation in 1441; it is therefore most likely that if the Venetians were acquainted with the use 54 of the press in taking impressions from wood-blocks, the Germans were so too, and for these more probable reasons, admitting the cuts in question to have been printed by means of a press:—First, the fact of those wood-cuts being discovered in Germany in the very district where we first hear of wood-engravers; and secondly, that if the Venetian wood-engravers were acquainted with the use of the press in taking impressions while the Germans were not, it is very unlikely that the latter would be able to undersell the Venetians in their own city. Until something like a probable reason shall be given for supposing the cuts in question to be productions “of Venice, or some other district of the territory then under the dominion of that republic,” I shall continue to believe that they were executed in the district in which they were discovered, and which has supplied to the collections of amateurs so many old wood engravings of a similar kind. No wood engravings executed in Italy, are known of a date earlier than those contained in the “Meditationes Johannis de Turre-cremata,” printed at Rome 1467,—and printed, be it observed, by a German, Ulrick Hahn. The circular wood engravings in the British Museum,II.18 which Mr. Ottley says are indisputably Italian, and of the old dry taste of the fifteenth century, can scarcely be referred to an earlier period than 1500, and my own opinion is that they are not older than 1510. The manner in which they are engraved is that which we find prevalent in Italian wood-cuts executed between 1500 and 1520.

With respect to the resemblance which the Annunciation bears to the style of the early Italian school,—I beg to observe that it equally resembles many of the productions of contemporary “schools” of England and France, as displayed in many of the drawings contained in old illuminated manuscripts. It would be no difficult matter to point out in many old German engravings attitudes at least as graceful as the Virgin’s; and as to her drapery, which is said to be “wholly unlike the angular sharpness, the stiffness and the flutter of the ancient German school,” I beg to observe that those peculiarities are not of so frequent occurrence in the works of German artists, whether sculptors, painters, or wood-engravers, who lived before 1450, as in the works of those who lived after that period. Angular sharpness and flutter in the draperies are not so characteristic of early German art generally, as of German art towards the end of the fifteenth, and in the early part of the sixteenth century.

55

Even the St. Bridget, which he considers to be of a date not later than the close of the fourteenth century,II.19 Mr. Ottley, with a German inscription before his eyes, is inclined to give to an artist of the Low Countries; and he kindly directs the attention of Coster’s partisans to the shield of arms—probably intended for those of Sweden—at the right-hand corner of the cut. Meerman had discovered a seal, having in the centre a shield charged with a lion rampant—the bearing of the noble family of Brederode—a label of three points, and the mark of illegitimacy—a bend sinister, and surrounded by the inscription, “S[igillum] Lowrens Janssoen,” which with him was sufficient evidence of its being the identical seal of Laurence, the Coster or churchwarden of Harlem.II.20

We thus perceive on what grounds the right of Germany to three of the oldest wood-cuts known is questioned; and upon what traits of resemblance they are ascribed to Italy and the Low Countries. By adopting Mr. Ottley’s mode of reasoning, it might be shown with equal probability that a very considerable number of early wood engravings—whether printed in books or separately—hitherto believed to be German, were really executed in Italy.

An old wood engraving of the martyrdom of St. Sebastian, of a quarto size, with a short prayer underneath, and the date 1437, apparently from the same block, was preserved in the monastery of St. Blaze, in the Black Forest on the confines of Suabia;II.21 and another, with the date 1443 inserted in manuscript, was pasted in a volume belonging to the library of the monastery of Buxheim. The latter is thus described by Von Murr: “Through the kindness of the celebrated librarian, Krismer, whom I have so often mentioned, I am enabled to give an account of an illuminated wood-cut, which at the latest must have been engraved in 1443. It is pasted on the inside of the cover of a volume which contains ‘Nicolai DunkelspülII.22 Sermonum Partem Hyemalem.’ It is of quarto 56 size, being seven and a half inches high, and five and a quarter wide, and is inclosed within a border of a single line. It is much soiled, as we perceive in the figures on cards which have been impressed by means of a rubber. The style in which it is executed is like that of no other wood-cut which I have ever seen. The cut itself represents three different subjects, the upper part of it being divided into two compartments, each three inches square, and separated from each other by means of a broad perpendicular line. In that to the right is seen St. Dorothy sitting in a garden, with the youthful Christ presenting flowers to her, of which she has her lap full. Before her stands a small hand-basket,—also full of flowers,—such as the ladies of Franconia and Suabia were accustomed to carry in former times. In the left compartment is seen St. Alexius, lying at the foot of a flight of steps, upon which a man is standing and emptying the contents of a pot upon the saint.II.23 Between these compartments there appears in manuscript the date ‘anno d’ni 1443.’ Both the ink and the characters correspond with those of the volume. This date indicates the time when the writer had finished the book and got it bound, as is more clearly proved by a memorandum at the conclusion. In the year 1483, before it came into the possession of the monastery of Buxheim, it belonged to Brother Jacobus Matzenberger, of the order of the Holy Ghost, and curate of the church of the Virgin Mary in Memmingen. The whole of the lower part of the cut is occupied with Christ bearing his cross, at the moment that he meets with his mother, whom one of the executioners appears to be driving away. Simon of Cyrene is seen assisting Christ to carry the cross. The engraving is executed in a very coarse manner.”II.24

In the Royal Library at Paris there is an ancient wood-cut of St. Bernardin, who is represented on a terrace, the pavement of which consists of alternate squares of yellow, red, and green. In his right hand the saint holds something resembling the consecrated wafer or host, in the midst of which is inscribed the name of Christ; and in his left a kind of oblong casket, on which are the words “Vide, lege, dulce nomen.” Upon a scroll above the head of the saint is engraved the sentence, “Ihesus semper sit in ore meo,” and behind him, on a black label, is his name in yellow letters, “Sanct’ Bernard’.” The cut is surrounded by a border of foliage, with the emblems of the four Evangelists at the four corners, and 57 at the foot are the five following lines, with the date, impressed from prominent lines:—

O . splendor . pudicitie . zelator . paupertatis . a

mator. innocentie . cultor . virginitatis . lustra

cors . apientie . protector . veritatis . thro

num . fulgidum . eterne . majestatis . para

nobis . additum . divine . pietatis . amen. (1454)

This rare cut was communicated to Jansen by M. Vanpraet, the well-known bibliographer and keeper of the Royal Library.II.25

“Having visited in my last tour,” says Heineken, after describing the St. Christopher, “a great many convents in Franconia, Suabia, Bavaria, and in the Austrian states, I everywhere discovered in their libraries many of those kinds of figures, engraved on wood, and pasted either at the beginning or the end of old volumes of the fifteenth century. I have indeed obtained several of them. These facts, taken altogether, have confirmed me in my opinion that the next step of the engraver in wood, after playing-cards, was to engrave figures of saints, which, being distributed and lost among the laity, were in part preserved by the monks, who pasted them in the earliest printed books with which they furnished their libraries.”II.26

A great many wood-cuts of devotional subjects, of a period probably anterior to the invention of book-printing by Gutenberg, have been discovered in Germany. They are all executed in a rude style, and many of them are coloured. It is not unlikely that the most of these woodcuts were executed at the instance of the monks for distribution among the common people as helps to devotion; and that each monastery, which might thus avail itself of the aid of wood engraving in the work of piety, would cause to be engraved the figure of its patron saint. The practice, in fact, of distributing such figures at monasteries and shrines to those who visit them, is not yet extinct on the Continent. In Belgium it is still continued, and, I believe, also in Germany, France, and Italy. The figures, however, are not generally impressions from wood-blocks, but are for the most part wholly executed by means of stencils. One of the latter class, representing the shrine of “Notre Dame de Hal,”—coloured in the most wretched taste with brick-dust red and shining green,—is 58 now lying before me. It was given to a gentleman who visited Halle, near Brussels, in 1829. It is nearly of the same size as many of the old devotional wood-cuts of Germany, being about four inches high, by two and three-quarters wide.II.27

The next step in the progress of wood engraving, subsequent to the production of single cuts, such as the St. Christopher, the Annunciation, and the St. Bridget, in each of which letters are sparingly introduced, was the application of the art to the production of those works which are known to bibliographers by the name of BLOCK-BOOKS: the most celebrated of which are the Apocalypsis, seu Historia Sancti Johannis; the Historia Virginis ex Cantico Canticorum; and the Biblia Pauperum. The first is a history, pictorial and literal, of the life and revelations of St. John the Evangelist, derived in part from the traditions of the church, but chiefly from the book of Revelations. The second is a similar history of the Virgin, as it is supposed to be typified in the Songs of Solomon; and the third consists of subjects representing some of the most important passages in the Old and New Testament, with texts either explaining the subject, or enforcing the example of duty which it may afford. With the above, the Speculum Humanæ Salvationis is usually, though improperly, classed, as the whole of the text, in that which is most certainly the first edition, is printed from movable metal types. In the others the explanatory matter is engraved on wood, on the same block with the subject to which it refers.

All the above books have been claimed by Meerman and other Dutch writers for their countryman, Laurence Coster: and although no date, either impressed or manuscript, has been discovered in any one copy from which the period of its execution might be ascertained,II.28 yet such appears to have been the clearness of the intuitive light which guided those authors, that they have assigned to each work the precise year in which it appeared. According to Seiz, the History of the Old and New Testament, otherwise called the Biblia Pauperum, appeared in 1432; the History of the Virgin in 1433; the Apocalypse in 1434; and the Speculum in 1439. For such assertions, however, he has not the slightest ground. That the three first might appear at some period between 1430 and 1450, is not unlikely;II.29 but that the Speculum—the text of which 59 in the first edition was printed from metal types—should be printed before 1460, is in the highest degree improbable.

Upon extremely slight grounds it has been conjectured that the Biblia Pauperum, the Apocalypse, and the Ars Moriendi,—another block-book,—were engraved before the year 1430. The Rev. T. H. Horne, “a gentleman long and well known for his familiar acquaintance with books printed abroad,” says Dr. Dibdin, “had a copy of each of the three books above mentioned, bound in one volume, upon the cover of which the following words were stamped: Hic liber relegatus fuit per Plebanum. ecclesie”—with the date, according to the best of the Rev. Mr. Horne’s recollection, 142(8). As he had broken up the volume, and had parted with the contents, he gave the above information on the strength of his memory alone. He was, however, confident that “the binding was the ancient legitimate one, and that the treatises had not been subsequently introduced into it, and that the date was 142 odd; but positively anterior to 1430.”II.30

In such a case as this, however, mere recollection cannot be admitted as decisive of the fact, more especially when we know the many instances in which mistakes have been committed in reading the numerals in ancient dates. At page 88 of his Inquiry, Mr. Ottley, catching at every straw that may help to support his theory of wood engraving having been practised by the Cunio and others in the fourteenth century, refers to a print which a Monsieur Thierry professed to have seen at Lyons, inscribed “Schoting of Nuremberg,” with the date 1384; and at p. 256 he alludes to it again in the following words: “The date 1384 on the wood-cut preserved at Lyons, said to have been executed at Nuremberg, appears, I know not why, to have been suspected.” It has been more than suspected; for, on examination, it has been found to be 1584. Paul Von Stettin published an account of a Biblia Pauperum, the date of which he supposed to be 1414; but which, when closely examined, was found to be 1474: and Baron Von Hupsch, of Cologne, published in 1787 an account of some wood-cuts which he supposed to have been executed in 1420; but which, in the opinion of Breitkopf, were part of the cuts of a Biblia Pauperum, in which it was probably intended to give the 60 explanations in moveable types underneath the cuts, and probably of a later date than 1470.II.31

It is surprising that the Rev. Mr. Horne, who is no incurious observer of books, but an author who has written largely on Bibliography, should not have carefully copied so remarkable a date, or communicated it to a friend, when it might have been confirmed by a careful examination of the binding; and still more surprising is it that such binding should have been destroyed. From the very fact of his not having paid more particular attention to this most important date, and from his having permitted the evidence of it to be destroyed, the Rev. Mr. Horne seems to be an incompetent witness. Who would think of calling a person to prove from recollection the date of an old and important deed, who, when he had it in his possession, was so little aware of its value as to throw it away? The three books in question, when covered by such a binding, would surely be much greater than when bound in any other manner. Such a volume must have been unique; and, if the date on the binding were correct, it must have been admitted as decisive of a fact interesting to every bibliographer in Europe. It is not even mentioned in what kind of numerals the date was expressed, whether in Roman or Arabic. If the numerals had been Arabic, we might very reasonably suppose that the Rev. Mr. Horne had mistaken a seven for a two, and that, instead of “142 odd,” the correct date was “147 odd.” In Arabic numerals, such as were used about the middle of the fifteenth century, the seven may very easily be mistaken for a two.

The earliest ancient binding known, on which a date is impressed, is, I believe, that described by Laire.II.32 It is that of a copy of “Sancti Hieronymi Epistolæ;” and the words, in the same manner as that of the binding of which the Rev. Mr. Horne had so accurate a recollection, were “stamped at the extremity of the binding, towards the edge of the squares.” It is only necessary to cite the words impressed on one of the boards, which were as follows:

“Illigatus est Anno Domini 1469
Per me Johannem
Richenbach Capellanum
In Gyslingen.”II.33

The numerals of the date it is to be observed were Arabic. In the library of Dr. Kloss of Frankfort, sold in London by Sotheby and Son in 1835, were two volumes, “St. Augustini de Civitat. Dei, Libri xxii. 61 1469,” and “St. Augustini Confessiones” of the same date; both of which were bound by “Johannes Capellanus in Gyslingen,” and who in the same manner had impressed his name on the covers with the date 1470. Both volumes had belonged to “Dominus Georgius Ruch de Gamundia.”II.34 That the volume formerly in the Rev. Mr. Horne’s possession was bound by the curate of Geisslingen I by no means pretend to say, though I am firmly of opinion that it was bound subsequent to 1470, and that the character which he supposed to be a two was in reality a different figure. It is worthy of remark that it appears to have been bound by the “Plebanus” of some church, a word which is nearly synonymous with “Capellanus.”II.35

As it does not come within the plan of the present volume to give a catalogue of all the subjects contained in the block-books to which it may be necessary to refer as illustrating the progress of wood engraving, I shall confine myself to a general notice of the manner in which the cuts are executed, with occasional observations on the designs, and such remarks as may be likely to explain any peculiarity of appearance, or to enable the reader to form a distinct idea of the subject referred to.

At whatever period the Apocalypse, the History of the Virgin, and the Biblia Pauperum may have been executed, the former has the appearance of being the earliest; and in the absence of everything like proof upon the point, and as the style in which it is engraved is certainly more simple than that of the other two, it seems entitled to be first noticed in tracing the progress of the art.

Of the Apocalypse,—or “Historia Sancti Johannis Evangelistæ ejusque Visiones Apocalypticæ,” as it is mostly termed by bibliographers, for the book itself has no title,—Heineken mentions no less than six editions, the earliest of which he considers to be that described by him at page 367 of his “Idée Générale d’une Collection complète d’Estampes.” He, however, declares that the marks by which he has assigned to each edition its comparative antiquity are not infallible. It is indeed very evident that the marks which he assumed as characteristic of the relative order of the different editions were merely arbitrary, and could by no means be admitted as of the slightest consequence in enabling any 62 person to form a correct opinion on the subject. He notices two editions as the first and second, and immediately after he mentions a circumstance which might almost entitle the third to take precedence of them both; and that which he saw last he thinks the oldest of all. The designs of the second edition described by him, he says, are by another master than those of the first, although the artist has adhered to the same subjects and the same ideas. The third, according to his observations, differs from the first and second, both in the subjects and the descriptive text. The fourth edition is from the same blocks as the third; the only difference between them being, that the fourth is without the letters in alphabetical order which indicate the succession of the cuts. The fifth differed from the third or fourth only in the text and the directing letters, as the designs were the same; the only variations that could be observed being extremely trifling. After having described five editions of the book, he decides that a sixth, which he saw after the others, ought to be considered the earliest of all.II.36 In all the copies which he had seen, the impressions had been taken by means of a rubber, in such a manner that each leaf contained only one engraving; the other side, which commonly bore the marks of the rubber, being without a cut. The impressions when collected into a volume faced each other, so that the first and last pages were blank.

The edition of the Apocalypse to which I shall now refer is that described by Heineken, at page 364, as the fifth; and the copy is that mentioned by him, at page 367, as then being in the collection of M. de Gaignat, and as wanting two cuts, Nos. 36 and 37. It is at present in the King’s Library at the British Museum.

It is a thin folio in modern red morocco binding, and has, when perfect, consisted of fifty wood engravings, with their explanatory text also cut in wood, generally within an oblong border of a single line, within the field of the engraving, and not added underneath, as in the Speculum Salvationis, nor in detached compartments, both above and below, as in the Biblia Pauperum. The paper, which is somewhat of a cream colour, is stout, with rather a coarse surface, and such as we find the most ancient books printed on. As each leaf has been pasted down on another of modern paper, in order to preserve it, the marks of the rubber at the back of each impression, as described by Heineken, cannot be seen.   see text The annexed outline is a reduced copy of a paper-mark, which may be perceived on some of the leaves. It is very like that numbered “vii.” at p. 224, vol. i. of Mr. Ottley’s Inquiry, and which he says occurs in the edition called the first Latin of the Speculum Salvationis. It is nearly the same as that which is to be seen in Earl Spencer’s “Historia Virginis;” and Santander 63 states that he has noticed a similar mark in books printed at Cologne by Ulric Zell, and Bart. de Unkel; at Louvain by John Veldener and Conrad Braen; and in books printed at Utrecht by Nic. Ketelaer and Gerard de Leempt.

The size of the largest cuts, as defined by the plain lines which form the border, is about ten and five-eighths inches high, by seven and six-eighths inches wide; of the smallest, ten and two-eighths inches high, by seven and three-eighths wide.II.37 The order in which they are to be placed in binding is indicated by a letter of the alphabet, which serves the same purpose as our modern signatures,—engraved in a conspicuous part of the cut. For instance, the first two, which, as well as the others, might either face each other or be pasted back to back, are each marked with the letter a; the two next with the letter b, and so on through the alphabet. As the alphabet—which has the i the same as the j, the v the same as the u, and has not the w—became exhausted at the forty-sixth cut, the forty-seventh and forty-eighth are marked with a character which was used to represent the words “et cetera;” and the forty-ninth and fiftieth with the terminal abbreviation of the letters “us.” In the copy described by Heineken, he observed that the directing letters m and n were wanting in the twenty-fourth and twenty-sixth cuts, and in the copy under consideration they are also omitted. The m, however, appears to have been engraved, though for some reason or other not to have been inked in taking an impression; for on a careful examination of this cut,—without being aware at the time of Heineken having noticed the omission,—I thought that I could very plainly discern the indention of the letter above one of the angels in the upper compartment of the print.

Of the forty-eight cutsII.38 contained in the Museum copy, the greater number are divided by a horizontal line, nearly in the middle, and thus each consists of two compartments; of the remainder, each is occupied by a single subject, which fills the whole page. In some, the explanatory text consists only of two or three lines; and in others it occupies so 64 large a space, that if it were set up in moderately sized type, it would be sufficient to fill a duodecimo page. The characters are different from those in the History of the Virgin and the Biblia Pauperum, and are smaller than those of the former, and generally larger and more distinctly cut than those of the latter; and although, as well as in the two last-named books, the words are much abbreviated, yet they are more easy to be made out than the text of either of the others. The impressions on the whole are better taken than those of the Biblia Pauperum, though in lighter-coloured ink, something like a greyish sepia, and apparently of a thinner body. It does not appear to have contained any oil, and is more like distemper or water-colour than printer’s ink. From the manner in which the lines are indented in the paper, in several of the cuts, it is evident that they must either have been subjected to a considerable degree of pressure or have been very hard rubbed.

Although some of the figures bear a considerable degree of likeness to others of the same kind in the Biblia Pauperum, I cannot think that the designs for both books were made by the same person. The figures in the different works which most resemble each other are those of saints and angels, whose form and expression have been represented according to a conventional standard, to which most of the artists of the period conformed, in the same manner as in representing the Almighty and Christ, whether they were painters, glass-stainers, carvers, or wood-engravers. In many of the figures the drapery is broken into easy and natural folds by means of single lines; and if this were admitted as a ground for assigning the cut of the Annunciation to Italy, with much greater reason might the Apocalypse be ascribed to the same country.

Without venturing to give an opinion whether the cuts were engraved in Germany, Holland, or in the Low Countries, the drawing of many of the figures appears to correspond with the idea that I have formed of the style of Greek art, such as it was in the early part of the fifteenth century. St. John was the favourite apostle of the Greeks, as St. Peter was of the church of Rome; and as the Revelations were more especially addressed to the churches of Greece, they were more generally read in that country than in Western Europe. Artists mostly copy, in the heads which they draw, the general expression of the countryII.39 to which they belong, and where they have received their first impressions; and in the Apocalypse the character of several of the heads appears to be decidedly Grecian. The general representation, too, of several visions would seem to have been suggested by a Greek who was familiar with that portion of the New Testament which was so generally perused in his native land, and whose annunciations and figurative prophecies were, in the early 65 part of the fifteenth century, commonly supposed by his countrymen to relate to the Turks, who at that time were triumphing over the cross. With them Mahomet was the Antichrist of the Revelations, and his followers the people bearing the mark of the beast, who were to persecute, and for a time to hold in bondage, the members of the church of Christ. As many Greeks, both artists and scholars, were driven from their country by the oppression of the Turks several years before the taking of Constantinople in 1453, I am induced to think that to a Greek we owe the designs of this edition of the Apocalypse. In the lower division of the twenty-third cut, m, representing the fight of Michael and his angels with the dragon, the following shields are borne by two of the heavenly host.

  see text

The crescent, as is well known, was one of the badges of Constantinople long previous to its capture by the Turks. The sort of cross in the other shield is very like that in the arms of the knights of St. Constantine, a military order which is said to have been founded at Constantinople by the Emperor Isaac Angelus Comnenus, in 1190. The above coincidences, though trifling, tend to support the opinion that the designs were made by a Greek artist. It is, however, possible, that the badges on the shields may have been suggested by the mere fancy of the designer, and that they may equally resemble the heraldic bearings of some order or of some individuals of Western Europe.

Though some of the designs are very indifferent, yet there are others which display considerable ability, and several of the single figures are decidedly superior to any that are contained in the other block-books. They are drawn with greater vigour and feeling; and though the designs of the Biblia Pauperum show a greater knowledge of the mechanism of art, yet the best of them, in point of expression and emphatic marking of character, are inferior to the best in the Apocalypse.

With respect to the engraving, the cuts are executed in the simplest manner, as there is not the least attempt at shading, by means of cross lines or hatchings, to be perceived in any one of the designs. The most difficult part of the engraver’s task, supposing the drawings to have been made by another person, would be the cutting of the letters, which in several of the subjects must have occupied a considerable portion of 66 time, and have required no small degree of care. The following is a reduced copy of the first cut.

  see text

In the upper portion of the subject, St. John is seen addressing four persons, three men and a woman; and the text at the top informs us of the success of his ministry: “Conversi ab idolis, per predicationem beati Johannis, Drusiana et ceteri.”—“By the preaching of St. John, Drusiana and others are withdrawn from their idols.” The letter a, a little above the saint’s outstretched hand, indicates that the cut is the first of the series. In the lower compartment St. John is seen baptizing Drusiana, who, as she stands naked in the font, is of very small size compared with the saint. The situation in which Drusiana is placed might be alleged in support of their peculiar tenets, either by the Baptists, who advocate immersion as the proper mode of administering the rite, or by those who consider sprinkling as sufficient; but in each case with a difficulty which it would not be easy to explain: for if Drusiana were to be baptized by immersion, the font is too small to allow her to be dipped overhead; and 67 if the rite were to be administered by mere sprinkling, why is she standing naked in the font? To the right of the cut are several figures, two of whom are provided with axes, who seem wishful to break open the door of the chapel in which St. John and his proselyte are seen. The inscription above their heads lets us know that they are—“Cultores ydolorum explorantes facta ejus;”—“Worshippers of idols watching the saint’s proceedings.”

The following cut is a copy of the eighteenth of the Apocalypse, which is illustrative of the XIth and XIIIth chapters of Revelations. The upper portion represents the execution of the two witnesses of the Lord, who are in the tablet named Enoch and Helyas, by the command of the beast which ascendeth out of the bottomless pit, and which is Antichrist. He is seen issuing his commands for the execution of the witnesses; and the face of the executioner who has just used his sword, and who is looking towards him with an expression of brutal exultation, might have served Albert Durer for that of the mocker in his cut of Christ crowned with thorns.

  see text

68

The inscription to the right, is the 7th verse of the XIth chapter, with the names of Enoch and Helyas inserted as those of the two witnesses: “Cum finierunt Enoch et Helyas testimonium suum, bestia quæ ascendit de abisso faciet contra eos bellum, et vincet eos et occidet illos.” In our translation the verse is rendered thus: “And when they shall have finished their testimony, the beast that ascendeth out of the bottomless pit shall make war against them, and shall overcome them and kill them.”

The tablet to the left contains the following inscription: “Et jacebunt corpora eorum in plateis, et non sinent poni in monumentis.” It is formed of two passages, in the 8th and 9th verses of the XIth chapter of Revelations, which are thus rendered in our version of the Bible: “And their dead bodies shall lie in the street, . . . and they of the people . . . shall not suffer their dead bodies to be put in graves.”

In the lower compartment Antichrist is seen working his miracles, uprooting the two olive trees, typical of the two witnesses whom he had caused to be slain.II.40 Two of his followers are seen kneeling as if worshipping him, while more to the left are the supporters of the true faith delivered into the hands of executioners. The design is illustrative of the XIIIth chapter of Revelations. The following is the inscription above the figure of Antichrist:—“Hic facit Antichristus miracula sua, et credentes in ipsum honorat, et incredentes variis interficit pœnis.”—“Here Antichrist is performing his miracles, honouring those who believe in him, and putting the incredulous to death by various punishments.” The leaves of the trees which Antichrist has miraculously uprooted are extremely like those of the tree of life engraved in one of the cuts of the Biblia Pauperum, and of which a copy will be found in a subsequent page.

In several of the cuts, the typical expressions which occur in the texts are explained. Thus, in cut eighth, we are informed that “Stolæ albæ animarum gloriam designant.”—“The white vestments denote the glory of departed souls.” In the lower compartment of the same cut, the “cæli recessio”—“the opening of the heavens”—is explained to be the communication of the Bible to the Gentiles. In the lower compartment of the ninth cut, “much incense” is said to signify the precepts of the Gospel; the “censers,” the hearts of the Apostles; and the “golden altar,” the Church.

The next block-book which demands notice is that named “Historia seu Providentia Virginis Mariæ, ex Cantico Canticorum:” that is, “The History or Prefiguration of the Virgin Mary, from the Song of Songs.” It is of small-folio size, and consists of sixteen leaves, printed on one side only by means of friction; and the ink is of a dark brown, approaching nearly to black. Each impressed page contains two subjects, 69 one above the other; the total number of subjects in the book is, consequently, thirty-two.

Of this book, according to the observations of Heineken, there are two editions; which, from variations noticed by him in the explanatory text, are evidently from different blocks; but, as the designs are precisely the same, it is certain that the one has been copied from the other.II.41 That which he considers to be the first edition, has, in his opinion, been engraved in Germany; the other, he thinks, was a copy of the original, executed by some engraver in Holland. The principal ground on which he determines the priority of the editions is, that in the one the text is much more correctly given than in the other; and he thence concludes that the most correct would be the second. In this opinion I concur; not that his rule will universally hold good, but that in this case the conclusion which he has drawn seems the most probable. The designs, it is admitted, are precisely the same; and as the cuts of the one would in all probability be engraved from tracings or transfers of the other, it is not likely that we should find such a difference in the text of the two editions if that of the first were correct. A wood-engraver—on this point I speak from experience—would be much more likely to commit literal errors in copying manuscript, than to deviate in cutting a fac-simile from a correct impression. Had the text of the first edition been correct,—considering that the designs of the one edition are exact copies of those of the other,—it is probable that the text of both would have been more nearly alike. But as there are several errors in the text of the first edition, it is most likely that many of them would be discovered and corrected by the person at whose instance the designs were copied for the second. Diametrically opposite to this conclusion is that of Mr. Ottley, who argues as follows:II.42 “Heineken endeavours to draw another argument in favour of the originality of the edition possessed by Pertusati, Verdussen, and the Bodleian library, from the various errors, in that edition, in the Latin inscriptions on the scrolls; which, he says, are corrected in the other edition. But it is evident that this circumstance makes in favour of an opposite conclusion. The artist who originally invented the work must have been well acquainted with Latin, since it is, in fact, no other than an union of many of the most beautiful verses of the Book of Canticles, with a series of designs illustrative of the divine mysteries supposed to be revealed in that sacred poem; and, consequently, we have reason to consider that edition the original in which the inscriptions are given with the most correctness; and to ascribe the gross blunders in the other to the ignorance of some ordinary wood-engraver by whom the work was copied.” Even granting the assumption that the 70 engraver of the edition, supposed by Mr. Ottley to be the first, was well acquainted with Latin, and that he who engraved the presumed second did not understand a word of that language, yet it by no means follows that the latter could not make a correct tracing of the engraved text lying before him. Because a draughtsman is unacquainted with a language, it would certainly be most erroneous to infer that he would be incapable of copying the characters correctly. Besides, though it does not benefit his argument a whit, it is surely assuming too much to assert that the artist who made the designs also selected the texts, and that he must have been well acquainted with Latin; and that he who executed Mr. Ottley’s presumed second edition was some ignorant ordinary wood-engraver. Did the artists who executed the fac-similes in Mr. Ottley’s work, or in Dr. Dibdin’s “Bibliotheca Spenceriana,” understand the abbreviated Latin which in many instances they had to engrave; and did they in consequence of their ignorance of that language copy incorrectly the original texts and sentences which were before them?

In a copy which Heineken considers to be of the second edition, belonging to the city of Harlem, that writer observed the following inscription, from a wood block, impressed, as I understand him, at the top of the first cut. “Dit is die voersinicheit va Marie der mod . godes . en is gehete in lath . Cāti.” This inscription—which Heineken says is “en langue Flamande, ou plûtôt en Plât-Alemand”—may be expressed in English as follows: “This is the prefiguration of Mary the mother of God, and is in Latin named the Canticles.” Heineken expresses no doubt of this inscription being genuine, though he makes use of it as an argument in support of his opinion, that the copy in which it occurs was one of later edition; “for it is well known,” he observes, “that the earliest editions of printed books are without titles, and more especially those of block-books.” As this inscription, however, has been found in the Harlem copy only, I am inclined to agree with Mr. Ottley in considering it as a silly fraud devised by some of the compatriots of Coster for the purpose of establishing a fact which it is, in reality, much better calculated to overthrow.II.43

Heineken, who appears to have had more knowledge than taste on the subject of art, declares the History of the Virgin to be “the most Gothic of all the block-books; that it is different from them both in the style of the designs and of the engraving; and that the figures are very like the ancient sculptures in the churches of Germany.” If by the term “Gothic” he means rude and tasteless, I differ with him entirely; for, though there be great sameness in the subjects, yet the figures, generally, are more gracefully designed than those of any other block-book 71 that I have seen. Compared with them, those of the Biblia Pauperum and the Speculum might be termed “Gothic” indeed.

  see text

The above group,—from that which Heineken considers the first edition,—in which the figures are of the size of the originals, is taken from the seventh subject in Mr. Ottley’s enumeration;II.44 that is, from the upper portion of the fourth cut.

The text is the 14th verse of the 1st chapter of the Song of Solomon: “Botrus cipri dilectus meus inter vineas enngadi;” which in our Bible is translated: “My beloved is unto me as a cluster of camphire in the vineyards of En-gedi.” In every cut the female figures are almost precisely the same, and the drapery and the expression scarcely vary. From the easy and graceful attitudes of his female figures, as well as from the 72 manner in which they are clothed, the artist may be considered as the Stothard of his day.

  see text

The two preceding subjects are impressed on the second leaf, in the order in which they are here represented, forming Nos. 3 and 4 in Mr. Ottley’s enumeration. They are reduced copies from the originals in the first edition, and afford a correct idea of a complete page.II.45

On the scroll to the left, in the upper subject, the words are intended for—“Trahe me, post te curremus in odore unguentorum tuorum.” They are to be found in the 4th and 3rd verses of the 1st chapter of the Song of Solomon. In our Bible the phrases are translated as follows: “Draw me, we will run after thee, . . . [in] the savour of thy good ointments.” 73 In the scroll to the right, the inscription is from the 14th verse of the IInd chapter: “Sonet vox tua in auribus meis, vox enim tua dulcis et facies tua decora:” which is thus rendered in our Bible: “Let me hear thy voice, for sweet is thy voice, and thy countenance is comely.”

On the scroll to the left, in the lower compartment, is the following inscription, from verse 10th, chapter IInd: “En dilectus meus loquitur mihi, Surge, propera, amica mea:” in our Bible translated thus: “My beloved spake, and said unto me, Rise up, my love, my fair one, and come away.” The inscription on the scroll to the right is from 1st verse of chapter IVth: “Quam pulchra es amica mea, quam pulchra es! Oculi tui columbarum, absque eo quod intrinsecus latet.” The translation of this passage in our Bible does not correspond with that of the Vulgate in the last clause: “Behold thou art fair, my love; behold thou art fair; thou hast doves’ eyes within thy locks.”

The style in which the cuts of the History of the Virgin are engraved indicates a more advanced state of art than those in the Apocalypse. The field of each cut is altogether better filled, and the subjects contain more of what an engraver would term “work;” and shadowing, which is represented by courses of single lines, is also introduced. The back-grounds are better put in, and throughout the whole book may be observed several indications of a perception of natural beauty; such as the occasional introduction of trees, flowers, and animals. A vine-stock, with its trellis, is happily and tastefully introduced at folio 4 and folio 10; and at folio 12 a goat and two sheep, drawn and engraved with considerable ability, are perceived in the background. Several other instances of a similar kind might be pointed out as proofs that the artist, whoever he might be, was no unworthy precursor of Albert Durer.

From a fancied delicacy in the engraving of the cuts of the History of the Virgin, Dr. Dibdin was led to conjecture that they were the “production of some metallic substance, and not struck off from wooden blocks.”II.46 This speculation is the result of a total ignorance of the practical part of wood engraving, and of the capabilities of the art; and the very process which is suggested involves a greater difficulty than that which is sought to be removed. But, in fact, so far from the engravings being executed with a delicacy unattainable on wood, there is nothing in them—so far as the mere cutting of fancied delicate lines is concerned—which a mere apprentice of the present day, using very ordinary tools, would not execute as well, either on pear-tree, apple-tree, or beech, the kinds of wood on which the earliest engravings are supposed to have been made. Working on box, there is scarcely a line in all the series which a skilful wood-engraver could not split. In a similar manner Mr. John 74 Landseer conjectured from the frequent occurrence of cross-hatching in the wood engravings of the sixteenth century, that they, instead of being cut on wood, had in reality been executed on type-metal; although, as is known to every wood-engraver, the execution of such hatchings on type-metal would be more difficult than on wood. When, in refutation of his opinion, he was shown impressions from such presumed blocks or plates of type-metal, which from certain marks in the impressions had been evidently worm-eaten, he—in the genuine style of an “ingenious disputant” who could

“Confute the exciseman and puzzle the vicar,—”

abandoned type-metal, and fortified his “stubborn opinion behind vegetable putties or pastes that are capable of being hardened—or any substance that is capable of being worm-eaten.”II.47 Such “commenta opinionum”—the mere figments of conjecture—only deserve notice in consequence of their extravagance.

The History of the Virgin, in the same manner as every other ancient block-book, has been claimed for Coster by those who ascribe to him the invention both of wood engraving and printing with moveable types; but if even the churchwarden of St. Bavon’s in Harlem ever had handled a graver, or made a design, or if he was even the cause of wood-cuts being engraved by others,—every one of which assertions I very much doubt,—I should yet feel strongly inclined to believe that the work in question was the production of an artist residing either in Suabia or Alsace.

Scarcely any person who has had an opportunity of examining the works of Martin Schön, or Schöngauer,—one of the earliest German copper-plate engravers,—who is said to have died in 1486, can fail, on looking over the designs in the History of the Virgin, to notice the resemblance which many of his female figures bear to those in the above-named work. The similarity is too striking to have been accidental. I am inclined to believe that Martin Schön must have studied—and diligently too—the subjects contained in the History, or that he had received his professional education in a school which might possibly be founded by the artist who designed and engraved the wood-cuts in question, or under a master who had thoroughly adopted their style.

Martin Schön was a native of Colmar in Alsace, where he was born about 1453, but was a descendant of a family, probably of artists, which originally belonged to Augsburg. Heineken and Von Murr both bear testimony,II.48 though indirectly, to the resemblance which his works bear to the designs in the History of the Virgin. The former states that the figures in the History are very like the ancient sculptures in the churches 75 of Germany, and Von Murr asserts that such sculptures were probably Martin Schön’s models.

In two or three of the designs in the History of the Virgin several shields of arms are introduced, either borne by figures, or suspended from a wall. As the heraldic emblems on such shields were not likely to be entirely suggested by the mere fancy of the artist, I think that most of them will be found to belong to Germany rather than to Holland; and the charge on one of them,—two fish back to back, which is rather remarkable, and by no means common, is one of the quarterings of the former Counts of Wirtemberg, the very district in which I am inclined to think the work was executed. I moreover fancy that in one of the cuts I can perceive an allusion to the Council of Basle, which in 1439 elected Amadeus of Savoy as Pope, under the title of Felix V, in opposition to Eugene IV. In order to afford those who are better acquainted with the subject an opportunity of judging for themselves, and of making further discoveries which may support my opinions if well-founded, or which may correct them if erroneous, I shall give copies of all the shields of arms which occur in the book. The following cut of four figures—a pope, two cardinals, and a bishop—occurs in the upper compartment of the nineteenth folio. The shield charged with a black eagle also occurs in the same compartment.

  see text

The preceding figures are seen looking over the battlements of a house in which the Virgin, typical of the Church, is seen in bed. On a scroll is inscribed the following sentence, from the Song of Solomon, chap. iii. v. 2: “Surgam et circumibo civitatem; per vicos et plateas queram quem diligit anima mea:” which is thus translated in our Bible: “I will rise now, and go about the city in the streets, and in the broad ways I will seek him whom my soul loveth.” In the same design, the Virgin, with her three attendants, are seen in a street, where two men on horseback 76 appear taking away her mantle. One of the men bears upon his shield the figure of a black eagle, the same as that which appears underneath the wood-cut above given. Upon a scroll is this inscription, from Solomon’s Song, chapter V. verse 7: “Percusserunt et vulneraverunt me, tulerunt pallium meum custodes murorum.” In our Bible the entire verse is thus translated: “The watchmen that went about the city found me; they smote me, they wounded me: the keepers of the walls took away my veil from me.”

As the incidents in the life of the Virgin, described in the Canticles, were assumed by commentators to be typical of the history of the Church, I am inclined to think that the above cut may contain an allusion to the disputes between Pope Eugene IV. and the Council assembled at Basle in 1439. The passage in the first inscription, “I will seek him whom my soul loveth,” might be very appropriately applied to a council which professed to represent the Church, and which had chosen for itself a new head. The second inscription would be equally descriptive of the treatment which, in the opinion of the same council, the Church had received from Eugene IV, whom they declared to be deposed, because “he was a disturber of the peace and union of the Church; a schismatic and a heretic; guilty of simony; perjured and incorrigible.” On the shield borne by the figure of a pope wearing a triple crown, is a fleur-de-lis; but whether or no this flower formed part of the armorial distinctions of Amadeus Duke of Savoy, whom the council chose for their new pope, I have not been able to ascertain. The lion borne by the second figure, a cardinal, is too general a cognizance to be assigned to any particular state or city. The charge on the shield borne by the third figure, also a cardinal, I cannot make out. The cross-keys on the bishop’s shield are the arms of the city of Ratisbon.

The following shields are borne by angels, who appear above the battlements of a wall in the lower compartment of folio 4, forming the eighth subject in Mr. Ottley’s enumeration.

  see text

On these I have nothing to remark further than that the double-headed eagle is the arms of the German empire. The other three I leave to be deciphered by others. The second, with an indented chief, and something like a rose in the field, will be found, I am inclined to think, to be the arms of some town or city in Wirtemberg or Alsace. I give the three inscriptions here, not that they are likely to throw any light on the subject, but because the third has not hitherto been deciphered. They are 77 all from the IVth chapter of the Song of Solomon. The first is from verse 12: “Ortus conclusus est soror, mea sposa; ortus conclusus, fons signatus:” in our translation of the Bible: “A garden enclosed is my sister, my spouse; a spring shut up, a fountain sealed.” The second is from verse 15: “Fons ortorum, puteus aquarum vivencium quæ fluunt impetu de Lybano:” in our Bible: “A fountain of gardens, a well of living waters, and streams from Lebanon.” The third is from verse 16: “Surge Aquilo; veni Auster, perfla ortum et fluant aromata illius:” in our Bible: “Awake, O north wind; and come, thou south; blow upon my garden, that the spices thereof may flow out.”

  see text

In the upper division of folio 15, which is the twenty-ninth subject in Mr. Ottley’s enumeration, the above shields occur. They are suspended on the walls of a tower, which is represented by an inscription as “the armoury whereon hang a thousand bucklers, all shields of mighty men.”II.49

On the first four I shall make no remark beyond calling the attention of those skilled in German heraldry to the remarkable charge in the first shield, which appears something like a cray-fish. The sixth, “two trouts hauriant and addorsed,” is one of the quarterings of the house of Wirtemberg as lords of Mompelgard. The seventh is charged with three crowns, the arms of the city of Cologne. The charge of the eighth I take to be three cinquefoils, which are one of the quarterings of the family of Aremberg. The cross-keys in the ninth are the arms of the city of Ratisbon.

The four following shields occur in the lower division of folio 15. They are borne by men in armour standing by the side of a bed. On a scroll is the following inscription, from the 7th and 8th verses of the third chapter of Solomon’s Song. “En lectulum Salomonis sexaginta fortes ambiunt, omnes tenentes gladios:” in our Bible: “Behold his bed, which is Solomon’s; three score valiant men are about it . . . . . they all hold swords.”

The first three of the shields on the following page I shall leave to be 78 assigned by others. The fourth, which is charged with a rose, was the arms of Hagenau, a town in Alsace.

  see text

As so little is known respecting the country where, and the precise time when, the principal block-books appeared,—of which the History of the Virgin is one,—I think every particular, however trifling, which may be likely to afford even a gleam of light, deserving of notice. It is for this reason that I have given the different shields contained in this and the preceding pages; not in the belief that I have made any important discovery, or established any considerable facts; but with the desire of directing to this subject the attention of others, whose further inquiries and comparisons may perhaps establish such a perfect identity between the arms of a particular district, and those contained in the volume, as may determine the probable locality of the place where it was executed. The coincidences which I have noticed were not sought for. Happening to be turning over Sebastian Munster’s Cosmography when a copy of the History of the Virgin was before me, I observed that the two fish in the arms of the Counts of Wirtemberg,II.50 and those in the 15th folio of the History, were the same. The other instances of correspondence were also discovered without search, from having occasionally, in tracing the progress of wood engraving, to refer to Merian’s Topographia.

Considering the thickness of the paper on which the block-books are printed,—if I may apply this term to them,—and the thin-bodied ink which has been used. I am at a loss to conceive how the early wood-engravers have contrived to take off their impressions so correctly; for in all the block-books which I have seen, where friction has evidently been the means employed to obtain the impression, I have only noticed two subjects in which the lines appeared double in consequence of the shifting of the paper. From the want of body in the ink, which appears in the Apocalypse to have been little more than water-colour, it is not likely the paper could be used in a damp state, otherwise the ink would run or spread; and, even if this difficulty did not exist, the paper in a damp state could not have borne the excessive rubbing which it appears to have received in order to obtain the impression.II.51 Even with 79 such printer’s ink as is used in the present day,—which being tenacious, renders the paper in taking an impression by means of friction much less liable to slip or shift,—it would be difficult to obtain clear impressions on thick paper from blocks the size of those which form each page of the Apocalypse, or the History of the Virgin.

Mr. Ottley, however, states that no less than two pages of the History of the Virgin have been engraved on the same block. His observations on this subject are as follows: “Upon first viewing this work, I was of opinion that each of the designs contained in it was engraved upon a separate block of wood: but, upon a more careful examination, I have discovered that the contents of each two pages—that is, four subjects—were engraved on the same block. The number of wooden blocks, therefore, from which the whole was printed, was only eight. This is proved in the first two pages of the copy before me;II.52 where, near the bottom of the two upper subjects, the block appears to have been broken in two, in a horizontal direction,—after it was engraved,—and joined together again; although not with such exactness but that the traces of the operation clearly show themselves. The traces of a similar accident are still more apparent in the last block, containing the Nos. 29, 30, 31, 32. The whole work was, therefore, printed on eight sheets of paper from the same number of engraved blocks, the first four subjects being printed from the same block upon the same sheet,—and so on with the rest; and, indeed, in Lord Spencer’s copy, each sheet, being mounted upon a guard, distinctly shows itself entire.”II.53

The appearance of a corresponding fracture in two adjacent pages would certainly render it likely that both were engraved on the same block; though I should like to have an opportunity of satisfying myself by inspection whether such appearances are really occasioned by a fracture or not; for it is rather singular that such appearances should be observable on the first and the last blocks only. I always reluctantly speculate, except on something like sufficient grounds; but as I have not seen a copy of the edition to which Mr. Ottley refers, I beg to ask if the traces of supposed fracture in the last two pages do not correspond with those in the first two? and if so, would it not be equally reasonable to infer that eight subjects instead of four were engraved on the same block? A block containing only two pages would be about seventeen inches by ten, allowing for inner margins; and to obtain clear impressions from it by means of friction, on dry thick paper, and with mere water-colour 80 ink, would be a task of such difficulty that I cannot conceive how it could be performed. No traces of points by which the paper might be kept steady on the block are perceptible; and I unhesitatingly assert that no wood-engraver of the present day could by means of friction take clear impressions from such a block on equally thick paper, and using mere distemper instead of printer’s ink. As the impressions in the History of the Virgin have unquestionably been taken by means of friction, it is evident to me that if the blocks were of the size that Mr. Ottley supposes, the old wood-engravers, who did not use a press, must have resorted to some contrivance to keep the paper steady, with which we are now unacquainted.

Heineken describes an edition of the Apocalypse consisting of forty-eight leaves, with cuts on one side only, which, when bound, form a volume of three “gatherings,” or collections, each containing sixteen leaves. Each of these gatherings is formed by eight folio sheets folded in the middle, and placed one within the other, so that the cuts are worked off in the following manner: On the outer sheet of the gathering, forming the first and the sixteenth leaf, the first and the sixteenth cuts are impressed, so that when the sheet is folded they face each other, and the first and the last pages are left blank. In a similar manner the 2nd and 15th; the 3d and 14th; the 4th and 13th; the 5th and 12th; the 6th and 11th; the 7th and 10th, and the 8th and 9th, are, each pair respectively, impressed on the same side of the same sheet. These sheets when folded for binding are then placed in such a manner that the first is opposite the second; the third opposite the fourth, and so on throughout the whole sixteen. Being arranged in this manner, two cuts and two blank pages occur alternately. The reason for this mode of arrangement was, that the blank pages might be pasted together, and the cuts thus appear as if one were impressed on the back of another. A familiar illustration of this mode of folding, adopted by the early wood-engravers before they were accustomed to impress their cuts on both sides of a leaf, is afforded by forming a sheet of paper into a little book of sixteen leaves, and numbering the second and third pages 1 and 2, leaving two pages blank; then numbering the fifth and sixth 3 and 4, and so to No. 16, which will stand opposite to No. 15, and have its back, forming the outer page of the gathering, unimpressed.

Of all the block-books, that which is now commonly called “Biblia Pauperum,”—the Bible of the Poor,—is most frequently referred to as a specimen of that kind of printing from wood-blocks which preceded typography, or printing by means of moveable characters or types. This title, however, has given rise to an error which certain learned bibliographers have without the least examination adopted, and have afterwards given to the public considerably enlarged, at least, if not 81 corrected.II.54 It has been gravely stated that this book, whose text is in abbreviated Latin, was printed for the use of the poor in an age when even the rich could scarcely read their own language. Manuscripts of the Bible were certainly at that period both scarce and costly, and not many individuals even of high rank were possessed of a copy; but to conclude that the first editions of the so-called “Biblia Pauperum” were engraved and printed for the use of the poor, appears to be about as legitimate an inference as to conclude that, in the present day, the reprints of the Roxburghe club were published for the benefit of the poor who could not afford to purchase the original editions. That a merchant or a wealthy trader might occasionally become the purchaser of “Biblia Pauperum,” I am willing to admit,—though I am of opinion that the book was never expressly intended for the laity;—but that it should be printed for the use of the poor, I cannot bring myself to believe. If the poor of Germany in the fifteenth century had the means of purchasing such books, and were capable of reading them, I can only say that they must have had more money to spare than their descendants, and have been more learned than most of the rich people throughout Europe in the present day. If the accounts which we have of the state of knowledge about 1450 be correct, the monk or friar who could read and expound such a work must have been esteemed as a person of considerable literary attainments.

The name “Biblia Pauperum” was unknown to Schelhorn and Schœpflin, and was not adopted by Meerman. Schelhorn, who was the first that published a fac-simile of one of the pages engraved on wood, gives it no distinctive name; but merely describes it as “a book which contained in text and figures certain histories and prophecies of the Old Testament, which, in the author’s judgment, were figurative of Christ, and of the works performed by him for the salvation of mankind.”II.55 Schœpflin calls it, “Vaticinia Veteris Testamenti de Christo;”II.56—“Prophecies of the Old Testament concerning Christ;” but neither this title, nor the description of Schelhorn, is sufficiently comprehensive; for the book contains not only prophecies and typical figures from the Old Testament, but also passages and subjects selected from the New. 82 The title which Meerman gives to it is more accurately descriptive of the contents: “Figuræ typicæ Veteris atque antitypicæ Novi Testamenti, seu Historia Jesu Christi in figuris;” that is, “Typical figures of the Old Testament and antitypical of the New, or the History of Jesus Christ pictorially represented.”II.57

Heineken appears to have been the first who gave to this book the name “Biblia Pauperum,” as it was in his opinion the most appropriate; “the figures being executed for the purpose of giving a knowledge of the Bible to those who could not afford to purchase a manuscript copy of the Scriptures.”II.58 This reason for the name is not, however, a good one: for, according to his own statement, the only copy which he ever saw with the title or inscription “Biblia Pauperum,” was a manuscript on vellum of the fourteenth century, in which the figures were drawn and coloured by hand.II.59 Meerman, however, though without adopting the title, had previously noticed the same manuscript, which in his opinion was as old as the twelfth or thirteenth century. As the word “Pauperum” formed part of the title of the book long before presumed cheap copies were printed from wood-blocks for the use of the poor, it could not be peculiarly appropriate as the title of an illumined manuscript on vellum, which the poor could as little afford to purchase as they could a manuscript copy of the Bible. In whatever manner the term “poor” became connected with the book, it is clear that the name “Biblia Pauperum” was not given to it in consequence of its being printed at a cheap rate for circulation among poor people. It is not indeed likely that its ancient title ever was “Biblia Pauperum;” while, on the contrary, there seems every reason to believe that Heineken had copied an abridged title and thus given currency to an error.

Heineken says that he observed the inscription, “Incipit Biblia Pauperum,” in a manuscript in the library at Wolfenbuttel, written on vellum in a Gothic character, which appeared to be of the fourteenth century. The figures, which were badly designed, were coloured in distemper, and the explanatory text was in Latin rhyme. It is surprising that neither Heineken nor any other bibliographer should have suspected that a word was wanting in the above supposed title, more especially as the word wanting might have been so readily suggested by another work so much resembling the pretended “Biblia Pauperum” that the one has 83 frequently been confounded with the other.II.60 In the proemium of this other work, which is no other than the “Speculum Salvationis,” the writer expressly states that he has compiled it “propter pauperes predicatores,”—for poor preachers.

Predictu’ p’hemiu’ hujus libri de conte’tis compilavi,

Et p’pter paup’es p’dicatores hoc apponere curavi;

Qui si forte nequieru’t totum librum sibi co’p’are,

Possu’t ex ipso p’hemio, si sciu’t p’dicare.

This preface of contents, stating what this book’s about,

For the sake of all poor preachers I have fairly written out;

If the purchase of the book entire should be above their reach,

This preface yet may serve them, if they know but how to preach.

That the other book might be called “Biblia Pauperum Predicatorum,” in consequence of its general use by mendicant preachers, I can readily believe; and no doubt the omission of the word “predicatorum” in the inscription copied by Heineken has given rise to the popular error, that the pretended “Biblia Pauperum” was a kind of cheap pictorial Bible, especially intended for the use of the poor. It is, in fact, a series of “skeleton sermons” ornamented with wood-cuts to warm the preacher’s imagination, and stored with texts to assist his memory. In speaking of this book in future, I shall always refer to it as the “Biblia Pauperum Predicatorum,”—“the Poor Preachers’ Bible;” for the continuance of its former title only tends, in my opinion, to disseminate an error.

Nyerup, who in 1784 published an “Account of such books as were read in schools in Denmark prior to the Reformation,”II.61 objected to the title “Biblia Pauperum,” as he had seen portions of a manuscript copy in which the drawings were richly coloured. The title which he preferred was Biblia Typico-Harmonica. In this objection, however, Camus does not concur: “It is not from the embellishments of a single copy,” he observes, “that we ought to judge of the current price of a book; and, besides, we must not forget to take into consideration the other motives which might suggest the title, ‘Bible of the Poor,’ for we have proofs that other abridgments of greater extent were called ‘Poor men’s books.’ Such is the ‘Biblia Pauperum’ of St. Bonaventure, consisting of extracts for the use of preachers, and the ‘Dictionarius Pauperum.’ Of the last the title is explained in the book itself: ‘Incipit summula omnibus verbi divini seminatoribus pernecessaria.’” It is surprising that Camus did 84 not perceive that the very titles which he cites militate against the opinion of the “Biblia” being intended for the use of poor men. St. Bonaventure’s work, and the Dictionary, which he refers to as instances of “Poor men’s books,” both bear on the very face of them a refutation of his opinion, for in the works themselves it is distinctly stated that they were compiled, not “ad usum pauperum hominum;” but “ad usum pauperum predicatorum, et verbi divini seminatorum:” not for the use of “poor men,” but for “poor preachers and teachers of the divine word.” Camus has unwittingly supplied a club to batter his own argument to pieces.

Of the “Biblia Pauperum Predicatorum,” there are, according to Heineken, five different editions with the text in Latin. Four of them contain each forty leaves, printed on one side only from wood-blocks by means of friction, and which differ from each other in so trifling a degree, that it is not unlikely that three of them are from the same set of blocks. The other edition,—the fifth described by Heineken—contains fifty leaves, printed in a similar manner, but apparently with the figures designed by a different artist. Besides the above, there are two different editions, also from wood-blocks, with the text in German: one with the date 1470; and the other, 1471 or 1475, for the last numeral appears as like a 1 as a 5. There are also two editions, one Latin, and the other German, with the text printed from moveable types by Albert Pfister, at Bamberg, about 1462.

Without pretending to decide on the priority of the first five editions,—as I have not been able to perceive any sufficient marks from which the order in which they were published might be ascertained,—I shall here give a brief account of a copy of that edition which Heineken ranks as the third. It is in the King’s Library at the British Museum, and was formerly in the collection of Monsieur Gaignat, at whose sale it was bought for George III.

It is a small folio of forty leaves, impressed on one side only, in order that the blank pages might be pasted together, so that two of the printed sides would thus form only one leaf. The order of the first twenty pages is indicated by the letters of the alphabet, from a to v, and of the second twenty by the same letters, having as a distinguishing mark a point both before and after them, thus: . a . In that which Heineken considers the first edition, the letters n, o, r, s, of the second alphabet, making pages 33, 34, 37, and 38, want those two distinguishing points, which, according to him, are to be found in each of the other three Latin editions of forty pages each. Mr. Ottley has, however, observed that Earl Spencer’s copy wants the points,—on each side of the letters n, o, r, s, of the second alphabet,—thus agreeing with that which Heineken calls the first edition, while in all other respects it answers the description which that writer 85 gives of the presumed second. Mr. Ottley says, that Heineken errs in asserting that the want of those points on each side of the said letters is a distinction exclusively belonging to the first edition, since the edition called by him the second is likewise without them.II.62 In fact, the variations noticed by Heineken are not only insufficient to enable a person to judge of the priority of the editions, but they are such as might with the greatest ease be introduced into a block after a certain number of copies had been taken off. Those which he considers as distinguishing marks might easily be broken away by the burnisher or rubber, and replaced by the insertion of other pieces, differing in a slight degree. From the trifling variations noticed by HeinekenII.63 in the first three editions, it is not unlikely that they were all taken from the same blocks. Each of the triangular ornaments in which he has observed a difference, might easily be re-inserted in the event of its being injured in taking an impression. The tiara of Moses, in page 35, letter . p . would be peculiarly liable to accident in taking an impression by friction, and I am disposed to think that a part of it has been broken off, and that in repairing it a trifling alteration has been made in the ornament on its top. Heineken, noticing the alteration, has considered it as a criterion of two different editions, while in all probability it only marks a trifling variety in copies taken from the same blocks.

On each page are four portraits,—two at the top, and two at the bottom,—intended for the prophets, and other holy men, whose writings are cited in the text. The middle part of the page between each pair of portraits consists of three compartments, each of which is occupied with a subject from the Old or the New Testament. In the 14th page, however, letter o, two of the compartments—that in the centre, and the adjoining one to the right—are both occupied by the same subject, Christ’s entry into Jerusalem. The greatest portion of the explanatory text is at the top on each side of the uppermost portraits; and on each side of those below there is a Leonine, or rhyming Latin, verse. A similar verse underneath those portraits forms the concluding line of each page. Texts of Scripture, and moral or explanatory sentences, having reference to the subjects in the three compartments, also appear on scrolls. The following cut, which is a reduced copy of the 14th page, letter k, will afford a better idea of the arrangement of the subjects, and of the explanatory texts, than any lengthened description.

The whole of this subject—both text and figures—appears intended to inculcate the necessity of restraining appetite. The inscription to the right, at the top, contains a reference to the 3rd chapter of Genesis, wherein there is to be found an account of the temptation and fall of 86 Adam and Eve, who were induced by the Serpent to taste the forbidden fruit. This temptation of our first parents through the medium of the palate, was, as may be gathered from the same inscription, figurative of the temptation of Christ after his fasting forty days in the wilderness, when the Devil came to him and said, “If thou be the Son of God, command that these stones be made bread.”

  see text

In the inscription to the left, reference is made to the 25th chapter of Genesis, as containing an account of Esau, who, in consequence of his unrestrained appetite, sold his birth-right for a mess of pottage.

In the compartments in the middle of the page, are three illustrations of the preceding text. In the centre is seen the pattern to imitate,—Christ resisting the temptation of the Devil; and on each side the examples to deter,—Adam and Eve with the forbidden fruit; and hungry Esau receiving the mess of pottage from Jacob.

87

Underneath the two half-length figures at the top, is inscribed “David 34,” and “Ysaie xxix.”II.64 The numerals are probably intended to indicate the chapters in the Psalms, and in the Prophecies of Isaiah, where the inscriptions on the adjacent scrolls are to be found. On similar scrolls, towards the bottom of the page, are references to the 7th chapter of the 2nd book of Kings, and to the 16th chapter of Job. The two half-length figures are most likely intended for the writers of those sacred books. The likenesses of the prophets and holy persons, thus introduced at the top and bottom of each page, are, as Schelhorn has observed,II.65 purely imaginary; for the same character is seldom seen twice with the same face. As most of the supposed figurative descriptions of Christ and his ministry are to be found in the Psalms, and in the Prophecies of Isaiah, the portraits of David and the last-named prophet are those which most frequently occur; and the designer seems to have been determined that neither the king nor the prophet should ever appear twice with the same likeness.

The rhyming verses are as follows. That to the right, underneath the subject of Adam and Eve:

Serpens vicit, Adam vetitam sibi sugerat escam.

The other, on the opposite side, underneath Jacob and Esau:

Lentis ob ardorem proprium male perdit honorem.

And the third, at the bottom of the page, underneath the two portraits:

Christum temptavit Sathanas ut eum superaret.

The following cuts are fac-similes, the size of the originals, of each of the compartments of the page referred to, and of which a reduced copy has been already given.

The first contains the representation of David and Isaiah, and the characters which follow the name of the former I consider to be intended for 34. They are the only instances in the volume of the use of Arabic, or rather Spanish numerals. The letter k, at the foot, is the “signature,” as a printer would term it, indicating the order of the page. On each side of it are portions of scrolls containing inscriptions, of which some of the letters are seen.

  see text

The next cut represents Satan tempting Christ by offering him stones to be converted into bread.

88

In the distance are seen the high mountain, to the top of which Christ was taken up by the Devil, and the temple from whose pinnacle Christ was tempted to cast himself down. The figure of Christ in this compartment is not devoid of sober dignity; nor is Satan deficient in diabolical ugliness; but, though clawed and horned proper, he wants the usual appendage of a tail. The deficiency is, however, in some degree compensated by giving to his hip the likeness of a fiendish face. In two or three other old wood engravings I have noticed a repulsive face indicated in a similar manner on the hip of the Devil. A person well acquainted with the superstitions of the fourteenth and fifteenth centuries may perhaps be able to give a reason for this. It may be intended to show that Satan, who is ever going about seeking whom he may devour, can see both before and behind.

  see text

The cut on the following page (90), which forms the compartment to the right, represents Adam and Eve, each with an apple: and the state in which Eve appears to be, is in accordance with an opinion maintained by several of the schoolmen of the twelfth and thirteenth centuries. The tree of knowledge is without fruit, and the serpent, with a human face, is seen twined round its stem. The form of the tree and the shape of the leaves are almost precisely the same as those of the olive-trees in the Apocalypse, uprooted by Antichrist. The character of the designs, however, in the two books is almost as different as the manner of the engraving. In the Apocalypse there is no attempt at shading, while in 89 the book under consideration it is introduced in every page, though merely by courses of single lines, as may be perceived in the drapery of Christ in the preceding cut, and in the trunk of the tree and in the serpent in the cut subjoined. In this cut the figure of Adam cannot be considered as a specimen of manly beauty; his face is that of a man who is past his prime, and his attitude is very like that of one of the splay-footed boors of Teniers. In point of personal beauty Eve appears to be a partner worthy of her husband; and though from her action she seems conscious that she is naked, yet her expression and figure are extremely unlike the graceful timidity and beautiful proportions of the Medicean Venus. The face of the serpent displays neither malignity nor fiendish cunning; but, on the contrary, is marked with an expression not unlike that of a Bavarian broom-girl. This manner of representing the temptation of our first parents appears to have been conventional 90 among the early German Formschneiders; for I have seen several old wood-cuts of this subject, in which the figures were almost precisely the same. Notwithstanding the bad drawing and the coarse engraving of the following cut, many of the same subject, executed in Germany between 1470 and 1510, are yet worse.

  see text

In the opposite cut, which forms the compartment to the left, Esau, who is distinguished by his bow and quiver, is seen receiving a bowl of pottage from his brother Jacob. At the far side of the apartment is seen a “kail-pot,” suspended from a “crook,” with something like a ham and a gammon of bacon hanging against the wall. This subject is treated in a style which is thoroughly Dutch. Isaac’s family appear to 91 have been lodged in a tolerably comfortable house, with a stock of provisions near the chimney nook; and his two sons are very like some of the figures in the pictures of Teniers, more especially about the legs.

  see text

The following cut, a copy of that which is the lowest in the page, represents the two prophets or inspired penmen, to whom reference is made on the two scrolls whose ends may be perceived towards the lower corners of each arch. The words underneath the figures are a portion of the last rhyming verse quoted at page 87. It is from a difference in the triangular ornament, above the pillar separating the two figures, though not in this identical page, that Heineken chiefly decides on three of the editions of this book; though nothing could be more easy than to 92 introduce another ornament of a similar kind, in the event of the original either being damaged in printing or intentionally effaced. In some of the earliest wood-blocks which remain undestroyed by the rough handling of time there are evident traces of several letters having been broken away, and of the injury being afterwards remedied by the introduction of a new piece of wood, on which the letters wanting were re-engraved.

  see text

The ink with which the cuts in the “Poor Preachers’ Bible” have been printed, is evidently a kind of distemper of the colour of bistre, lighter than in the History of the Virgin, and darker than in the Apocalypse. In many of the cuts certain portions of the lines appear surcharged with ink,—sometimes giving to the whole page rather a blotched appearance,—while other portions seem scarcely to have received any.II.66 This appearance is undoubtedly in consequence of the light-bodied ink having, from its want of tenacity, accumulated on the block where the line was thickest, or where two lines met, leaving the thinner portions adjacent with scarce any colouring at all. The block must, in my opinion, have been charged with such ink by means of something like a brush, and not by means of a ball. In some parts of the cuts—more especially where there is the greatest portion of text—small 93 white spaces may be perceived, as if a graver had been run through the lines. On first noticing this appearance, I was inclined to think that it was owing to the spreading of the hairs of the brush in inking, whereby certain parts might have been left untouched. The same kind of break in the lines may be observed, however, in some of the impressions of the old wood-cuts published by Becker and Derschau,II.67 and which are worked off by means of a press, and with common printer’s ink. In these it is certainly owing to minute furrows in the grain of the wood; and I am now of opinion that the same cause has occasioned a similar appearance in the cuts of the “Biblia Pauperum Predicatorum.” Mr. Ottley, speaking of the impressions in Earl Spencer’s copy, makes the following remarks: “In many instances they have a sort of horizontally striped and confused appearance, which leads me to suppose that they were taken from engravings executed on some kind of wood of a coarse grain.”II.68 This correspondence between Earl Spencer’s copy and that in the King’s Library at the British Museum tends to confirm my opinion that there are not so many editions of the book as Heineken,—from certain accidental variations,—has been induced to suppose.

The manner in which the cuts are engraved, and the attempts at something like effect in the shading and composition, induce me to think that this book is not so old as either the Apocalypse or the History of the Virgin. That it appeared before 1428, as has been inferred from the date which the Rev. Mr. Horne fancied that he had seen on the ancient binding, I cannot induce myself to believe. It is more likely to have been executed at some time between 1440 and 1460; and I am inclined to think that it is the production of a Dutch or Flemish, rather than a German artist.

A work, from which the engraved “Biblia Pauperum Predicatorum” is little more than an abstract, appears to have been known in France and Germany long before block-printing was introduced. Of such a work there were two manuscript copies in the National Library at Paris; the one complete, and the other—which, with a few exceptions, had been copied from the first—imperfect. The work consisted of a brief summary of the Bible, arranged in the following manner. One or two phrases in Latin and in French formed, as it were, the text; and each text was followed by a moral reflection, also in Latin and in French. Each 94 article, which thus consisted of two parts, was illustrated by two drawings, one of which related to the historical fact, and the other to the moral deduced from it. The perfect copy consisted of four hundred and twenty-two pages, on each of which there were eight drawings, so that the number contained in the whole volume was upwards of five thousand. In some of the single drawings, which were about two and one-third inches wide, by three and one-third inches high, Camus counted not less than thirty heads.II.69

In a copy of the “Biblia Pauperum Predicatorum” from wood-blocks, Heineken observed written: “S. Ansgarius est autor hujus libri,”—St. Ansgarius is the author of this book. St. Ansgarius, who was a native of France, and a monk of the celebrated Abbey of Corbey, was sent into Lower Saxony, and other places in the north, for the purpose of reclaiming the people from paganism. He was appointed the first bishop of Hamburg in 831, and in 844 Bishop of Bremen, where he died in 864.II.70 From a passage cited by Heineken from Ornhielm’s Ecclesiastical History of Sweden and Gothland, it appears that Ansgarius was reputed to have compiled a similar book;II.71 and Heineken observes that it might be from this passage that the “Biblia Pauperum Predicatorum” was ascribed to the Bishop of Hamburg.

In the cloisters of the cathedral at Bremen, Heineken saw two bas-reliefs sculptured on stone, of which the figures, of a moderate size, were precisely the same as those in two of the pages—the first and eighth—of the German “Biblia Pauperum Predicatorum.” The inscriptions, which were in Latin, were the same as in block-book. He thinks it very probable that the other arches of the cloisters were formerly ornamented in the same manner with the remainder of the subjects, but that the sculptures had been destroyed in the disturbances which had occurred in Bremen. Though he by no means pretends that the cuts were engraved in the time of Ansgarius, he thinks it not impossible that the sculptures might be executed at that period according to the bishop’s directions. This last passage is one of the most silly that occurs in Heineken’s book.II.72 It is just about as likely that the cuts in the “Biblia Pauperum Predicatorum” were engraved in the time of Ansgarius, as that the bas-reliefs in the cloisters of the cathedral of Bremen should have been sculptured under his direction.

95

The book usually called the “Speculum Humanæ Salvationis,”II.73—the Mirror of Human Salvation,—which is ascribed by Hadrian Junius to Lawrence Coster, has been more frequently the subject of discussion among bibliographers and writers who have treated of the origin of printing, than any other work. A great proportion, however, of what has been written on the subject consists of groundless speculation; and the facts elicited, compared with the conjectures propounded, are as “two grains of wheat to a bushel of chaff.” It would be a waste of time to recite at length the various opinions that have been entertained with respect to the date of this book, the manner in which the text was printed, and the printer’s name. The statements and the theories put forth by Junius and Meerman in Coster’s favour, so far as the execution of the Speculum is concerned, are decidedly contradicted by the book itself. Without, therefore, recapitulating arguments which are contradicted by established facts, I shall endeavour to give a correct account of the work, leaving those who choose to compare it, and reconcile it if they can, with the following assertions made by Coster’s advocates: 1. that the Speculum was first printed by him in Dutch with wooden types; 2. that while engraving a Latin edition on blocks of wood he discovered the art of printing with moveable letters; 3. that the Latin edition, in which the text is partly from moveable types and partly from wood-blocks, was printed by Coster’s heirs and successors, their moveable types having been stolen by John Gutemberg before the whole of the text was set up.

The Speculum which has been the subject of so much discussion is of a small folio size, and without date or printer’s name. There are four editions of it known to bibliographers, all containing the same cuts; two of those editions are in Latin, and two in Dutch. In the Latin editions the work consists of sixty-three leaves, five of which are occupied by an introduction or prologue, and on the other fifty-eight are printed the cuts and explanatory text. The Dutch editions, though containing the same number of cuts as the Latin, consist of only sixty-two leaves each, as the preface occupies only four. In all those editions the leaves are printed on one side only. Besides the four editions above noticed, which have been ascribed to Coster and have excited so much controversy, there are two or three others in which the cuts are more coarsely engraved, and probably executed, at a later period, in Germany. There is also a quarto edition of the Speculum, printed in 1483, at Culemburg, by John Veldener, and ornamented with the identical cuts of the folio editions ascribed to Coster and his heirs.

The four controverted editions of the Speculum may be considered as holding a middle place between block-books,—which are wholly executed, 96 both text and cuts, by the wood-engraver,—and books printed with moveable types: for in three of the editions the cuts are printed by means of friction with a rubber or burnisher, in the manner of the History of the Virgin, and other block-books, while the text, set in moveable type, has been worked off by means of a press; and in a fourth edition, in which the cuts are taken in the same manner as in the former, twenty pages of the text are printed from wood-blocks by means of friction, while the remainder are printed in the same manner as the whole of the text in the three other editions; that is, from moveable metal types, and by means of a press.

There are fifty-eight cuts in the Speculum, each of which is divided into two compartments by a slender column in the middle. In all the editions the cuts are placed as head-pieces at the top of each page, having underneath them, in two columns, the explanatory text. Under each compartment the title of the subject, in Latin, is engraved on the block.

The following reduced copy of the first cut will give an idea of their form, as every subject has pillars at the side, and is surmounted by an arch in the same style.

  see text

The style of engraving in those cuts is similar to those of the Poor Preachers’ Bible. The former are, however, on the whole executed with greater delicacy, and contain more work. The shadows and folds of the drapery in the first forty-eight cuts are indicated by short parallel lines, which are mostly horizontal. In the forty-ninth and subsequent cuts, as has been noticed by Mr. Ottley, a change in the mode of indicating the shades and the folds in the draperies is perceptible; for the short parallel lines, instead of being horizontal as in the former, are mostly slanting. Heineken observes, that to the forty-eighth cut inclusive, the chapters in the printed work are conformable with the old Latin manuscripts; and 97 as a perceptible change in the execution commences with the forty-ninth, it is not unlikely that the cuts were engraved by two different persons. The two following cuts are fac-similes of the compartments of the first, of which a reduced copy has been previously given.

  see text

In the above cut, its title, “Casus Luciferi,”—the Fall of Lucifer,—is engraved at the bottom; and the subject represented is Satan and the rebellious angels driven out of heaven, as typical of man’s disobedience and fall. The following are the first two lines of the column of text underneath the cut in the Latin editions:

Inchoatur speculum humanae salvacionis

In quo patet casus hominis et modus repactionis.

Which may be translated into English thus:

In the Mirror of Salvation here is represented plain

The fall of man, and by what means he made his peace again.

The following is the right-hand compartment of the same cut. The 98 title of this subject, as in all the others, is engraved at the bottom; the contracted words when written in full are, “Deus creavit hominem ad ymaginem et similitudinem suam,”—God created man after his own image and likeness.

  see text

The first two lines of the text in the column underneath this cut are,

Mulier autem in paradiso est formata

De costis viri dormienti est parata.

That is, in English rhyme of similar measure,

The woman was in Paradise for man an help meet made,

From Adam’s rib created as he asleep was laid.

The cuts in all the editions are printed in light brown or sepia colour which has been mixed with water, and readily yields to moisture. The impressions have evidently been taken by means of friction, as the back of the paper immediately behind is smooth and shining from the action of the rubber or burnisher, while on the lower part of the page at the back 99 of the text, which has been printed with moveable types, there is no such appearance. In the second Latin edition, in which the explanatory text to twenty of the cutsII.74 has been printed from engraved wood-blocks by means of friction, the reverse of those twenty pages presents the same smooth appearance as the reverse of the cuts. In those twenty pages of text from engraved wood-blocks the ink is lighter-coloured than in the remainder of the book which is printed from moveable types, though much darker than that of the cuts. It is, therefore, evident that the two impressions,—the one from the block containing the cut, and the other from the block containing the text,—have been taken separately. In the pages printed from moveable types, the ink, which has evidently been compounded with oil, is full-bodied, and of a dark brown colour, approaching nearly to black. In the other three editions, one Latin and two Dutch, in which the text is entirely from moveable types, the ink is also full-bodied and nearly jet black, forming a strong contrast with the faint colour of the cuts.

The plan of the Speculum is almost the same as that of the Poor Preachers’ Bible, and is equally as well entitled as the latter to be called “A History typical and anti-typical of the Old and New Testament.” Several of the subjects in the two books are treated nearly in the same manner, though in no single instance, so far as my observation goes, is the design precisely the same in both. In several of the cuts of the Speculum, in the same manner as in the Poor Preachers’ Bible, one compartment contains the supposed type or prefiguration, and the other its fulfilment; for instance: at No. 17 the appearance of the Lord to Moses in the burning bush is typical of the Annunciation; at No. 23 the brazen bath in the temple of Solomon is typical of baptism; at No. 31 the manna provided for the children of Israel in the Desert is typical of the Lord’s Supper; at No. 45 the Crucifixion is represented in one compartment, and in the other is Tubal-Cain, the inventor of iron-work, and consequently of the nails with which Christ was fixed to the cross; and at No. 53 the descent of Christ to Hades, and the liberation of the patriarchs and fathers, is typified by the escape of the children of Israel from Egypt.

Though most of the subjects are from the Bible or the Apocrypha, yet there are two or three which the designer has borrowed from profane history: such as Semiramis contemplating the hanging gardens of Babylon; the Sibyl and Augustus; and Codrus king of Athens incurring death in order to secure victory to his people.

The Speculum Salvationis, as printed in the editions previously noticed, is only a portion of a larger work with the same title, and 100 ornamented with similar designs, which had been known long before in manuscript. Heineken says, at page 478 of his Idée Générale, that the oldest copy he ever saw was in the Imperial Library at Vienna; and, at page 468, he observes that it appeared to belong to the twelfth century.

The manuscript work, when complete, consisted of forty-five chapters in rhyming Latin, to which was prefixed an introduction containing a list of them. Each of the first forty-two chapters contained four subjects, the first of which was the principal, and the other three illustrative of it. To each of these chapters were two drawings, every one of which, as in the printed copies of the work, consisted of two compartments. The last three chapters contained each eight subjects, and each subject was ornamented with a design.II.75 The whole number of separate illustrations in the work was thus one hundred and ninety-two. The printed folio editions contain only fifty-eight cuts, or one hundred and sixteen separate illustrations.

Though the Speculum from the time of the publication of Junius’s workII.76 had been confidently claimed for Coster, yet no writer, either for or against him, appears to have particularly directed his attention to the manner in which the work was executed before Fournier, who in 1758, in a dissertation on the Origin and Progress of the Art of Wood-engraving,II.77 first published some particulars respecting the work in question, which induced Meerman and Heineken to speculate on the priority of the different editions. Mr. Ottley, however, has proved, in a manner which carries with it the certainty of mathematical demonstration, that the conjectures of both the latter writers respecting the priority of the editions of the Speculum are absolutely erroneous. To elicit the truth does not, with respect to this work, seem to have been the object of those two writers. Both had espoused theories on its origin without much inquiry with respect to facts, and each presumed that edition to be the first which seemed most likely to support his own speculations.

Heineken, who assumed that the work was of German origin, insisted that the first edition was that in which the text is printed partly from moveable types and partly from letters engraved on wood-blocks, and that the Dutch editions were executed subsequently in the Low Countries. The Latin edition with the text entirely printed from moveable types he is pleased to denominate the second, and to assert, contrary to the evidence which the work itself affords, that the type resembles that of Faust and 101 Scheffer, and that the cuts in this second Latin edition, as he erroneously calls it, are coarser and not so sharp as those in the Latin edition which he supposes to be the first.

Fournier’s discoveries with respect to the execution of the Speculum seem to have produced a complete change as to its origin in the opinions of Meerman; who, in 1757, the year before Fournier’s dissertation was printed, had expressed his belief, in a letter to his friend Wagenaar, that what was alleged in favour of Coster being the inventor of printing was mere gratuitous assertion; that the text of the Speculum was probably printed after the cuts, and subsequent to 1470; that there was not a single document, nor an iota of evidence, to show that Coster ever used moveable types; and lastly, that the Latin was prior to the Dutch edition of the Speculum, as was apparent from the Latin names engraved at the foot of the cuts, which certainly would have been in Dutch had the cuts been originally destined for a Dutch edition.II.78 In the teeth of his own previous opinions, having apparently gained a new light from Fournier’s discoveries, Meerman, in his Origines Typographicæ, printed in 1765, endeavours to prove that the Dutch edition was the first, and that it was printed with moveable wooden types by Coster. The Latin edition in which the text is printed partly from moveable types and partly from wood-blocks he supposes to have been printed by Coster’s heirs after his decease, thus endeavouring to give credibility to the story of Coster having died of grief on account of his types being stolen, and to encourage the supposition that his heirs in this edition supplied the loss by having engraved on blocks of wood those pages which were not already printed.

Fournier’s discoveries relative to the manner in which the Speculum was executed were: 1st, that the cuts and the text had been printed at separate times, and that the former had been printed by means of friction; 2d, that a portion of the text in one of the Latin editions had been printed from engraved wood-blocks.II.79 Fournier, who was a type-founder and wood-engraver, imagined that the moveable types with which the Speculum was printed were of wood. He also asserted that Faust and Scheffer’s Psalter and an early edition of the Bible were printed with moveable wooden types. Such assertions are best 102 answered by a simple negative, leaving the person who puts them forth to make out a probable case.

The fact having been established that in one of the editions of the Speculum a part of the text was printed from wood-blocks, while the whole of the text in the other three was printed from moveable types, Heineken, without diligently comparing the editions with each other in order to obtain further evidence, decides in favour of that edition being the first in which part of the text is printed from wood-blocks. His reasons for supposing this to be the first edition, though specious in appearance, are at variance with the facts which have since been incontrovertibly established by Mr. Ottley, whose scrutinizing examination of the different editions has clearly shown the futility of all former speculations respecting their priority. The argument of Heineken is to this effect: “It is improbable that a printer who had printed an edition wholly with moveable types should afterwards have recourse to an engraver to cut for him on blocks of wood a portion of the text for a second edition; and it is equally improbable that a wood-engraver who had discovered the art of printing with moveable types, and had used them to print the entire text of the first edition, should, to a certain extent, abandon his invention in a second by printing a portion of the text from engraved blocks of wood.” The following is the order in which he arranges the different editions:

1. The Latin edition in which part of the text is printed from wood-blocks.

2. The Latin edition in which the text is entirely printed from moveable types.

3. The Dutch edition with the text printed wholly from moveable types, supposed by Meerman to be the first edition of all.II.80

4. The Dutch edition with the text printed wholly from moveable types, and which differs only from the preceding one in having the two pages of text under cuts No. 45 and 56 printed in a type different from the rest of the book.

The preceding arrangement—including Meerman’s opinion respecting the priority of the Dutch edition—rests entirely on conjecture, and is almost diametrically contradicted in every instance by the evidence afforded by the books themselves; for through the comparisons and investigations of Mr. Ottley it is proved, to an absolute certainty, that the Latin edition supposed by Heineken to be the second is the earliest of all; that the edition No. 4, called the second Dutch, is the next in order to the actual first Latin; and that the two editions, No. 1 and No. 3, respectively proclaimed by Heineken and Meerman as the earliest, 103 have been printed subsequently to the other two.II.81 Which of the pretended first editions was in reality the last, has not been satisfactorily determined; though there seems reason to believe that it was the Latin one which has part of the text printed from wood-blocks.

It is well known to every person acquainted with the practice of wood-engraving, that portions of single lines in such cuts as those of the Speculum are often broken out of the block in the process of printing. If two books, therefore, containing the same wood-cuts, but evidently printed at different times, though without a date, should be submitted to the examination of a person acquainted with the above fact and bearing it in mind, he would doubtless declare that the copy in which the cuts were most perfect was first printed, and that the other in which parts of the cuts appeared broken away was of a later date. If, on comparing other copies of the same editions he should find the same variations, the impression on his mind as to the priority of the editions would amount to absolute certainty. The identity of the cuts in all the four editions of the Speculum being unquestionable, and as certain minute fractures in the lines of some of them, as if small portions of the block had been broken out in printing, had been previously noticed by Fournier and Heineken, Mr. Ottley conceived the idea of comparing the respective cuts in the different editions, with a view of ascertaining the order in which they were printed. He first compared two copies of the edition called the first Latin with a copy of that called the second Dutch, and finding, that, in several of the cuts of the former, parts of lines were wanting which in the latter were perfect, he concluded that the miscalled second Dutch edition was in fact of an earlier date than the pretended first Latin edition of Heineken. In further comparing the above editions with the supposed second Latin edition of Heineken and the supposed first Dutch edition of Meerman, he found that the cuts in the miscalled second Latin edition were the most perfect of all; and that the cuts in Heineken’s first Latin and Meerman’s first Dutch editions contained more broken lines than the edition named by those authors the second Dutch. The conclusion which he arrived at from those facts was irresistible, namely, that the earliest edition of all was that called by Heineken the second Latin; and that the edition called the second Dutch was the next in order. As the cuts in the copies examined of the pretended first Latin and Dutch editions contained similar fractures, it could not be determined with certainty which was actually the last.

104

As it is undoubted that the cuts of all the editions have been printed separately from the text, it has been objected that Mr. Ottley’s examination has only ascertained the order in which the cuts have been printed, but by no means decided the priority of the editions of the entire book. All the cuts, it has been objected, might have been taken by the engraver before the text was printed in a single edition, and it might thus happen that the book first printed with text might contain the last, and consequently the most imperfect cuts. This exception, which is founded on a very improbable presumption, will be best answered by the following facts established on a comparison of the two Latin, and which, I believe, have not been previously noticed:—On closely comparing those pages which are printed with moveable types in the true second edition with the corresponding pages in that edition which is properly the first, it was evident from the different spelling of many of the words, and the different length of the lines, that they had been printed at different times: but on comparing, however, those pages which are printed in the second edition from engraved wood-blocks with the corresponding pages, from moveable type, in the first edition, I found the spelling and the length of the lines to be the same. The page printed from the wood-block was, in short, a fac-simile of the corresponding page printed from moveable types. So completely did they correspond, that I have no doubt that an impression of the page printed from moveable types had been “transferred,”II.82 as engravers say, to the block. In the last cutII.83 of the first edition I noticed a scroll which was quite black, as if meant to contain an inscription which the artist had neglected to engrave; and in the second edition I perceived that the black was cut away, thus having the part intended for the inscription white. Another proof, in addition to those adduced by Mr. Ottley of that Latin edition being truly the first in which the whole of the text is printed from moveable types.

Though there can no longer be a doubt in the mind of any impartial person of that Latin edition, in which part of the text is printed from engraved wood-blocks, and the rest from moveable types, being later than the other; yet the establishment of this fact suggests a question, as to the cause of part of the text of this second Latin edition being printed from wood-blocks, which cannot perhaps be very satisfactorily answered. 105 All writers previous to Mr. Ottley, who had noticed that the text was printed partly from moveable types and partly from wood-blocks, decided, without hesitation, that this edition was the first; and each, accordingly as he espoused the cause of Gutemberg or Coster, proceeded to theorise on this assumed fact. As their arguments were founded in error, it cannot be a matter of surprise that their conclusions should be inconsistent with truth. The fact of this edition being subsequent to that in which the text is printed wholly from moveable types has been questioned on two grounds: 1st. The improbability that the person who had printed the text of a former edition entirely from moveable types should in a later edition have recourse to the more tedious operation of engraving part of the text on wood-blocks. 2d. Supposing that the owner of the cuts had determined in a later edition to engrave the text on blocks of wood, it is difficult to conceive what could be his reason for abandoning his plan, after twenty pages of the text were engraved, and printing the remainder with moveable types.

Before attempting to answer those objections, I think it necessary to observe that the existence of a positive fact can never be affected by any arguments which are grounded on the difficulty of accounting for it. Objections, however specious, can never alter the immutable character of truth, though they may affect opinions, and excite doubts in the minds of persons who have not an opportunity of examining and judging for themselves.

With respect to the first objection, it is to be remembered that in all the editions, the text, whether from wood-blocks or moveable types, has been printed separately from the cuts; consequently the cuts of the first edition might be printed by a wood-engraver, and the text set up and printed by another person who possessed moveable types. The engraver of the cuts might not be possessed of any moveable types when the text of the first edition was printed; and, as it is a well-known fact that wood-engravers continued to execute entire pages of text for upwards of thirty years after the establishment of printing with moveable types, it is not unlikely that he might attempt to engrave the text of a second edition and print the book solely for his own advantage. This supposition is to a certain extent corroborated by the fact of the twenty pages of engraved text in the second Latin edition being fac-similes of the twenty corresponding pages of text from moveable types in the first.

To the second objection every day’s experience suggests a ready answer; for scarcely anything is more common than for a person to attempt a work which he finds it difficult to complete, and, after making some progress in it, to require the aid of a kindred art, and abandon his original plan.

As the first edition of the Speculum was printed subsequent to the 106 discovery of the art of printing with moveable types, and as it was probably printed in the Low Countries, where the typographic art was first introduced about 1472, I can discover no reason for believing that the work was executed before that period. Santander, who was so well acquainted with the progress of typography in Belgium and Holland, is of opinion that the Speculum is not of an earlier date than 1480. In 1483 John Veldener printed at Culemburg a quarto edition of the Speculum, in which the cuts are the same as in the earlier folios. In order to adapt the cuts to this smaller edition Veldener had sawn each block in two, through the centre pillar which forms a separation between the two compartments in each of the original engravings. Veldener’s quarto edition, which has the text printed on both sides of the paper from moveable types, contains twelve more cuts than the older editions, but designed and executed in the same style.II.84 If Lawrence Coster had been the inventor of printing with moveable types, and if any one folio edition of the Speculum had been executed by him, we cannot suppose that Veldener, who was himself a wood-engraver, as well as a printer, would have been ignorant of those facts. He, however, printed two editions of the Fasciculus Temporum,—one at Louvain in 1476, and the other at Utrecht in 1480,—a work which contains a short notice of the art of printing being discovered at Mentz, but not a syllable concerning its discovery at Harlem by Lawrence Coster. The researches of Coster’s advocates have clearly established one important fact, though an unfortunate one for their argument; namely, that the Custos or Warden of St. Bavon’s was not known as a printer to one of his contemporaries. The citizens of Harlem, however, have still something to console themselves with: though Coster may not be the inventor of printing, there can be little doubt of Junius, or his editor, being the discoverer of Coster,—

“Est quoddam prodire tenus, si non datur ultra.”

There is in the Print Room of the British Museum a small volume of wood-cuts, which has not hitherto been described by any bibliographer, nor by any writer who has treated on the origin and progress of wood engraving. It appears to have been unknown to Heineken, Breitkopf, Von Murr, and Meerman; and it is not mentioned, that I am aware of, either by Dr. Dibdin or Mr. Douce, although it certainly was submitted to the inspection of the latter. It formerly belonged to the late Sir George Beaumont, by whom it was bequeathed to the Museum; but where he obtained it I have not been able to learn. It consists of an 107 alphabet of large capital letters, formed of figures arranged in various attitudes; and from the general character of the designs, the style of the engraving, and the kind of paper on which the impressions have been taken, it evidently belongs to the same period as the Poor Preachers’ Bible. There is only one cut on each leaf, the back being left blank as in most of the block-books, and the impressions have been taken by means of friction. The paper at the back of each cut has a shining appearance when held towards the light, in consequence of the rubbing which it has received; and in some it appears as if it had been blacked with charcoal, in the same manner that some parts of the cartoons were blacked which have been pricked through by the tapestry worker. The ink is merely a distemper or water-colour, which will partly wash out by the application of hot water, and its colour is a kind of sepia. Each leaf, which is about six inches high, by three and six-eighths wide, consists of a separate piece of paper, and is pasted, at the inner margin, on to a slip either of paper or parchment, through which the stitching of the cover passes. Whether the paper has been cut in this manner before or after that the impressions were taken, I am unable to determine.II.85

The greater part of the letter A is torn out, and in that which remains there are pin-marks, as if it had been traced by being pricked through. The letters S, T, and V are also wanting. The following is a brief description of the letters which remain. The letter B is composed of five figures, one with a pipe and tabor, another who supports him, a dwarf, an old man kneeling, and an old woman with a staff. C, a youthful figure rending open the jaws of a lion, with two grotesque heads like those of satyrs. D, a man on horseback, and a monk astride on a fiendish-looking monster. E, two grotesque heads, a figure holding the horn of one of them, and another figure stretching out a piece of cloth. F, a tall figure blowing a trumpet, and a youth beating a tabor, with an animal like a dog at their feet.II.86 G, David with Goliath’s head, and a figure stooping, who appears to kiss a flagellum. H, a figure opening the jaws of a dragon. I, a tall man embracing a woman. K, a female with a wreath, a youth kneeling, an old man on his knees, and a young man with his heels uppermost. [Engraved as a specimen at page 109.] L, a man with a long sword, as if about to pierce a figure reclining. [Engraved as 108 a specimen at page 110.] M, two figures, each mounted on a kind of monster; between them, an old man. N, a man with a sword, another mounted on the tail of a fish. O, formed of four grotesque heads. P, two figures with clubs. Q, formed of three grotesque heads, similar to those in O. R, a tall, upright figure, another with something like a club in his hand; a third, with his heels up, blowing a horn. X, composed of four figures, one of which has two bells, and another has one; on the shoulder of the upper figure to the right a squirrel may be perceived. Y, a figure with something like a hairy skin on his shoulder; another thrusting a sword through the head of an animal. Z, three figures; an old man about to draw a dagger, a youth lying down, and another who appears as if flying. [Engraved as a specimen at page 111.] The last cut is the ornamental flower, of which a copy is given at page 112.

In the same case with those interesting, and probably unique specimens of early wood engraving, there is a letter relating to them, dated 27th May, 1819, from Mr. Samuel Lysons to Sir George Beaumont, from which the following is an extract: “I return herewith your curious volume of ancient cuts. I showed it yesterday to Mr. Douce, who agrees with me that it is a great curiosity. He thinks that the blocks were executed at Harlem, and are some of the earliest productions of that place. He has in his possession most of the letters executed in copper, but very inferior to the original cuts. Before you return from the Continent I shall probably be able to ascertain something further respecting them.” What might be Mr. Douce’s reasons for supposing that those cuts were executed at Harlem I cannot tell; though I am inclined to think that he had no better foundation for his opinion than his faith in Junius, Meerman, and other advocates of Lawrence Coster, who unhesitatingly ascribe every early block-book to the spurious “Officina Laurentiana.”

In the manuscript catalogue in the Print Room of the British Museum the volume is thus described by Mr. Ottley: “Alphabet of initial letters composed of grotesque figures, wood engravings of the middle of the fifteenth century, apparently the work of a Dutch or Flemish artist; the impressions taken off by friction in the manner of the early block-books. . . . I perceive the word ‘London’ in small characters written upon the blade of a sword in one of the cuts, [the letter L,] and I suspect they were engraved in England.”

  see text

As to whether these cuts were engraved in England or no I shall not venture to give an opinion. I am, however, satisfied that they were neither designed nor engraved by the artists who designed and engraved the cuts in the Apocalypse, the History of the Virgin, and the Poor Preachers’ Bible. With respect to drawing, expression, and engraving, the cuts of the Alphabet are decidedly superior to those of every block-book, 109 and generally to all wood engravings executed previous to 1500, with the exception of such as are by Albert Durer, and those contained in the Hypnerotomachia, an Italian rhapsody, with wood-cuts supposed to have been designed by Raffaele or Andrea Mantegna, and printed by Aldus at Venice, 1499. Although the cuts of the Alphabet may not have been engraved in England, it is, however, certain that the volume had been at rather an early period in the possession of an Englishman. The cover consists of a double fold of thick parchment, on the inside of which, between the folds, there is written in large old English characters what I take to be the name “Edwardus Lowes.” On the blank side of the last leaf there is a sketch of a letter commencing “Right reverent and wershipfull masters and frynds; In the moste loweliste maner that I canne or may, I here recomende me, duely glade to her of yor good 110 prosperitye and welth.” The writing, as I have been informed, is of the period of Henry VIII; and on the slips of paper and parchment to which the inner margins of the leaves are pasted are portions of English manuscripts, which are probably of the same date. There can, however, be little doubt that the leaves have been mounted, and the volume covered, about a hundred years subsequent to the engraving of the cuts.

  see text

I agree with Mr. Ottley in thinking that those cuts were engraved about the middle of the fifteenth century, but I can perceive nothing in them to induce me to suppose they were the work of a Dutch artist; and I am as little inclined to ascribe them to a German. The style of the drawing is not unlike what we see in illuminated French manuscripts of the middle of the fifteenth century; and as the only two engraved words which occur in the volume are French, I am rather inclined to suppose that the artist who made the drawings was a native of France. 111 The costume of the female to whom the words are addressed appears to be French; and the action of the lover kneeling seems almost characteristic of that nation. No Dutchman certainly ever addressed his mistress with such an air. He holds what appears to be a ring as gracefully as a modern Frenchman holds a snuff-box, and upon the scroll before him are engraved a heart, and the words which he may be supposed to utter, “Mon Ame.” At page 109, is a fac-simile of the cut referred to, the letter K, of the size of the original, and printed in the same kind of colour.

  see text

Upon the sword-blade in the original cut of the following letter, L, there is written in small characters, as Mr. Ottley has observed, the word “London;” and in the white space on the right, or upper side, of the figure lying down, there appears written in the same hand the name “Bethemsted.” In this name the letter B is not unlike a W; and I have heard it conjectured that the name might be that of 112 John Wethamstede, abbot of St. Alban’s, who was a great lover of books, and who died in 1440. This conjecture, however, will not hold good, for the letter is certainly intended for a B; and in the cut of the letter B there is written “R. Beths.,” which is in all probability intended for an abbreviation of the name, “Bethemsted,” which occurs in another part of the book. The ink with which these names are written is nearly of the same colour as that of the cuts. The characters appear to be of an earlier date than those on the reverse of the last leaf.

  see text

The cut at page 111, is that of the letter Z, which stands the wrong way in consequence of its not having been drawn reversed upon the block. The subject might at first sight be supposed to represent the angel staying Abraham when about to sacrifice Isaac; but on examining the cut more closely it will be perceived that the figure which might be mistaken for an angel is without wings, and appears to be in the act of supplicating the old man, who with his left hand holds him by the hair.

113

The opposite cut, which is the last in the book, is an ornamental flower designed with great freedom and spirit, and surpassing everything of the kind executed on wood in the fifteenth century. I speak not of the style of engraving, which, though effective, is coarse; but of the taste displayed in the drawing. The colour of the cuts on pages 109, 110, 111, from the late Sir George Beaumont’s book, will give the reader, who has not had an opportunity of examining the originals, some idea of the colour in which the cuts of the Apocalypse, the History of the Virgin, the Poor Preachers’ Bible, and the Speculum, are printed; which in all of them is a kind of sepia, in some inclining more to a yellow, and in others more to a brown.

In the volume under consideration we may clearly perceive that the art of wood engraving had made considerable progress at the time the cuts were executed. Although there are no attempts at cross-hatching, which was introduced about 1486, yet the shadows are generally well indicated, either by thickening the line, or by courses of short parallel lines, marking the folds of the drapery, or giving the appearance of rotundity to the figures. The expression of the heads displays considerable talent, and the wood-engraver who at the present time could design and execute such a series of figures, would be entitled to no small degree of commendation. Comparing those cuts with such as are to be seen in books typographically executed between 1461II.87 and 1490, it is surprising that the art of wood engraving should have so materially declined when employed by printers for the illustration of their books. The best of the cuts printed with letter-press in the period referred to are decidedly inferior to the best of the early block-books.

As it would occupy too much space, and would be beyond the scope of the present treatise to enter into a detail of the contents of all the block-books noticed by Heineken, I shall give a brief description of that named “Ars Memorandi,” and conclude the chapter with a list of such others as are chiefly referred to by bibliographers.

The “Ars Memorandi” is considered by SchelhornII.88 and by Dr. Dibdin as one of the earliest block-books, and in their opinion I concur. Heineken, however,—who states that the style is almost the same as in the figures of the Apocalypse,—thinks that it is of later date than the Poor Preachers’ Bible and the History of the 114 Virgin. It is of a quarto size, and consists of fifteen cuts, with the same number of separate pages of text also cut on wood, and printed on one side of each leaf only by means of friction.II.89 At the foot of each page of text is a letter of the alphabet, commencing with a, indicating the order in which they are to follow each other. In every cut an animal is represented,—an eagle, an angel, an ox, or a lion,—emblematic of the Evangelist whose Gospel is to be impressed on the memory. Each of the animals is represented standing upright, and marked with various signs expressive of the contents of the different chapters. To the Gospel of St. John, with which the book commences, three cuts with as many pages of text are allotted. St. Matthew has five cuts, and five pages of text. St. Mark three cuts and three pages of text; and St. Luke four cuts and four pages of text.II.90

“It is worthy of observation,” says J. C. Von Aretin, in his Essay on the earliest Results of the Invention of Printing, “that this book, which the most intelligent bibliographers consider to be one of the earliest of its kind, should be devoted to the improvement of the memory, which, though divested of much of its former importance by the invention of writing, was to be rendered of still less consequence by the introduction of printing.”II.91

The first cut is intended to express figuratively the first six chapters of St. John’s Gospel. The upright eagle is the emblem of the saint, and the numerals are the references to the chapters. The contents of the first chapter are represented by the dove perched on the eagle’s head, and the two faces,—one of an old, the other of a young man,—probably intended for those of Moses and Christ.II.92 The lute on the breast of the eagle, with something like three bellsII.93 suspended from it, indicate the contents of the second chapter, and are supposed by Schelhorn to refer to the marriage of Cana. The numeral 3, in Schelhorn’s opinion, relates to “nonnihil apertum et prosectum circa ventrem,” which he thinks may be intended as a reference to the words of Nicodemus: “Nunquid homo senex potest in ventrem matris suæ 115 iterum introire et renasci?” Between the feet of the eagle is a water-bucket surmounted by a sort of coronet or crown, intended to represent the principal events narrated in the 4th chapter, which are Christ’s talking with the woman of Samaria at the well, and his healing the son of a nobleman at Capernaum. The 5th chapter is indicated by a fish above the eagle’s right wing, which is intended to bring to mind the pool of Bethesda. The principal event related in the 6th chapter, Christ feeding the multitude, is indicated by the two fishes and five small loaves above the eagle’s left wing. The cross within a circle, above the fishes, is emblematic of the consecrated wafer in the Lord’s supper, as celebrated by the church of Rome.II.94

  see text

The above reduced copy of the cut will afford some idea of the manner in which the memory is to be assisted in recollecting the first six chapters of St. John. Those who wish to know more respecting this curious book are referred to Schelhorn’s Amœnitates Literariæ, tom. i. 116 pp. 1-17; Heineken, Idée Générale, pp. 394, 395; and to Dr. Dibdin’s Bibliotheca Spenceriana, vol. i. p. 4, where a copy is given of the first cut relating to the Gospel of St. Matthew.

Block-books containing both text and figures were executed long after the introduction of typography, or printing by means of moveable types; but the cuts in such works are decidedly inferior to those executed at an earlier period. The book entitled “Die Kunst Cyromantia,”II.95 which consists chiefly of text, is printed from wood-blocks on both sides of each leaf by means of a press. At the conclusion of the title is the date 1448; but this is generally considered to refer to the period when the book was written, and not the time when it was engraved. On the last page is the name: “jorg schapff zu augspurg.” If this George Schapff was a wood-engraver of Augsburg, the style of the cuts in the book sufficiently declares that he must have been one of the very lowest class. More wretched cuts were never chiselled out by a printer’s apprentice as a head-piece to a half-penny ballad.

Of the block-book entitled “Ars Moriendi,” Heineken enumerates no less than seven editions, of which one is printed on both sides of the leaves, and by means of a press. Besides these he mentions another edition, impressed on one side of the paper only, in which appear the following name and date: “Hans eporer, 1473, hat diss puch pruffmo er.”II.96

Of the book named in German “Der Entkrist”—Antichrist—printed from wood-blocks, Heineken mentions two editions. In that which he considers the first, containing thirty-nine cuts, each leaf is printed on one side only by means of friction; in the other, which contains thirty-eight cuts, is the “brief-maler’s” or wood-engraver’s name: “Der jung hanss priffmaler hat das puch zu nurenberg, 1472.”

At Nuremberg, in the collection of a physician of the name of Treu, Heineken noticed a small volume in quarto, consisting of thirty-two wood-cuts of Bible subjects, underneath each of which were fifteen verses in German, engraved on the same block. Each leaf was printed on one side only, and the impressions, which were in pale ink, had been taken by means of friction.

The early wood-engravers, besides books of cuts, executed others 117 consisting of text only, of which several portions are preserved in public libraries in Germany,II.97 France, and Holland; and although it is certain that block-books continued to be engraved and printed several years after the invention of typography, there can be little doubt that editions of the grammatical primer called the “Donatus,” from the name of its supposed compiler, were printed from wood-blocks previous to the earliest essays of Gutemberg to print with moveable types. It is indeed asserted that Gutemberg himself engraved, or caused to be engraved on wood, a “Donatus” before his grand invention was perfected.

In the Royal Library at Paris are preserved the two old blocks of a “Donatus” which are mentioned by Heineken at page 257 of his Idée Générale. They are both of a quarto form; but as the one contains twenty lines and the other only sixteen, and as there is a perceptible difference in the size of the letters, it is probable that they were engraved for different editions.II.98 Those blocks were purchased in Germany by a Monsieur Faucault, and after passing through the hands of three other book-collectors they came into the possession of the Duke de la Vallière, at whose sale they were sold for two hundred and thirty livres. In De Bure’s catalogue of the La Vallière library, impressions are given from the original blocks. The letters in both those blocks, though differing in size, are of the same proportions and form; and Heineken and Fischer consider that they bear a great resemblance to the characters of Faust and Scheffer’s Psalter, printed with moveable types in 1457, although the latter are considerably larger.

The art of wood engraving, having advanced from a single figure with merely a name cut underneath it, to the impression of entire pages of text, was now to undergo a change. Moveable letters formed of metal, and wedged together within an iron frame, were to supersede the engraved page; and impressions, instead of being taken by the slow and tedious process of friction, were now to be obtained by the speedy and powerful action of the press. If the art of wood engraving suffered a temporary decline for a few years after the general introduction of typography, it was only to revive again under the protecting influence of the PRESS; by means of which its productions were to be multiplied a hundred fold, and, instead of being confined to a few towns, were to be disseminated throughout every part of Europe.

II.1 A stencil is a piece of pasteboard, or a thin plate of metal, pierced with lines and figures, which are communicated to paper, parchment, or linen, by passing a brush charged with ink or colour over the stencil.

II.2 Cards—Carten—are mentioned in a book of bye-laws of Nuremberg, between 1380 and 1384. They are included in a list of games at which the burghers might indulge themselves, provided they ventured only small sums. “Awzgenommen rennen mit Pferder, Schiessen mit Armbrusten, Carten, Schofzagel, Pretspil, und Kugeln, umb einen pfenink zwen zu vier poten.” That is: “always excepting horse-racing, shooting with cross-bows, cards, shovel-board, tric-trac, and bowls, at which a man may bet from twopence to a groat.”—C. G. Von Murr, Journal zur Kunstsgesch. 2 Theil, S. 99.

II.3 In the town-books of Nuremberg a Hans Formansneider occurs so early as 1397, which De Murr says is not meant for “wood engraver,” but is to be read thus: Hans Forman, Schneider; that is, “Ihon Forman, maister-fashionere,” or, in modern phrase, “tailor.” The word “Karter” also occurs in the same year, but it is meant for a carder, or wool-comber, and not for a card-maker.—C. G. Von Murr, Journal, 2 Theil, S. 99.

II.4 “Conscioscia che l’arte e mestier delle carte & figure stampide, che se fano in Venesia è vegnudo a total deffaction, e questo sia per la gran quantità de carte a zugar, e fegure depente stampide, le qual vien fate de fuora de Venezia.” The curious document in which the above passage occurs was discovered by Temanza, an Italian architect, in an old book of rules and orders belonging to the company of Venetian painters. His discovery, communicated in a letter to Count Algarotti, appeared in the Lettere Pittoriche, tom. v. p. 320, et sequent. and has since been quoted by every writer who has written upon the subject.

II.5 This celebrated version, in the Mœso-Gothic language, is preserved in the library of Upsal in Sweden.

II.6 Osservazioni sulla Chirotipografia, ossia Antica Arte di Stampare a mano. Opera di D. Vincenzo Requeno. Roma 1810, 8vo.

II.7 Fuseli, at p. 85 of Ottley’s Inquiry; and Breitkopf, Versuch d. Ursprungs der Spielkarten Zu erforschen, 2 Theil, S. 175.

II.8 Fournier, Dissertation sur l’Origine et les Progrès de l’Art de Graver en Bois, p. 79; and Papillon, Traité de la Gravure en Bois, tom. i. p. 20, and Supplement, p. 80.

II.9 “Liber iste, Laus Virginis intitulatus, continet Lectiones Matutinales accommodatas Officio B. V. Mariæ per singulos anni dies,” &c. At the beginning of the volume is the following memorandum: “Istum librum legavit domna Anna filia domni Stephani baronis de Gundelfingen, canonica in Büchow Aule bte. Marie v’ginis in Buchshaim ord’is Cartusieñ prope Memingen Augusten. dyoc.”—Von Murr, Journal, 2 Theil, S. 104-105.

II.10 A fac-simile, of the size of the original, is given in Von Murr’s Journal, vol. ii. p. 104, and in Ottley’s Inquiry, vol. i. p. 90, both engraved on wood. There is an imitation engraved on copper, in Jansen’s Essai sur l’Origine de la Gravure, tom. i.

II.11 The following announcement appears in the colophon of the Nuremberg Chronicle. “Ad intuitum autem et preces providorum civium Sebaldi Schreyer et Sebastiani Romermaister hunc librum Anthonius Koberger Nurembergiæ impressit. Adhibitis tamen viris mathematicis pingendique arte peritissimis, Michaele Wolgemut et Wilhelmo Pleydenwurff, quorum solerti accuratissimaque animadversione tum civitatum tum illustrium virorum figuræ insertæ sunt. Consummatum autem duodecima mensis Julii. Anno Salutis ñre 1493.”

II.12 As great a neglect of the rules of perspective may be seen in several of the cuts in the famed edition of Theurdanck, Nuremberg, 1517, which are supposed to have been designed by Hans Burgmair, and engraved by Hans Schaufflein.

II.13 See Brand’s Popular Antiquities, vol. i. pp. 359-364.—Bohn’s edition.

II.14 Bibliographical and Picturesque Tour, by the Rev. T. F. Dibdin, D.D. p. 58, vol. ii. second edition, 1829. The De Murr to whom Dr. Dibdin alludes, is C. G. Von Murr, editor of the Journal of Arts and General Literature, published at Nuremberg in 1775 and subsequent years. Von Murr was the first who published, in the second volume of his journal, a fac-simile, engraved on wood by Sebast. Roland, of the Buxheim St. Christopher, from a tracing sent to him by P. Krismer, the librarian of the convent. Von Murr, in his Memorabilia of the City of Nuremberg, mentions that Breitkopf had seen a duplicate impression of the Buxheim St. Christopher in the possession of M. De Birkenstock at Vienna.

II.15 There is every reason in the world to suppose that this wood-cut was executed either in Nuremberg or Augsburg. Buxheim is situated almost in the very heart of Suabia, the circle in which we find the earliest wood engravers established. Buxheim is about thirty English miles from Ulm, forty-four from Augsburg, and one hundred and fifteen from Nuremberg. Von Murr does not notice the pretensions of Ulm, which on his own grounds are stronger than those of his native city, Nuremberg.

II.16 Von Murr, Journal, 2 Theil, S. 105, 106.

II.17 St. Bridget was a favourite saint in Germany, where many religious establishments of the rule of St. Saviour, introduced by her, were founded. A folio volume, containing the life, revelations, and legends of St. Bridget, was published by A. Koberger, Nuremberg, 1502, with the following title: “Das puch der Himlischen offenbarung der Heiligen wittiben Birgitte von dem Kunigreich Schweden.”

II.18 Those cuts consist of illustrations of the New Testament. There are ten of them, apparently a portion of a larger series, in the British Museum; and they are marked in small letters, a. b. c. d. e. f. g. h. i. k. n. That which is marked g. also contains the words “Opus Jacobi.” In this cut a specimen of cross-hatching may be observed, which was certainly very little practised—if at all—in Italy, before 1500.

II.19 Mr. Ottley’s reason for considering this cut to be so old is, that “after that period [1400] an artist, who was capable of designing so good a figure, could scarcely have been so grossly ignorant of every effect of linear perspective, as was evidently the case with the author of the performance before us.”—Inquiry, p. 87. Offences, however, scarcely less gross against the rules of linear perspective, are to be found in the wood-cuts in the Adventures of Sir Theurdank, 1517, many of which contain figures superior to that of St. Bridget. Errors in perspective are indeed frequent in the designs of many of the most eminent of Albert Durer’s contemporaries, although in other respects the figures may be correctly drawn, and the general composition good.

II.20 An engraving of this seal is given in the first volume of Meerman’s Origines Typographicæ.

II.21 Heineken, Neue Nachrichten von Künstlern und Kunstsachen. Dresden und Leipzig, 1786, S. 143.

II.22 In the Table des Matières to Jansen’s Essai sur l’Origine de la Gravure, Paris, 1808, we find “Dünkelspül (Nicolas) graveur Allemand en 1443.” After this specimen of accuracy, it is rather surprising that we do not find St. Alexius referred to also as “un graveur Allemand.”

II.23 St. Alexius returning unknown to his father’s house, as a poor pilgrim, was treated with great indignity by the servants.

II.24 Von Murr, Journal, 2 Theil, S. 113-115.

II.25 Jansen, Essai sur l’Origine de la Gravure, tom. i. p. 237. Jansen’s own authority on subjects connected with wood engraving is undeserving of attention. He is a mere compiler, who scarcely appears to have been able to distinguish a wood-cut from a copper-plate engraving.

II.26 Idée Générale, p. 251. Hartman Schedel, the compiler of the Nuremberg Chronicle, was accustomed to paste both old wood-cuts and copper-plate engravings within the covers of his books, many of which were preserved in the Library of the Elector of Bavaria at Munich.—Idée Gén. p. 287; and Von Murr, Journal, 2 Theil, S. 115.

II.27 Heineken thus speaks of those old devotional cuts: “On trouve dans la Bibliothèque de Wolfenbüttel de ces sortes d’estampes, qui représentent différens sujets de l’histoire sainte et de dévotion, avec du texte vis à vis de la figure, tout gravé en bois. Ces pièces sont de la même grandeur que nos cartes à jouer: elles portent 3 pouces de hauteur sur 2 pouces 6 lignes de largeur.”—Idée Générale, p. 249.

II.28 A copy of the Speculum belonging to the city of Harlem had at the commencement, “Ex Officina Laurentii Joannis Costeri. Anno 1440.” But this inscription had been inserted by a modern hand—Idée Générale, p. 449.

II.29 In the catalogue of Dr. Kloss’s Library, No. 2024, is a “Historia et Apocalypsis Johannis Evangelistæ,” imperfect, printed from wooden blocks. The following are the observations of the editor or compiler of the catalogue: “At the end of the volume is a short note, written by Pope Martin V., who occupied the papal chair from 1417 to 1431. This appears to accord with the edition described by Heineken at page 360, excepting in the double a, No. 3 and 4.” If the note referred to were genuine, and actually written in the book, a certain date would be at once established. The information, however, comes in a questionable shape, as the English rédacteur’s power of ascertaining who were the writers of ancient MS. notes appears little short of miraculous.

II.30 Bibliotheca Spenceriana, vol. i. p. 4, cited in Ottley’s Inquiry, vol. i. p. 99.

II.31 Singer’s Researches into the History of Playing-cards, p. 107.

II.32 Index Librorum ab inventa Typographia ad annum 1500, No. 37.

II.33 Mr. Bohn is in possession of a similarly bound volume, namely, “Astexani de Ast, Scrutinium Scripturarum,” printed by Mentelin, without date, but about 1468, on the pig-skin covers of which is printed in bold black letter, Per me Rich-en-bach illigatus in Gysslingen 1470.

II.34 “Catalogue of the Library of Dr. Kloss of Frankfort,” Nos. 460 and 468. Geisslingen is about fifteen miles north-west of Ulm in Suabia, and Gemund about twelve miles northward of Geisslingen.

II.35 Mr. Singer, at page 101 of his Researches into the History of Playing-cards, speaks of “one Plebanus of Augsburg,” as if Plebanus were a proper name. It has nearly the same meaning as our own word “Curate.” “Plebanus, Parœcus, Curio, Sacerdos, qui plebi præest; Italis, Piovano; Gallo-Belgis, Pleban. Balbus in Catholico: ‘Plebanus, dominus plebis, Presbyter, qui plebem regit.’—Plebanum vero maxime vocant in ecclesiis cathedralibus seu collegiatis canonicum, cui plebis earum jurisdictioni subditæ cura committitur.”—Du Cange, Glossarium, in verbo “Plebanus.”

II.36 Idée Générale, pp. 334-370.

II.37 In the copy of the Biblia Pauperum in the British Museum,

Inches. Inches.
The largest cut is 10-4/8 high, and 7-5/8 wide.
The smallest  — 10-1/8   —  — 7-5/8   —

In the Historia Virginis, also in the British Museum,

The largest cut is 10-3/8 high, and 7-2/8 wide.
The smallest  — 9-7/8   —   — 6-7/8   —

II.38 The two which are wanting are those numbered 36 and 37—that is, the second s, and the first t—in Heineken’s collation. Although there is a memorandum at the commencement of the book that those cuts are wanting, yet the person who has put in the numbers, in manuscript, at the foot of each, has not noticed the omission, but has continued the numbers consecutively, marking that 36 which in a perfect copy is 38, and so on to the rest. A reference to Heineken from those manuscript numbers subsequent to the thirty-fifth cut would lead to error.

II.39 Witness Rembrandt, who never gets rid of the Dutch character, no matter how elevated his subject may be.

II.40 Revelations, chap. xi. verses 3d and 4th.

II.41 Idée Générale, p. 376.

II.42 Inquiry, vol. i. p. 140.

II.43 Inquiry, p. 140.

II.44 Inquiry, p. 144, vol. i.

II.45 The copy from which the preceding specimens are given was formerly the property of the Rev. C. M. Cracherode, by whom it was left, with the rest of his valuable collection of books, to the British Museum.

II.46 Bibliotheca Spenceriana, vol. i. p. 36. Mr. Ottley cites the passage at p. 139, vol i. of his Inquiry, for the purpose of expressing his dissent from the theory.

II.47 Landseer’s Lectures on the Art of Engraving, pp. 201-205, 8vo. London, 1807.

II.48 Heineken, Idée Générale, p. 374. Von Murr, Journal, 2 Theil, S. 43.

II.49 Song of Solomon, chap. iv. verse 4.

II.50 Those arms are to be seen in Sebastiana Munsteri Cosmographia, cap. De Regione Wirtenbergensi, p. 592. Folio, Basiliæ, apud Henrichum Petri, 1554.

II.51 The backs of many of the old wood-cuts which have been taken by means of friction, still appear bright in consequence of the rubbing which the paper has sustained in order to obtain the impression. They would not have this appearance if the paper had been used in a damp state.

II.52 This must have been a copy of that which Heineken calls the second edition; no such appearances of a fracture or joining are to be seen in the first.

II.53 Inquiry, p. 142.

II.54 “It is a manual or kind of Catechism of the Bible,” says the Rev. T. H. Horne, “for the use of young persons and of the common people, whence it derives its name Biblia Pauperum,—the Bible of the Poor,—who were thus enabled to acquire, at a comparatively low price, an imperfect knowledge of some of the events recorded in the Scripture.”—Introduction to the Critical Study of the Scriptures, vol. ii. p. 224-5. The young and the poor must have been comparatively learned at that period to be able to read cramped Latin, when many a priest could scarcely spell his breviary.

II.55 J. G. Schelhorn, Amœnitates Literariæ, tom. iv. p. 297. 8vo. Francofurt. & Lips. 1730. Lichtenberger, Initia Typographica, p. 4, says erroneously, that Schelhorn’s fac-simile was engraved on copper. It is on wood, as Schelhorn himself states at p. 296.

II.56 J. D. Schœpflin, Vindiciæ Typographicæ, p. 7, 4to. Argentorati, 1760.

II.57 Ger. Meerman, Origines Typographicæ, P. 1, p. 241. 4to. Hagæ Comit. 1765.

II.58 Idée Générale, p. 292, note.

II.59 Camus, speaking of one of those manuscripts compared with the block-book, observes: “Ce dernier abrégé méritoit bien le nom de Biblia Pauperum, par comparison aux tableaux complets de la Bible que je viens d’indiquer. Des ouvrages tels que les tableaux complets ne pouvoient être que Biblia Divitum.”—Notice d’un Livre imprimé à Bamberg en 1462, p. 12, note. 4to. Paris, 1800.

II.60 “Entre ces abrégés [de la Bible] on remarque le Speculum Humanæ Salvationis et le Biblia Pauperum. Ces deux ouvrages ont beaucoup d’affinité entre eux pour le volume, le choix des histoires, les moralités, la composition des tableaux. Ils existent en manuscrits dans plusieurs bibliothèques.”—Camus, Notice d’un Livre, &c. p. 12.

II.61 “Librorum qui ante Reformationem in scholis Daniæ legebantur, Notitia. Hafniæ, 1784;” referred to by Camus, Notice d’un Livre, &c. p. 10.

II.62 Inquiry, vol. i. p. 129.

II.63 Idée Générale, p. 307, 308.

II.64 The passages referred to are probably the 8th, 9th, and 10th verses of the xxxivth Psalm; and the 8th verse of the xxixth chapter of Isaiah.

II.65 “Has autem icones ex sola sculptoris imaginatione et arbitrio fluxisse vel inde liquet, quod idem scriptor sacer in diversis foliis diversa plerumque et alia facie delineatus sistatur, sicuti, v. g. Esaias ac David, sæpius obvii, Protei instar, varias induerunt in hoc opere formas.”—Amœnitates Literariæ, tom. iv. p. 297.

II.66 Schelhorn has noticed a similar appearance in the old block-book entitled “Ars Memorandi:” “Videas hic nonnunquam literas atramento confluenti deformatas, ventremque illarum, alias album et vacuum, atramentaria macula repletum.” Amœnitat. Liter. tom. i. p. 7.

II.67 This collection of wood engravings from old blocks was published in three parts, large folio, at Gotha in 1808, 1810, and 1816, under the following title: “Holzschnitte alter Deutscher Meister in den Original-Platten gesammelt von Hans Albrecht Von Derschau: Als ein Beytrag zur Kunstgeschichte herausgegeben, und mit einer Abhandlung über die Holzschneidekunst und deren Schicksale begleitet von Rudolph Zacharias Becker.” The collector has frequently mistaken rudeness of design, and coarseness of execution, for proofs of antiquity.

II.68 Inquiry, vol. i. p. 130.

II.69 Notice d’un Livre, &c. p. 11.

II.70 Heineken, Idée Générale, p. 319.

II.71 Ornhielm’s book was printed in 4to. at Stockholm, 1689. The passage referred to is as follows: “Quos per numeros et signa conscripsisse cum [Ansgarium] libros Rembertus memorat indigitatos pigmentorum vocabulo, eos continuisse, palam est, quasdam aut e divinarum literarum, aut pie doctorum patrum scriptis, pericopas et sententias.”

II.72 “Ces conjectures sont foibles; elles ont été attaquées par Erasme Nyerup dans un écrit publié à Copenhague en 1784. . . . . Nyerup donne à penser que Heinecke a reconnu lui-même, dans la suite, la foiblesse de ses conjectures.”—Camus, Notice d’un Livre, &c. p. 9.

II.73 It is sometimes named “Speculum Figuratum;” and Junius in his account of Coster’s invention calls it “Speculum Nostræ Salutis.”

II.74 The cuts which have the text printed from wood-blocks are Nos. 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 16, 17, 21, 22, 26, 27, 46, and 55.—Heineken, Idée Générale, p. 444.

II.75 Heineken, Idée Générale, p. 474.

II.76 The “Batavia” or Junius, in which the name of Lawrence Coster first appears as a printer, was published in 1588.

II.77 Dissertation sur l’Origine et les Progrès de l’Art de Graver en Bois. Par M. Fournier le Jeune, 8vo. Paris, 1758.

II.78 A French translation of Meerman’s letter, which was originally written in Dutch, is given by Santander in his Dictionnaire Bibliographique, tom. i. pp. 14-18, 8vo. Bruxelles, 1805.

II.79 Dissertation, pp. 29-32. The many mistakes which Fournier commits in his Dissertation, excite a suspicion that he was either superficially acquainted with his subject, or extremely careless. He published two or three other small works on the subject of engraving and printing,—after the manner of “Supplements to an Appendix,”—the principal of which is entitled “De l’Origine et des Productions de l’Imprimerie primitive en taille de bois; avec une refutation des préjugés plus ou moins accredités sur cet art; pour servir de suite à la Dissertation sur l’Origine de l’Art de graver en bois. Paris, 1759.”

II.80 Heineken seems inclined to consider this as the second Dutch edition; and he only mentions it as the first Dutch edition because it is called so by Meerman.—Idée Gén. pp. 453, 454.

II.81 Inquiry into the Origin and Early History of Engraving, pp. 205-217. Though differing from Mr. Ottley in the conclusions which he draws from the facts elicited by him respecting the priority of the editions of the Speculum, I bear a willing testimony to the value of his discoveries on this subject, which may rank among the most interesting that have resulted from bibliographical research.

II.82 Wood-engravers of the present day are accustomed to transfer an old impression from a cut or a page of letter-press to a block in the following manner. They first moisten the back of the paper on which the cut or letter-press is printed with a mixture of concentrated potash and essence of lavender in equal quantities, which causes the ink to separate readily from the paper; next, when the paper is nearly dry, the cut or page is placed above a prepared block, and by moderate pressure the ink comes off from the paper, and leaves an impression upon the wood.

II.83 The subject is Daniel explaining to Belshazzar the writing on the wall.

II.84 Heineken gives an account of those twelve additional cuts at page 463 of his Idée Générale. It appears that Veldener also published in the same year another edition of the Speculum, also in quarto, containing the same cuts as the older folios, but without the twelve above mentioned.

II.85 see text The following is a reduced copy of the paper-mark, which appears to be a kind of anchor with a small cross springing from a ball or knob at the junction of the arms with the shank. It bears a considerable degree of resemblance to the mark given at page 62, from an edition of the Apocalypse. An anchor is to be found as a paper-mark in editions of the Apocalypse, and of the Poor Preachers’ Bible. According to Santander, a similar paper-mark is to be found in books printed at Cologne, Louvain, and Utrecht, from about 1470 to 1480.

II.86 The initial F, at the commencement of this chapter, is a reduced copy of the letter here described.

II.87 The first book with moveable types and wood-cuts both printed by means of the press is the Fables printed at Bamberg, by Albert Pfister, “Am Sant Valentinus tag,” 1461.

II.88 “Nostrum vero libellum, cujus gratia hæc præfati sumus, intrepide, si non primum artis inventæ fœtum, certe inter primos fuisse asseveramus.”—Amœnitates Literariæ, tom. i. p. 4.

II.89 Heineken had seen two editions of this book, and he gives fac-similes of their titles, which are evidently from different blocks. The title at full length is as follows: Ars memorandi notabilis per figuras Ewangelistarum hic ex post descriptam quam diligens lector diligenter legat et practiset per signa localia ut in practica experitur.”—“En horridum et incomtum dicendi genus, Priscianumque misere vapulantem!” exclaims Schelhorn.

II.90 Heineken, Idée Générale, p. 394.

II.91 Über die frühesten universal historischen Folgen der Erfindung der Buchdruckerkunst, von J. Christ. Freyherrn Von Aretin, S. 18. 4to. Munich, 1808.

II.92 “For the law was given by Moses, but grace and truth came by Jesus Christ.”—St. John’s Gospel, chap. i. v. 17.

II.93 “Forte tamen ea, quæ tintinnabulis haud videntur dissimilia, nummulariorum loculos et pecuniæ receptacula referunt.”—Schelhorn, Amœnit. Liter. tom. i. p. 10.

II.94 The following are the contents of the first page, descriptive of the cut: “Evangelium Johannis habet viginti unum capittula. Primum. In principio erat verbum de eternitate verbi et de trinitate. Secundum capittulum. Nupcie facte sunt in Chana Galilee et qualiter Christus subvertit mensas nummulariorum. Tertium capittulum. Erat antem homo ex Phariseis Nycodemus nomine. Quartum capittulum. Qualiter Ihesus peciit a muliere Samaritana bibere circum puteum Jacob et de regulo. Quintum capittulum. De probatica piscina ubi dixit Ihesus infirmo Tolle grabatum tuum & vade. Sextum capittulum. De refectione ex quinque panibus & duobus piscibus Et de ewkaristia.”—Schelhorn, Amœnit. Lit. tom. i. p. 9.

II.95 This work on Palmistry was composed in German by a Doctor Hartlieb, as is expressed at the beginning: “Das nachgeschriben buch von der hand hätt zu teutsch gemacht Doctor Hartlieb.” Specimens of the first and the last pages, and of one of the cuts, are given in Heineken’s Idée Générale, plates 27 and 28.

II.96 I am of opinion that this is the same person who executed the cuts for a German edition of the Poor Preachers’ Bible in 1475. His name does not appear; but on a shield of arms there is a spur, which may be intended as a rebus of the name; in the same manner as Albert Durer’s surname appears in his coat of arms, a pair of doors,—Durer, or, as his father’s name was sometimes spelled, Thurer.

II.97 Aretin says that in the Royal Library at Munich there are about forty books and about a hundred single leaves printed from engraved wood-blocks.—Über die Folgen, &c. S. 6.

II.98 Meerman had an old block of a Donatus, which was obtained from the collection of a M. Hubert of Basle, and which appeared to belong to the same edition as that containing sixteen lines in the Royal Library at Paris.—Heineken, Idée Générale, p. 258.

Errors in Chapter II

The term Formschneider, which was originally used
Fornschneider

lustra / cors . apientie
printed as shown: probably error for “lustra / tor . sapientie”

much better calculated to overthrow.II.43
overthrow.”

“Confute the exciseman and puzzle the vicar,—”
close quote missing

On these I have nothing to remark further
futher

not in the belief that I have made any important discovery
final t in “important” invisible

not so old as either the Apocalypse or the History of the Virgin
Apocalpyse

Mulier autem in paradiso est formata
formato

David with Goliath’s head
Goliah’s

The title at full length is as follows: “Ars memorandi
open quote missing

Footnote II.2

That is: “always excepting
open quote missing

Footnote II-7

der Spielkarten Zu erforschen,
Zuerforschen

118

CHAPTER III.
INVENTION OF TYPOGRAPHY.

The discovery of desroches.—the stamping of lodewyc van vaelbeke.—early “prenters” of antwerp and bruges not typographers.—cologne chronicle.—donatuses printed in holland.—gutemberg’s birth and family—progress of his invention—his law-suit with the drytzehns at strasburg—his return to mentz, and partnership with faust—partnership dissolved.—possibility of printing with wooden types examined.—supposed early productions of gutemberg and faust’s press.—proofs of gutemberg having a press of his own.—the vocabulary printed at elfeld.—gutemberg’s death and epitaphs.—invention of printing claimed for lawrence coster.—the account given by junius—contradicted, altered, and amended at will by meerman, koning, and others.—works pretended to be printed with coster’s types.—the horarium discovered by enschedius.

Before proceeding to trace the progress of wood engraving in connexion with typography, it appears necessary to give some account of the invention of the latter art. In the following brief narrative of Gutemberg’s life, I shall adhere to positive facts; and until evidence equally good shall be produced in support of another’s claim to the invention, I shall consider him as the father of typography. I shall also give Hadrian Junius’s account of the invention of wood engraving, block-printing, and typography by Lawrence Coster, with a few remarks on its credibility. Some of the conjectures and assertions of Meerman, Koning, and other advocates of Coster, will be briefly noticed, and their inconsistency pointed out. To attempt to refute at length the gratuitous assumptions of Coster’s advocates, and to enter into a detail of all their groundless arguments, would be like proving a medal to be a forgery by a long dissertation, when the modern fabricator has plainly put his name in the legend. The best proof of the fallacy of Coster’s claims to the honour of having discovered the art of printing with moveable types is to be found in the arguments of those by whom they have been supported.

119

Meerman, with all his research, has not been able to produce a single fact to prove that Lawrence Coster, or Lawrence Janszoon as he calls him, ever printed a single book; and it is by no means certain that his hero is the identical Lawrence Coster mentioned by Junius. In order to suit his own theory he has questioned the accuracy of the statements of Junius, and has thus weakened the very foundation of Coster’s claims. The title of the custos of St. Bavon’s to the honour of being the inventor of typography must rest upon the authenticity of the account given by Junius; and how far this corresponds with established facts in the history of wood engraving and typography I leave others to decide for themselves.

Among the many fancied discoveries of the real inventor of the art of printing, that of Monsieur Desroches, a member of the Imperial Academy of Sciences and Belles Lettres at Brussels, seems to require an especial notice. In a paper printed in the transactions of that society,III.1 he endeavoured to prove, that the art of printing books was practised in Flanders about the beginning of the fourteenth century; and one of the principal grounds of his opinion was contained in an old chronicle of Brabant, written, as is supposed, by one Nicholas le Clerk, [Clericus,] secretary to the city of Antwerp. The chronicler, after having described several remarkable events which happened during the government of John II. Duke of Brabant, who died in 1312, adds the following lines:

In dieser tyt sterf menschelyc

Die goede vedelare Lodewyc;

Die de beste was die voor dien

In de werelt ye was ghesien

Van makene ende metter hant;

Van Vaelbeke in Brabant

Alsoe was hy ghenant.

Hy was d’erste die vant

Van Stampien die manieren

Diemen noch hoert antieren.

This curious record, which Monsieur Desroches considered as so plain a proof of “die goede vedelare Lodewyc” being the inventor of printing, may be translated in English as follows:

This year the way of all flesh went

Ludwig, the fidler most excellent;

For handy-work a man of name;

From Vaelbeke in Brabant he came.

120

He was the first who did find out

The art of beating time, no doubt,

(Displaying thus his meikle skill,)

And fidlers all practise it still.III.2

The laughable mistake of Monsieur Desroches in supposing that fidler Ludwig’s invention, of beating time by stamping with the foot, related to the discovery of printing by means of the press, was pointed out in 1779 by Monsieur Ghesquiere in a letter printed in the Esprit des Journaux.III.3 In this letter Monsieur Ghesquiere shows that the Flemish word “Stampien,” used by the chronicler in his account of the invention of the “good fidler Ludwig,” had not a meaning similar to that of the word “stampus” explained by Ducange, but that it properly signified “met de voet kleppen,”—to stamp or beat with the feet.

In support of his opinion of the antiquity of printing, Monsieur Desroches refers to a manuscript in his possession, consisting of lives of the saints and a chronicle written in the fourteenth century. At the end of this manuscript was a catalogue of the books belonging to the monastery of Wiblingen, the writing of which was much abbreviated, and which appeared to him to be of the following century. Among other entries in the catalogue was this: “(It.) dōicali īpv̄o lībo ſtmp̄to ī bappiro nō s͞crpō.” On supplying the letters wanting Monsieur Desroches says that we shall have the following words: “Item. Dominicalia in parvo libro stampato in bappiro [papyro,] non scripto;” that is, “Item. Dominicals [a form of prayer or portion of church service] in a small book printed [or stamped] on paper, not written.” In the abbreviated word ſtm̄p̄to, he says that the letter m could not very well be distinguished; but the doubt which might thus arise he considers to be completely resolved by the words “non scripto,” and by the following memorandum which occurs, in the same hand-writing, at the foot of the page: “Anno Dñi 1340 viguit q̄ fēt stāpā Dñatos,”— 121 “In 1340 he flourished who caused Donatuses to be printed.” If the catalogue were really of the period supposed by Monsieur Desroches, the preceding extracts would certainly prove that the art of printing or stamping books, though not from moveable types, was practised in the fourteenth century; but, as the date has not been ascertained, its contents cannot be admitted as evidence on the point in dispute. Monsieur Ghesquiere is inclined to think that the catalogue was not written before 1470; and, as the compiler was evidently an ignorant person, he thinks that in the note, “Anno Domini 1340 viguit qui fecit stampare Donatos,” he might have written 1340 instead of 1440.

Although it has been asserted that the wood-cut of St. Christopher with the date 1423, and the wood-cut of the Annunciation—probably of the same period—were printed by means of a press, yet I consider it exceedingly doubtful if the press were employed to take impressions from wood-blocks before Gutemberg used it in his earliest recorded attempts to print with moveable types. I believe that in every one of the early block-books, where opportunity has been afforded of examining the back of each cut, unquestionable evidence has been discovered of their having been printed, if I may here use the term, by means of friction. Although there is no mention of a press which might be used to take impressions before the process between Gutemberg and the heirs of one of his partners, in 1439, yet “Prenters” were certainly known in Antwerp before his invention of printing with moveable types was brought to perfection. Desroches in his Essay on the Invention of Printing gives an extract from an order of the magistracy of Antwerp, in the year 1442, in favour of the fellowship or guild of St. Luke, called also the Company of Painters, which consisted of Painters, Statuaries, Stone-cutters, Glass-makers, Illuminators, and “Prenters”. This fellowship was doubtless similar to that of Venice, in whose favour a decree was made by the magistracy of that city in 1441, and of which some account has been given, at page 43, in the preceding chapter. There is evidence of a similar fellowship existing at Bruges in 1454; and John Mentelin, who afterwards established himself at Strasburg as a typographer or printer proper, was admitted a member of the Painters’ Company of that city as a “Chrysographus” or illuminator in 1447.III.4

Whether the “Prenters” of Antwerp in 1442 were acquainted with the use of the press, or not, is uncertain; but there can be little doubt of their not being Printers, as the word is now generally understood; that is, persons who printed books with moveable types. They were most likely block-printers, and such as engraved and printed cards and 122 images of saints; and it would seem that typographers were not admitted members of the society; for of all the early typographers of Antwerp the name of one only, Mathias Van der Goes, appears in the books of the fellowship of St. Luke; and he perhaps may have been admitted as a wood-engraver, on account of the cuts in an herbal printed with his types, without date, but probably between 1485 and 1490.

Ghesquiere, who successfully refuted the opinion of Desroches that typography was known at Antwerp in 1442, was himself induced to suppose that it was practised at Bruges in 1445, and that printed books were then neither very scarce nor very dear in that city.III.5 In an old manuscript journal or memorandum book of Jean-le-Robèrt, abbot of St. Aubert in the diocese of Cambray, he observed an entry stating that the said abbot had purchased at Bruges, in January 1446, a “Doctrinale gette en mole” for the use of his nephew. The words “gette en mole” he conceives to mean, “printed in type;” and he thinks that the Doctrinale mentioned was the work which was subsequently printed at Geneva, in 1478, under the title of Le Doctrinal de Sapience, and at Westminster by Caxton, in 1489, under the title of The Doctrinal of Sapyence. The Abbé Mercier de St. Leger, who wrote a reply to the observations of Ghesquiere, with greater probability supposes that the book was printed from engraved wood-blocks, and that it was the “Doctrinale Alexandri Galli,” a short grammatical treatise in monkish rhyme, which at that period was almost as popular as the “Donatus,” and of which odd leaves, printed on both sides, are still to be seen in libraries which are rich in early specimens of printing.

Although there is every reason to believe that the early Printers of Antwerp and Bruges were not acquainted with the use of moveable types, yet the mention of such persons at so early a period, and the notice of the makers “of cards and printed figures” at Venice in 1441, sufficiently declare that, though wood engraving might be first established as a profession in Suabia, it was known, and practised to a considerable extent, in other countries previous to 1450.

The Cologne Chronicle, which was printed in 1499, has been most unfairly quoted by the advocates of Coster in support of their assertions; and the passage which appeared most to favour their argument they have ascribed to Ulric Zell, the first person who established a press at Cologne. A shrewd German,III.6 however, has most clearly shown, from the same chronicle, that the actual testimony of Ulric Zell is directly in opposition 123 to the claims advanced by the advocates of Coster. The passage on which they rely is to the following effect: “Item: although the art [of printing] as it is now commonly practised, was discovered at Mentz, yet the first conception of it was discovered in Holland from the Donatuses, which before that time were printed there.” This we are given to understand by Meerman and Koning is the statement of Ulric Zell. A little further on, however, the Chronicler, who in the above passage appears to have been speaking in his own person from popular report, thus proceeds: “But the first inventor of printing was a citizen of Mentz, though born at Strasburg,III.7 named John Gutemberg: Item: from Mentz the above-named art first came to Cologne, afterwards to Strasburg, and then to Venice. This account of the commencement and progress of the said art was communicated to me by word of mouth by that worthy person Master Ulric Zell of Hanau, at the present time [1499] a printer in Cologne, through whom the said art was brought to Cologne.” At this point the advocates of Coster stop, as the very next sentence deprives them of any advantage which they might hope to gain from the “impartial testimony of the Cologne Chronicle,” the compiler of which proceeds as follows: “Item: there are certain fanciful people who say that books were printed before; but this is not true; for in no country are books to be found printed before that time.”III.8

That “Donatuses” and other small elementary books for the use of schools were printed from wood-blocks previous to the invention of typography there can be little doubt; and it is by no means unlikely that they might be first printed in Holland or in Flanders. At any rate an opinion seems to have been prevalent at an early period that the idea of printing with moveable types was first derived from a “Donatus,”III.9 printed from wood-blocks. In the petition of Conrad Sweinheim and Arnold Pannartz, two Germans, who first established 124 a press at Rome, addressed to Pope Sixtus IV. in 1472, stating the expense which they had incurred in printing books, and praying for assistance, they mention amongst other works printed by them, “Donati pro puerulis, unde IMPRIMENDI INITIUM sumpsimus;” that is: “Donatuses for boys, whence we have taken the beginning of printing.” If this passage is to be understood as referring to the origin of typography, and not to the first proofs of their own press, it is the earliest and the best evidence on the point which has been adduced; for it is very likely that both these printers had acquired a knowledge of their art at Mentz in the very office where it was first brought to perfection.

About the year 1400, Henne, or John Gænsfleisch de Sulgeloch, called also John Gutemberg zum Jungen, appears to have been born at Mentz. He had two brothers; Conrad who died in 1424, and Friele who was living in 1459. He had also two sisters, Bertha and Hebele, who were both nuns of St. Claire at Mentz. Gutemberg had an uncle by his father’s side, named Friele, who had three sons, named John, Friele, and Pederman, who were all living in 1459.

Gutemberg was descended of an honourable family, and he himself is said to have been by birth a knight.III.10 It would appear that the family had been possessed of considerable property. They had one house in Mentz called zum Gænsfleisch, and another called zum Gudenberg, or Gutenberg, which Wimpheling translates, “Domum boni montis.” The local name of Sulgeloch, or Sorgenloch, was derived from the name of a village where the family of Gænsfleisch had resided previous to their removing to Mentz. It seems probable that the house zum Jungen at Mentz came into the Gutembergs’ possession by inheritance. It was in this house, according to the account of Trithemius, that the printing business was carried on during his partnership with Faust.III.11

When Gutemberg called himself der Junge, or junior, it was doubtless to distinguish himself from Gænsfleisch der Elter, or senior, a name which frequently occurs in the documents printed by Koehler. Meerman has fixed upon the latter name for the purpose of giving to Gutemberg a brother of the same christian name, and of making him the thief who stole Coster’s types. He also avails himself of an error committed by Wimpheling and others, who had supposed John Gutemberg and John Gænsfleisch to be two different persons. In two deeds of sale, however, of the date 1441 and 1442, entered in the Salic book of the church of 125 St. Thomas at Strasburg, he is thus expressly named: “Joannes dictus Gensfleisch alias nuncupatus Gutenberg de Moguncia, Argentinæ commorans;” that is, “John Gænsfleisch, otherwise named Gutemberg, of Mentz, residing at Strasburg.”III.12 Anthony à Wood, in his History of the University of Oxford, calls him Tossanus; and Chevillier, in his Origine de l’Imprimerie de Paris, Toussaints. SeizIII.13 is within an ace of making him a knight of the Golden Fleece. That he was a man of property is proved by various documents; and those writers who have described him as a person of mean origin, or as so poor as to be obliged to labour as a common workman, are certainly wrong.

From a letter written by Gutemberg in 1424 to his sister Bertha it appears that he was then residing at Strasburg; and it is also certain that in 1430 he was not living at Mentz; for in an act of accommodation between the nobility and burghers of that city, passed in that year with the authority of the archbishop Conrad III., Gutemberg is mentioned among the nobles “die ytzund nit inlendig sint”—“who are not at present in the country.” In 1434 there is positive evidence of his residing at Strasburg; for in that year he caused the town-clerk of Mentz to be arrested for a sum of three hundred florins due to him from the latter city, and he agreed to his release at the instance of the magistrates of Strasburg within whose jurisdiction the arrest took place.III.14 In 1436 he entered into partnership with Andrew Drytzehn and others; and there is every reason to believe that at this period he was engaged in making experiments on the practicability of printing with moveable types, and that the chief object of his engaging with those persons was to obtain funds to enable him to perfect his invention.

From 1436 to 1444 the name of Gutemberg appears among the “Constaflers” or civic nobility of Strasburg. In 1437 he was summoned before the ecclesiastical judge of that city at the suit of Anne of Iron-Door,III.15 for breach of promise of marriage. It would seem that he afterwards fulfilled his promise, for in a tax-book of the city of Strasburg, Anne Gutemberg is mentioned, after Gutemberg had returned to Mentz, as paying the toll levied on wine.

Andrew Drytzehn, one of Gutemberg’s partners, having died in 1438, his brothers George and Nicholas instituted a process against Gutemberg to compel him either to refund the money advanced by their brother, or to admit them to take his place in the partnership. From the depositions 126 of the witnesses in this cause, which, together with the decision of the judges, are given at length by Schœpflin, there can be little doubt that one of the inventions which Gutemberg agreed to communicate to his partners was an improvement in the art of printing, such as it was at that period.

The following particulars concerning the partnership of Gutemberg with Andrew Drytzehn and others are derived from the recital of the case contained in the decision of the judges. Some years before his death, Andrew Drytzehn expressed a desire to learn one of Gutemberg’s arts, for he appears to have been fond of trying new experiments, and the latter acceding to his request taught him a method of polishing stones, by which he gained considerable profit. Some time afterwards, Gutemberg, in company with a person named John Riff, began to exercise a certain art whose productions were in demand at the fair of Aix-la-Chapelle. Andrew Drytzehn, hearing of this, begged that the new art might be explained to him, promising at the same time to give whatever premium should be required. Anthony Heilman also made a similar request for his brother Andrew Heilman.III.16 To both these applications Gutemberg assented, agreeing to teach them the art; it being stipulated that the two new partners were to receive a fourth part of the profits between them; that Riff was to have another fourth; and that the remaining half should be received by the inventor. It was also agreed that Gutemberg should receive from each of the new partners the sum of eighty florins of gold payable by a certain day, as a premium for communicating to them his art. The great fair of Aix-la-Chapelle being deferred to another year, Gutemberg’s two new partners requested that he would communicate to them without reserve all his wonderful and rare inventions; to which he assented on condition that to the former sum of one hundred and sixty florins they should jointly advance two hundred and fifty more, of which one hundred were to be paid immediately, and the then remaining seventy-five florins due by each were to be paid at three instalments. Of the hundred florins stipulated to be paid in ready money, Andrew Heilman paid fifty, according to his engagement, while Andrew Drytzehn only paid forty, leaving ten due. The term of the partnership for carrying on the “wonderful art” was fixed at five years; and it was also agreed that if any of the partners should die within that period, his interest in the utensils and stock should become vested in the surviving partners, who at the completion of the term were to pay to the heirs of the deceased the sum of one 127 hundred florins. Andrew Drytzehn having died within the period, and when there remained a sum of eighty-five florins unpaid by him, Gutemberg met the claim of his brothers by referring to the articles of partnership, and insisted that from the sum of one hundred florins which the surviving partners were bound to pay, the eighty-five remaining unpaid by the deceased should be deducted. The balance of fifteen florins thus remaining due from the partnership he expressed his willingness to pay, although according to the terms of the agreement it was not payable until the five years were expired, and would thus not be strictly due for some years to come. The claim of George Drytzehn to be admitted a partner, as the heir of his brother, he opposed, on the ground of his being unacquainted with the obligations of the partnership; and he also denied that Andrew Drytzehn had ever become security for the payment of any sum for lead or other things purchased on account of the business, except to Fridelin von Seckingen, and that this sum (which was owing for lead) Gutemberg himself paid. The judges having heard the allegations of both parties, and having examined the agreement between Gutemberg and Andrew Drytzehn, decided that the eighty-five florins which remained unpaid by the latter should be deducted from the hundred which were to be repaid in the event of any one of the partners dying; and that Gutemberg should pay the balance of fifteen florins to George and Nicholas Drytzehn, and that when this sum should be paid they should have no further claim on the partnership.III.17

From the depositions of some of the witnesses in this process, there can scarcely be a doubt that the “wonderful art” which Gutemberg was attempting to perfect was typography or printing with moveable types. FournierIII.18 thinks that Gutemberg’s attempts at printing, as may be gathered from the evidence in this cause, were confined to printing from wood-blocks; but such expressions of the witnesses as appear to relate to printing do not favour this opinion. As Gutemberg lived near the monastery of St. Arbogast, which was without the walls of the city, it appears that the attempts to perfect his invention were carried on in the house of his partner Andrew Drytzehn. Upon the death of the latter, Gutemberg appears to have been particularly anxious that “four pieces” which were in a “press” should be “distributed,”—making use of the very word which is yet used in Germany to express the distribution or separation of a form of types—-so that no person should know what they were.

Hans Schultheis, a dealer in wood, and Ann his wife, depose to the following effect: After the death of Andrew Drytzehn, Gutemberg’s 128 servant, Lawrence Beildeck, came to their house, and thus addressed their relation Nicholas Drytzehn: “Your deceased brother Andrew had four “pieces” placed under a press, and John Gutemberg requests that you will take them out and lay them separately [or apart from each other] upon the press so that no one may see what it is.”III.19

Conrad Saspach states that one day Andrew Heilman, a partner of Gutemberg’s, came to him in the Merchants’ Walk and said to him, “Conrad, as Andrew Drytzehn is dead, and as you made the press and know all about it, go and take the piecesIII.20 out of the press and separate [zerlege] them so that no person may know what they are.” This witness intended to do as he was requested, but on making inquiry the day after St. Stephen’s DayIII.21 he found that the work was removed.

Lawrence Beildeck, Gutemberg’s servant, deposes that after Andrew Drytzehn’s death he was sent by his master to Nicholas Drytzehn to tell him not to show the press which he had in his house to any person. Beildeck also adds that he was desired by Gutemberg to go to the presses, and to open [or undo] the press which was fastened with two screws, so that the “pieces” [which were in it] should fall asunder. The said “pieces” he was then to place in or upon the press, so that no person might see or understand them.

Anthony Heilman, the brother of one of Gutemberg’s partners, states that he knew of Gutemberg having sent his servant shortly before Christmas both to Andrew Heilman and Andrew Drytzehn to bring away all the “forms” [formen] that they might be separated in his presence, as he found several things in them of which he disapproved.III.22 The same witness also states that he was well aware of many people being wishful 129 to see the press, and that Gutemberg had desired that they should send some person to prevent its being seen.

Hans Dünne, a goldsmith, deposed that about three years before, he had done work for Gutemberg on account of printing alone to the amount of a hundred florins.III.23

As Gutemberg evidently had kept his art as secret as possible, it is not surprising that the notice of it by the preceding witnesses should not be more explicit. Though it may be a matter of doubt whether his invention was merely an improvement on block-printing, or an attempt to print with moveable types, yet, bearing in mind that express mention is made of a press and of printing, and taking into consideration his subsequent partnership with Faust, it is morally certain that Gutemberg’s attention had been occupied with some new discovery relative to printing at least three years previous to December 1439.

If Gutemberg’s attempts when in partnership with Andrew Drytzehn and others did not extend beyond block-printing, and if the four “pieces” which were in the press are assumed to have been four engraved blocks, it is evident that the mere unscrewing them from the “chase” or frame in which they might be enclosed, would not in the least prevent persons from knowing what they were; and it is difficult to conceive how the undoing of the two screws would cause “the pieces” to fall asunder. If, however, we suppose the four “pieces” to have been so many pages of moveable types screwed together in a frame, it is easy to conceive the effect of undoing the two screws which held it together. On this hypothesis, Gutemberg’s instructions to his servant, and Anthony Heilman’s request to Conrad Saspach, the maker of the press, that he would take out the “pieces” and distribute them, are at once intelligible. If Gutemberg’s attempts were confined to block-printing, he could certainly have no claim to the discovery of a new art, unless indeed we are to suppose that his invention consisted in the introduction of the press for the purpose of taking impressions; but it is apparent that his anxiety was not so much to prevent people seeing the press as to keep them ignorant of the purpose for which it was employed, and to conceal what was in it.

The evidence of Hans Dünne the goldsmith, though very brief, is in favour of the opinion that Gutemberg’s essays in printing were made with moveable types of metal; and it also is corroborated by the fact of lead being one of the articles purchased on account of the partnership. It is certain that goldsmiths were accustomed to engrave letters and figures upon silver and other metals long before the art of copper-plate printing was introduced; and Fournier not attending to the distinction 130 between simple engraving on metal and engraving on a plate for the purpose of taking impressions on paper, has made a futile objection to the argument of Bär,III.24 who very naturally supposes that the hundred florins which Hans Dünne received from Gutemberg for work done on account of printing alone, might be on account of his having cut the types, the formation of which by means of punches and matrices was a subsequent improvement of Peter Scheffer. It is indeed difficult to conceive in what manner a goldsmith could earn a hundred florins for work done on account of printing, except in his capacity as an engraver; and as I can see no reason to suppose that Hans Dünne was an engraver on wood, I am inclined to think that he was employed by Gutemberg to cut the letters on separate pieces of metal.

There is no evidence to show that Gutemberg succeeded in printing any books at Strasburg with moveable types: and the most likely conclusion seems to be that he did not. As the process between him and the Drytzehns must have given a certain degree of publicity to his invention, it might be expected that some notice would have been taken of its first-fruits had he succeeded in making it available in Strasburg. On the contrary, all the early writers in the least entitled to credit, who have spoken of the invention of printing with moveable types, agree in ascribing the honour to Mentz, after Gutemberg had returned to that city and entered into partnership with Faust. Two writers, however, whose learning and research are entitled to the highest respect, are of a different opinion. “It has been doubted,” says Professor Oberlin, “that Gutemberg ever printed books at Strasburg. It is, nevertheless, probable that he did; for he had a press there in 1439, and continued to reside in that city for five years afterwards. He might print several of those small tracts without date, in which the inequality of the letters and rudeness of the workmanship indicate the infancy of the art. Schœpflin thinks that he can identify some of them; and the passages cited by him clearly show that printing had been carried on there.”III.25 It is, however, to be remarked that the passages cited by Schœpflin, and referred to by Oberlin, 131 by no means show that the art of printing had been practised at Strasburg by Gutemberg; nor do they clearly prove that it had been continuously carried on there by his partners or others to the time of Mentelin, who probably established himself there as a printer in 1466.

It has been stated that Gutemberg’s first essays in typography were made with wooden types; and Daniel Specklin, an architect of Strasburg, who died in 1589, professed to have seen some of them. According to his account there was a hole pierced in each letter, and they were arranged in lines by a string being passed through them. The lines thus formed like a string of beads were afterwards collected into pages, and submitted to the press. Particles and syllables of frequent occurrence were not formed of separate letters, but were cut on single pieces of wood. We are left to conjecture the size of those letters; but if they were sufficiently large to allow of a hole being bored through them, and to afterwards sustain the action of the press, they could not well be less than the missal types with which Faust and Scheffer’s Psalter is printed. It is however likely that Specklin had been mistaken; and that he had supposed some old initial letters, large enough to admit of a hole being bored through them without injury, to have been such as were generally used in the infancy of the art.

In 1441 and 1442, Gutemberg, who appears to have been always in want of money, executed deeds of sale to the dean and chapter of the collegiate church of St. Thomas at Strasburg, whereby he assigned to them certain rents and profits in Mentz which he inherited from his uncle John Leheymer, who had been a judge in that city. In 1443 and 1444 Gutemberg’s name still appears in the rate or tax book of Strasburg; but after the latter year it is no longer to be found. About 1445, it is probable that he returned to Mentz, his native city, having apparently been unsuccessful in his speculations at Strasburg. From this period to 1450 it is likely that he continued to employ himself in attempts to perfect his invention of typography. In 1450 he entered into partnership with John Faust, a goldsmith and native of Mentz, and it is from this year that Trithemius dates the invention. In his Annales Hirsaugienses, under the year 1450, he gives the following account of the first establishment and early progress of the art. “About this time [1450], in the city of Mentz upon the Rhine, in Germany, and not in Italy as some have falsely stated, this wonderful and hitherto unheard of art of printing was conceived and invented by John Gutemberg, a citizen of Mentz. He had expended nearly all his substance on the invention; and being greatly pressed for want of means, was about to abandon it in despair, when, through the advice and with the money furnished by John Faust, also a citizen of Mentz, he completed his undertaking. At first they printed the vocabulary called the Catholicon, from letters cut on blocks of wood. 132 These letters however could not be used to print anything else, as they were not separately moveable, but were cut on the blocks as above stated. To this invention succeeded others more subtle, and they afterwards invented a method of casting the shapes, named by them matrices, of all the letters of the Roman alphabet, from which they again cast letters of copper or tin, sufficient to bear any pressure to which they might be subjected, and which they had formerly cut by hand. As I have heard, nearly thirty years ago, from Peter Scheffer, of Gernsheim, citizen of Mentz, who was son-in-law of the first inventor, great difficulties attended the first establishment of this art; for when they had commenced printing a Bible they found that upwards of four thousand florins had been expended before they had finished the third quaternion [or quire of four sheets]. Peter Scheffer, an ingenious and prudent man, at first the servant, and afterwards, as has been already said, the son-in-law of John Faust, the first inventor, discovered the more ready mode of casting the types, and perfected the art as it is at present exercised. These three for some time kept their method of printing a secret, till at length it was divulged by some workmen whose assistance they could not do without. It first passed to Strasburg, and gradually to other nations.”III.26

As Trithemius finished the work which contains the preceding account in 1514, Marchand concludes that he must have received his information from Scheffer about 1484, which would be within thirty-five years of Gutemberg’s entering into a partnership with Faust. Although Trithemius had his information from so excellent an authority, yet the account which he has thus left is far from satisfactory. Schœpflin, amongst other objections to its accuracy, remarks that Trithemius is wrong in stating that the invention of moveable types was subsequent to Gutemberg’s connexion with Faust, seeing that the former had previously employed them at Strasburg; and he also observes that in the learned abbot’s account there is no distinct mention made of moveable letters cut by hand, but that we are led to infer that the improvement of casting types from matrices immediately followed the printing of the Catholicon from wood-blocks. The words of Trithemius on this point are as follows: “Post hæc, inventis successerunt subtiliora, inveneruntque modum fundendi formas omnium Latini alphabeti litterarum, quas ipsi matrices nominabant, ex quibus rursum æneos sive stanneos characteres fundebant ad omnem pressuram sufficientes quos prius manibus sculpebant.” From this passage it might be objected in opposition to the opinion of Schœpflin:III.27 1. That the “subtiliora,”—more subtle contrivances, mentioned before the invention of casting moveable letters, may relate to the cutting 133 of such letters by hand. 2. That the word “quos” is to be referred to the antecedent “æneos sive stanneos characteres,”—letters of copper or tin,—and not to the “characteres in tabulis ligneis scripti,”—letters engraved on wood-blocks,—which are mentioned in a preceding sentence. The inconsistency of Trithemius in ascribing the origin of the art to Gutemberg, and twice immediately afterwards calling Scheffer the son-in-law of “the first inventor,” Faust, is noticed by Schœpflin, and has been pointed out by several other writers.

In 1455 the partnership between Gutemberg and Faust was dissolved at the instance of the latter, who preferred a suit against his partner for the recovery, with interest, of certain sums of money which he had advanced. There is no mention of the time when the partnership commenced in the sentence or award of the judge; but Schwartz infers, from the sum claimed on account of interest, that it must have been in August 1449. It is probable that his conclusion is very near the truth; for most of the early writers who have mentioned the invention of printing at Mentz by Gutemberg and Faust, agree in assigning the year 1450 as that in which they began to practise the new art. It is conjectured by Santander that Faust, who seems to have been a selfish character,III.28 sought an opportunity of quarrelling with Gutemberg as soon as Scheffer had communicated to him his great improvement of forming the letters by means of punches and matrices.

The document containing the decision of the judges was drawn up by Ulric Helmasperger, a notary, on 6th November, 1455, in the presence of Peter Gernsheim [Scheffer], James Faust, the brother of John, Henry Keffer, and others.III.29 From the statement of Faust, as recited in this instrument, it appears that he had first advanced to Gutemberg eight hundred florins at the annual interest of six per cent., and afterwards eight hundred florins more. Gutemberg having neglected to pay the interest, there was owing by him a sum of two hundred and fifty florins on account of the first eight hundred; and a further sum of one hundred and forty on account of the second. In consequence of Gutemberg’s 134 neglecting to pay the interest, Faust states that he had incurred a further expense of thirty-six florins from having to borrow money both of Christians and Jews. For the capital advanced by him, and arrears of interest, he claimed on the whole two thousand and twenty florins.III.30

In answer to these allegations Gutemberg replied: that the first eight hundred florins which he received of Faust were advanced in order to purchase utensils for printing, which were assigned to Faust as a security for his money. It was agreed between them that Faust should contribute three hundred florins annually for workmen’s wages and house-rent, and for the purchase of parchment, paper, ink, and other things.III.31 It was also stipulated that in the event of any disagreement arising between them, the printing materials assigned to Faust as a security should become the property of Gutemberg on his repaying the sum of eight hundred florins. This sum, however, which was advanced for the completion of the work, Gutemberg did not think himself bound to expend on book-work alone; and although it was expressed in their agreement that he should pay six florins in the hundred for an annual interest, yet Faust assured him that he would not accept of it, as the eight hundred florins were not paid down at once, as by their agreement they ought to have been. For the second sum of eight hundred florins he was ready to render Faust an account. For interest or usury he considered that he was not liable.III.32

The judges, having heard the statements of both parties, decided that Gutemberg should repay Faust so much of the capital as had not been expended in the business; and that on Faust’s producing witnesses, or swearing that he had borrowed upon interest the sums advanced, Gutemberg should pay him interest also, according to their agreement. Faust having made oath that he had borrowed 1550 florins, which he paid over to Gutemberg, to be employed by him for their common benefit, and that he had paid yearly interest, and was still liable on account of the same, the notary, Ulric Helmasperger, signed his attestation of the award on 135 6th November, 1455.III.33 It would appear that Gutemberg not being able to repay the money was obliged to relinquish the printing materials to Faust.

Salmuth, who alludes to the above document in his annotations upon Pancirollus, has most singularly perverted its meaning, by representing Gutemberg as the person who advanced the money, and Faust as the ingenious inventor who was sued by his rich partner. “From this it evidently appears,” says he, after making Gutemberg and Faust exchange characters, “that Gutemberg was not the first who invented and practised typography; but that some years after its invention he was admitted a partner by John Faust, to whom he advanced money.” If for “Gutemberg” we read “Faust,” and vice versâ, the account is correct.

Whether Faust, who might be an engraver as well as a goldsmith, assisted Gutemberg or not by engraving the types, does not appear. It is stated that Gutemberg’s earliest productions at Mentz were an alphabet cut on wood, and a Donatus executed in the same manner. Trithemius mentions a “Catholicon” engraved on blocks of wood as one of the first books printed by Gutemberg and Faust, and this Heineken thinks was the same as the Donatus.III.34 Whatever may have been the book which Trithemius describes as a “Catholicon,” it certainly was not the “Catholicon Joannis Januensis,” a large folio which appeared in 1460 without the name or residence of the printer, but which is supposed to have been printed by Gutemberg after the dissolution of his partnership with Faust.

It has been stated that previous to the introduction of metal types Gutemberg and Faust used moveable types of wood; and Schœpflin speaks confidently of such being used at Strasburg by Mentelin long after Scheffer had introduced the improved method of forming metal types by means of punches and matrices. On this subject, however, Schœpflin’s opinion is of very little weight, for on whatever relates to the practice of typography or wood engraving he was very slightly informed. He fancies that all the books printed at Strasburg previous to the appearance of Vincentii Bellovacensis Speculum Historiale in 1473, were printed with moveable types of wood. It is, however, doubtful if ever a single book was printed in this manner.

136

Willett in his Essay on Printing, published in the eleventh volume of the Archæologia, not only says that no entire book was ever printed with wooden types, but adds, “I venture to pronounce it impossible.” He has pronounced rashly. Although it certainly would be a work of considerable labour to cut a set of moveable letters of the size of what is called Donatus type, and sufficient to print such a book, yet it is by no means impossible. That such books as “Eyn Manung der Cristenheit widder die durken,” of which a fac-simile is given by Aretin, and the first and second Donatuses, of which specimens are given by Fischer, might be printed from wooden types I am perfectly satisfied, though I am decidedly of opinion that they were not. Marchand has doubted the possibility of printing with wooden types, which he observes would be apt to warp when wet for the purpose of cleaning; but it is to be observed that they would not require to be cleaned before they were used.

Fournier, who was a letter-founder, and who occasionally practised wood engraving, speaks positively of the Psalter first printed by Faust and Scheffer in 1457, and again in 1459, being printed with wooden types; and he expresses his conviction of the practicability of cutting and printing with such types, provided that they were not of a smaller size than Great Primer Roman. Meerman shows the possibility of using such types; and Camus caused two lines of the Bible, supposed to have been printed by Gutemberg, to be cut in separate letters on wood, and which sustained the action of the press.III.35 Lambinet says, it is certain that Gutemberg cut moveable letters of wood, but he gives no authority for the assertion; and I am of opinion that no unexceptionable testimony on this point can be produced. The statements of Serarius and Paulus Pater,III.36 who profess to have seen such ancient wooden types at Mentz, are entitled to as little credit as Daniel Specklin, who asserted that he had seen such at Strasburg. They may have seen large initial letters of wood with holes bored through, but scarcely any lower-case letters which were ever used in printing any book.

That experiments might be made by Gutemberg with wooden types I can believe, though I have not been able to find any sufficient authority for the fact. Of the possibility of cutting moveable types of a certain size in wood, and of printing a book with them, I am convinced from experiment; and could convince others, were it worth the expense, by 137 printing a fac-simile, from wooden types, of any page of any book which is of an earlier date than 1462. But, though convinced of the possibility of printing small works in letters of a certain size, with wooden types, I have never seen any early specimens of typography which contained positive and indisputable indications of having been printed in that manner. It was, until of late, confidently asserted by persons who pretended to have a competent knowledge of the subject, that the text of the celebrated Adventures of Theurdank, printed in 1517, had been engraved on wood-blocks, and their statement was generally believed. There cannot, however, now be a doubt in the mind of any person who examines the book, and who has the slightest knowledge of wood engraving and printing, of the text being printed with metal types.

During the partnership of Gutemberg and Faust it is likely that they printed some works, though there is scarcely one which can be assigned to them with any degree of certainty. One of the supposed earliest productions of typography is a letter of indulgence conceded on the 12th of August, 1451, by Pope Nicholas V, to Paulin Zappe, counsellor and ambassador of John, King of Cyprus. It was to be in force for three years from the 1st of May, 1452, and it granted indulgence to all persons who within that period should contribute towards the defence of Cyprus against the Turks. Four copies of this indulgence are known, printed on vellum in the manner of a patent or brief. The characters are of a larger size than those of the “Durandi Rationale,” 1459, or of the Latin Bible printed by Faust and Scheffer in 1462. The following date appears at the conclusion of one of the copies: “Datum Erffurdie sub anno Domini m cccc liiij, die vero quinta decima mensis novembris.” The words which are here printed in Italic, are in the original written with a pen. A copy of the same indulgence discovered by Professor Gebhardi is more complete. It has at the end, a “Forma plenissimæ absolutionis et remissionis in vita et in mortis articulo,”—a form of plenary absolution and remission in life and at the point of death. At the conclusion is the following date, the words in Italics being inserted with a pen: “Datum in Luneborch anno Domini m cccc l quinto, die vero vicesima sexta mensis Januarii.” Heineken, who saw this copy in the possession of Breitkopf, has observed that in the original date, m cccc liiij, the last four characters had been effaced and the word quinto written with a pen; but yet in such a manner that the numerals iiij might still be perceived. In two copies of this indulgence in the possession of Earl Spencer, described by Dr. Dibdin in the Bibliotheca Spenceriana, vol. i. p. 44, the final units (iiij) have not had the word “quinto” overwritten, but have been formed with a pen into the numeral V. In the catalogue of Dr. Kloss’s library, No. 1287, it is stated that a fragment of a “Donatus” there described, consisting of two leaves of parchment, is printed 138 with the same type as the Mazarine Bible; and it is added, on the authority of George Appleyard, Esq., Earl Spencer’s librarian, that the “Littera Indulgentiæ” of Pope Nicholas V, in his lordship’s possession, contains two lines printed with the same type. Breitkopf had some doubts respecting this instrument; but a writer in the Jena Literary Gazette is certainly wrong in supposing that it had been ante-dated ten years. It was only to be in force for three years; and Pope Nicholas V, by whom it was granted, died on the 24th March, 1455.III.37 Two words, UNIVERSIS and PAULINUS, which are printed in capitals in the first two lines, are said to be of the same type as those of a Bible of which Schelhorn has given a specimen in his “Dissertation on an early Edition of the Bible,” Ulm, 1760.

The next earliest specimen of typography with a date is the tract entitled “Eyn Manung der Cristenkeit widder die durken,”—An Appeal to Christendom against the Turks,—which has been alluded to at page 136. A lithographic fac-simile of the whole of this tract, which consists of nine printed pages of a quarto size, is given by Aretin at the end of his “Essay on the earliest historical results of the invention of Printing,” published at Munich in 1808. This “Appeal” is in German rhyme, and it consists of exhortations, arranged under every month in the manner of a calendar, addressed to the pope, the emperor, to kings, princes, bishops, and free states, encouraging them to take up arms and resist the Turks. The exhortation for January is addressed to Pope Nicholas V, who died, as has been observed, in March 1455. Towards the conclusion of the prologue is the date “Als man zelet noch din’ geburt offenbar m.cccc.lv. iar sieben wochen und iiii do by von nativitatis bis esto michi.” At the conclusion of the exhortation for December are the following words: “Eyn gut selig nuwe Jar:” A happy new year! From these circumstances Aretin is of opinion that the tract was printed towards the end of 1454. M. Bernhart, however, one of the superintendents of the Royal Library at Munich, of which Aretin was the principal director, has questioned the accuracy of this date; and from certain allusions in the exhortation for December, has endeavoured to show that the correct date ought to be 1472.III.38

Fischer in looking over some old papers discovered a calendar of a folio size, and printed on one side only, for 1457. The letters, according to his description, resemble those of a Donatus, of which he has given a specimen in the third part of his Typographic Rarities, and he supposes that both the Donatus and the Calendar were printed by Gutemberg.III.39 139 It is, however, certain that the Donatus which he ascribed to Gutemberg was printed by Peter Scheffer, and in all probability after Faust’s death; and from the similarity of the type it is likely that the Calendar was printed at the same office. Fischer, having observed that the large ornamental capitals of this Donatus were the same as those in the Psalter printed by Faust and Scheffer in 1457, was led most erroneously to conclude that the large ornamental letters of the Psalter, which were most likely of wood, had been cut by Gutemberg. The discovery of a Donatus with Peter Scheffer’s imprint has completely destroyed his conjectures, and invalidated the arguments advanced by him in favour of the Mazarine Bible being printed by Gutemberg alone.

As Trithemius and the compiler of the Cologne Chronicle have mentioned a Bible as one of the first books printed by Gutemberg and Faust, it has been a fertile subject of discussion among bibliographers to ascertain the identical edition to which the honour was to be awarded. It seems, however, to be now generally admitted that the edition called the MazarineIII.40 is the best entitled to that distinction. In 1789 Maugerard produced a copy of this edition to the Academy of Metz, containing memoranda which seem clearly to prove that it was printed at least as early as August 1456. As the partnership between Gutemberg and Faust was only dissolved in November 1455, it is almost impossible that such could have been printed by either of them separately in the space of eight months; and as there seems no reason to believe that any other typographical establishment existed at that period, it is most likely that this was the identical edition alluded to by Trithemius as having cost 4,000 florins before the partners, Gutemberg and Faust, had finished the third quaternion, or quire of four sheets.

The copy produced by Maugerard is printed on paper, and is now in the Royal Library at Paris. It is bound in two volumes; and every complete page consists of two columns, each containing forty-two lines. At the conclusion of the first volume the person by whom it was rubricatedIII.41 and bound has written the following memorandum: “Et sic est finis prime partis biblie. Scr. Veteris testamenti. Illuminata seu rubricata et illuminata p’ henricum Albeh alius Cremer anno dn’i m.cccc.lvi festo Bartholomei apli—Deo gratias—alleluja.” At the end of the second 140 volume the same person has written the date in words at length: “Iste liber illuminatus, ligatus & completus est p’ henricum Cremer vicariū ecclesie collegatur Sancti Stephani maguntini sub anno D’ni millesimo quadringentesimo quinquagesimo sexto festo assumptionis gloriose virginis Marie. Deo gracias alleluja.III.42 FischerIII.43 says that this last memorandum assigns “einen spätern tag”—a later day—to the end of the rubricator’s work. In this he is mistaken; for the feast of the Assumption of the Virgin, when the second volume was finished, is on the 15th of August: while the feast of St. Bartholomew, the day on which he finished the first, falls on August 24th. Lambinet,III.44 who doubts the genuineness of those inscriptions, makes the circumstance of the second volume being finished nine days before the first, a ground of objection. This seeming inconsistency however can by no means be admitted as a proof of the inscriptions being spurious. It is indeed more likely that the rubricator might actually finish the second volume before the first, than that a modern forger, intent to deceive, should not have been aware of the objection.

The genuineness of the inscriptions is, however, confirmed by other evidence which no mere conjecture can invalidate. On the last leaf of this Bible there is a memorandum written by Berthold de Steyna, vicar of the parochial church of “Ville-Ostein,”III.45 to the sacrist of which the Bible belonged. The sum of this memorandum is that on St. George’s day [23d April] 1457 there was chaunted, for the first time by the said Berthold, the mass of the holy sacrament. In the Carthusian monastery without the walls of Mentz, SchwartzIII.46 says that he saw a copy of this edition, the last leaves of which were torn out; but that in an old catalogue he perceived an entry stating that this Bible was presented to the monastery by Gutemberg and Faust. If the memorandum in the catalogue could be relied on as genuine, it would appear that this Bible had been completed before the dissolution of Gutemberg and Faust’s partnership in November 1455.

Although not a single work has been discovered with Gutemberg’s imprint, yet there cannot be a doubt of his having established a press of his own, and printed books at Mentz after the partnership between him and Faust had been dissolved. In the chronicle printed by Philip de Lignamine at Rome in 1474, it is expressly stated, under the year 1458, 141 that there were then two printers at Mentz skilful in printing on parchment with metal types. The name of one was Cutemberg, and the other Faust; and it was known that each of them could print three hundred sheets in a day.III.47 On St. Margaret’s day, 20th July, 1459, Gutemberg, in conjunction with his brother Friele and his cousins John, Friele, and Pederman, executed a deed in favour of the convent of St. Clara at Mentz, in which his sister Hebele was a nun. In this document, which is preserved among the archives of the university of Mentz, there occurs a passage, “which makes it as clear,” says Fischer, who gives the deed entire, “as the finest May-day noon, that Gutemberg had not only printed books at that time, but that he intended to print more.” The passage alluded to is to the following effect: “And with respect to the books which I, the above-named John, have given the library of the said convent, they shall remain for ever in the said library; and I, the above-named John, will furthermore give to the library of the said convent all such books required for pious uses and the service of God,—whether for reading or singing, or for use according to the rules of the order,—as I, the above-named John, have printed or shall hereafter print.”III.48

That Gutemberg had a press of his own is further confirmed by a bond or deed of obligation executed by Dr. Conrad Homery on the Friday after St. Matthias’ day, 1468, wherein he acknowledges having received “certain forms, letters, utensils, materials, and other things belonging to printing,” left by John Gutemberg deceased; and he binds himself to the archbishop Adolphus not to use them beyond the territory of Mentz, and in the event of his selling them to give a preference to a person belonging to that city.

The words translated “certain forms, letters, utensils, materials, and other things belonging to printing,” in the preceding paragraph, are in the original enumerated as: “etliche formen, buchstaben, instrument, gezuge und anders zu truckwerck gehoerende.” As there is a distinction made between “formen” and “buchstaben,”—literally, “forms” and “letters,”—Schwartz is inclined to think that by “formen” engraved wood-blocks might be meant, and he adduces in favour of his opinion the word “formen-schneider,” the old German name for a wood-engraver. One or more pages of type when wedged into a rectangular iron frame called a “chase,” and ready for the press, is termed a “form” both by English and German printers; but Schwartz thinks that such were not the “forms” 142 mentioned in the document. As there appears to be a distinction also between “instrument” and “gezuge,”—translated utensils and materials,—he supposes that the latter word may be used to signify the metal of which the types were formed. He observes that German printers call their old worn-out types “der Zeug”—literally, “stuff,” and that the mixed metal of which types are composed is also known as “der Zeug, oder Metall.”III.49 It is to be remembered that the earliest printers were also their own letter-founders.

The work called the Catholicon, compiled by Johannes de Balbis, Januensis, a Dominican, which appeared in 1460 without the printer’s name, has been ascribed to Gutemberg’s press by some of the most eminent German bibliographers. It is a Latin dictionary and introduction to grammar, and consists of three hundred and seventy-three leaves of large folio size. Fischer and others are of opinion that a Vocabulary, printed at Elfeld,—in Latin, Altavilla,—near Mentz, on 6th November, 1467, was executed with the same types. At the end of this work, which is a quarto of one hundred and sixty-five leaves, it is stated to have been begun by Henry Bechtermuntze, and finished by his brother Nicholas, and Wigand Spyess de Orthenberg.III.50 A second edition of the same work, printed by Nicholas Bechtermuntze, appeared in 1469. The following extract from a letter written by Fischer to Professor Zapf in 1803, contains an account of his researches respecting the Catholicon and Vocabulary: “The frankness with which you retracted your former opinions respecting the printer of the Catholicon of 1460, and agreed with me in assigning it to Gutemberg, demands the respect of every unbiassed inquirer. I beg now merely to mention to you a discovery that I have made which no longer leaves it difficult to conceive how the Catholicon types should have come into the hands of Bechtermuntze. From a monument which stands before the high altar of the church of Elfeld it is evident that the family of Sorgenloch, of which that of Gutemberg or Gænsfleisch was a branch, was connected with the family of Bechtermuntze by marriage. The types used by Bechtermuntze were not only similar to those formerly belonging to Gutemberg, but were the very same, as I always maintained, appealing to the principles of the type-founder’s art. They had come into the possession of Bechtermuntze by inheritance, on the death of Gutemberg, and hence Dr. Homery’s reclamation.”III.51

143

Zapf, to whom Fischer’s letter is addressed, had previously communicated to Oberlin his opinion that the types of the Catholicon were the same as those of an Augustinus de Vita Christiana, 4to, without date or printer’s name, but having at the end the arms of Faust and Scheffer. In his account, printed at Nuremberg, 1803, of an early edition of “Joannis de Turre-cremata explanatio in Psalterium,” he acknowledged that he was mistaken; thus agreeing with Schwartz, Meerman, Panzer, and Fischer, that no book known to be printed by Faust and Scheffer is printed with the same types as the Catholicon and the Vocabulary.

Although there can be little doubt of the Catholicon and the Elfeld Vocabulary being printed with the same types, and of the former being printed by Gutemberg, yet it is far from certain that Bechtermuntze inherited Gutemberg’s printing materials, even though he might be a relation. It is as likely that Gutemberg might sell to the brothers a portion of his materials and still retain enough for himself. If they came into their possession by inheritance, which is not likely, Gutemberg must have died some months previous to 4th November, 1467, the day on which Nicholas Bechtermuntze and Wygand Spyess finished the printing of the Vocabulary. If the materials had been purchased by Bechtermuntze in Gutemberg’s lifetime, which seems to be the most reasonable supposition, Conrad Homery could have no claim upon them on account of money advanced to Gutemberg, and consequently the types and printing materials which after his death came into Homery’s possession, could not be those employed by the brothers Bechtermuntze in their establishment at Elfeld.III.52

By letters patent, dated at Elfeld on St. Anthony’s day, 1465, Adolphus, archbishop and elector of Mentz, appointed Gutemberg one of his courtiers, with the same allowance of clothing as the rest of the nobles attending his court, with other privileges and exemptions. From this period Fischer thinks that Gutemberg no longer occupied himself with business as a printer, and that he transferred his printing materials to Henry Bechtermuntze. “If Wimpheling’s account be true,” says Fischer, “that Gutemberg became blind in his old age, we need no longer be surprised that during his lifetime his types and utensils should come into 144 the possession of Bechtermuntze.” The exact period of Gutemberg’s decease has not been ascertained, but in the bond or deed of obligation executed by Doctor Conrad Homery the Friday after St. Matthias’s day,III.53 1468, he is mentioned as being then dead. He was interred at Mentz in the church of the Recollets, and the following epitaph was composed by his relation, Adam Gelthaus:III.54

“D. O. M. S.

“Joanni Genszfleisch, artis impressoriæ repertori, de omni natione et lingua optime merito, in nominis sui memoriam immortalem Adam Gelthaus posuit. Ossa ejus in ecclesia D. Francisci Moguntina feliciter cubant.”

From the last sentence it is probable that this epitaph was not placed in the church wherein Gutemberg was interred. The following inscription was composed by Ivo Wittich, professor of law and member of the imperial chamber at Mentz:

“Jo. Guttenbergensi, Moguntino, qui primus omnium literas ære imprimendas invenit, hac arte de orbe toto bene merenti Ivo Witigisis hoc saxum pro monimento posuit M.D.VII.

This inscription, according to Serarius, who professes to have seen it, and who died in 1609, was placed in front of the school of law at Mentz. This house had formerly belonged to Gutemberg, and was supposed to be the same in which he first commenced printing at Mentz in conjunction with Faust.III.55

From the documentary evidence cited in the preceding account of the life of Gutemberg, it will be perceived that the art of printing with moveable types was not perfected as soon as conceived, but that it was a work of time. It is highly probable that Gutemberg was occupied with his invention in 1436; and from the obscure manner in which his “admirable discovery” is alluded to in the process between him and the Drytzehns in 1439, it does not seem likely that he had then proceeded beyond making experiments. In 1449 or 1450, when the sum of 800 florins was advanced by Faust, it appears not unreasonable to suppose that he had so far improved his invention, as to render it practically available without reference to Scheffer’s great improvement in casting the types from matrices formed by punches, which was most likely discovered between 1452 and 1455.III.56 About fourteen years must have 145 elapsed before Gutemberg was enabled to bring his invention into practice. The difficulties which must have attended the first establishment of typography could only have been surmounted by great ingenuity and mechanical knowledge combined with unwearied perseverance. After the mind had conceived the idea of using moveable types, those types, whatever might be the material employed, were yet to be formed, and when completed they were to be arranged in pages, divided by proper spaces, and bound together in some manner which the ingenuity of the inventor was to devise. Nor was his invention complete until he had contrived a Press, by means of which numerous impressions from his types might be perfectly and rapidly obtained.

Mr. Ottley, at page 285 of the first volume of his Researches, informs us that “almost all great discoveries have been made by accident;” and at page 196 of the same volume, when speaking of printing as the invention of Lawrence Coster, he mentions it as an “art which had been at first taken up as the amusement of a leisure hour, became improved, and was practised by him as a profitable trade.” Let any unbiassed person enter a printing-office; let him look at the single letters, let him observe them formed into pages, and the pages wedged up in forms; let him see a sheet printed from one of those forms by means of the press; and when he has seen and considered all this, let him ask himself if ever, since the world began, the amusement of an old man practised in his hours of leisure was attended with such a result? “Very few great discoveries,” says Lord Brougham, “have been made by chance and by ignorant persons, much fewer than is generally supposed.—They are generally made by persons of competent knowledge, and who are in search of them.”III.57

Having now given some account of the grounds on which Gutemberg’s claims to the invention of typography are founded, it appears necessary to give a brief summary, from the earliest authorities, of the pretensions of Lawrence Coster not only to the same honour, but to something more; for if the earliest account which we have of him be true, he was not only the inventor of typography, but of block-printing also.

The first mention of Holland in connexion with the invention of typography occurs in the Cologne Chronicle, printed by John Kœlhoff in 1499, wherein it is said that the first idea of the art was suggested by the Donatuses printed in Holland; it being however expressly stated in 146 the same work that the art of printing as then practised was invented at Mentz. In a memorandum, which has been referred to at page 123, written by Mariangelus Accursius, who flourished about 1530, the invention of printing with metal types is erroneously ascribed to Faust; and it is further added, that he derived the idea from a Donatus printed in Holland from a wood-block. That a Donatus might be printed there from a wood-block previous to the invention of typography is neither impossible nor improbable; although I esteem the testimony of Accursius of very little value. He was born and resided in Italy, and it is not unlikely, as has been previously observed, that he might derive his information from the Cologne Chronicle.

John Van Zuyren, who died in 1594, is said to have written a book to prove that typography was invented at Harlem; but it never was printed, and the knowledge that we have of it is from certain fragments of it preserved by Scriverius, a writer whose own uncorroborated testimony on this subject is not entitled to the slightest credit. The substance of Zuyren’s account is almost the same as that of Junius, except that he does not mention the inventor’s name. The art according to him was invented at Harlem, but that while yet in a rude and imperfect state it was carried by a stranger to Mentz, and there brought to perfection.

Theodore Coornhert, in the dedication of his Dutch translation of Tully’s Offices to the magistrates of Harlem, printed in 1561, says that he had frequently heard from respectable people that the art of printing was invented at Harlem, and that the house where the inventor lived was pointed out to him. He proceeds to relate that by the dishonesty of a workman the art was carried to Mentz and there perfected. Though he says that he was informed by certain respectable old men both of the inventor’s name and family, yet, for some reason or other, he is careful not to mention them. When he was informing the magistrates of Harlem of their city being the nurse of so famous a discovery, it is rather strange that he should not mention the parent’s name. From the conclusion of his dedication we may guess why he should be led to mention Harlem as the place where typography was invented. It appears that he and certain friends of his, being inflamed with a patriotic spirit, designed to establish a new printing-office at Harlem, “in honour of their native city, to the profit of others, and for their own accommodation, and yet without detriment to any person.” His claiming the invention of printing for Harlem was a good advertisement for the speculation.

The next writer who mentions Harlem as the place where printing was invented is Guicciardini, who in his Description of the Low Countries, first printed at Antwerp in 1567, gives the report, without vouching for its truth, as follows: “In this place, it appears, not only from the general opinion of the inhabitants and other Hollanders, but from the 147 testimony of several writers and from other memoirs, that the art of printing and impressing letters on paper such as is now practised, was invented. The inventor dying before the art was perfected or had come into repute, his servant, as they say, went to live at Mentz, where making this new art known, he was joyfully received; and applying himself diligently to so important a business, he brought it to perfection and into general repute. Hence the report has spread abroad and gained credit that the art of printing was first practised at Mentz. What truth there may be in this relation, I am not able, nor do I wish, to decide; contenting myself with mentioning the subject in a few words, that I might not prejudice [by my silence the claims of] this district.”III.58

It is evident that the above account is given from mere report. What other writers had previously noticed the claims of Harlem, except Coornhert and Zuyren, remain yet to be discovered. They appear to have been unknown to Guicciardini’s contemporary, Junius, who was the first to give a name to the Harlem inventor; a “local habitation” had already been provided for him by Coornhert.

The sole authority for one Lawrence Coster having invented wood-engraving, block-printing, and typography, is Hadrian Junius, who was born at Horn in North Holland, in 1511. He took up his abode at Harlem in 1560. During his residence in that city he commenced his Batavia,—the work in which the account of Coster first appeared,—which, from the preface, would seem to have been finished in January, 1575. He died the 16th June in the same year, and his book was not published until 1588, twelve years after his decease.III.59 In this work, which is a topographical and historical account of Holland, or more properly of the country included within the limits of ancient Batavia, we find the first account of Lawrence Coster as the inventor of typography. Almost every succeeding advocate of Coster’s pretensions has taken the liberty of 148 altering, amplifying, or contradicting the account of Junius according as it might suit his own line of argument; but not one of them has been able to produce a single solitary fact in confirmation of it. Scriverius, Seiz, Meerman, and Koning are fertile in their conjectures about the thief that stole Coster’s types, but they are miserably barren in their proofs of his having had types to be stolen. “If the variety of opinions,” observes Naude, speaking of Coster’s invention, “may be taken as an indication of the falsehood of any theory, it is impossible that this should be true”. Since Naude’s time the number of Coster’s advocates has been increased by Seiz, Meerman, and Koning;III.60 who, if they have not been able to produce any evidence of the existence of Lawrence Coster as a printer, have at least been fertile in conjectures respecting the thief. They have not strengthened but weakened the Costerian triumphal arch raised by Junius, for they have all more or less knocked a piece of it away; and even where they have pretended to make repairs, it has merely been “one nail driving another out.”

Junius’s account of Coster is supposed to have been written about 1568; and in order to do justice to the claims of Harlem I shall here give a faithful translation of the “document,”—according to Mr. Ottley,—upon which they are founded. After alluding, in a preliminary rhetorical flourish, to Truth being the daughter of Time, and to her being concealed in a well, Junius thus proceeds to draw her out.

“If he is the best witness, as Plutarch says, who, bound by no favour and led by no partiality, freely and fearlessly speaks what he thinks, my testimony may deservedly claim attention. I have no connexion through kindred with the deceased, his heirs, or his posterity, and I expect on this account neither favour nor reward. What I have done is performed through a regard to the memory of the dead. I shall therefore relate what I have heard from old and respectable persons who have held offices in the city, and who seriously affirmed that they had heard what they told from their elders, whose authority ought justly to entitle them to credit.”

About a hundred and twenty-eight years ago,III.61 Lawrence John, called the churchwarden or keeper,III.62 from the profitable and honourable office which his family held by hereditary right, dwelt in a large house, which is yet standing entire, opposite the Royal Palace. This is 149 the person who now on the most sacred ground of right puts forth his claims to the honour of having invented typography, an honour so nefariously obtained and possessed by others. Walking in a neighbouring wood, as citizens are accustomed to do after dinner and on holidays, he began to cut letters of beech-bark, with which for amusement, the letters being inverted as on a seal, he impressed short sentences on paper for the children of his son-in-law. Having succeeded so well in this, he began to think of more important undertakings, for he was a shrewd and ingenious man; and, in conjunction with his son-in-law Thomas Peter, he discovered a more glutinous and tenacious kind of ink, as he found from experience that the ink in common use occasioned blots. This Thomas Peter left four sons, all of whom were magistrates; and I mention this that all may know that the art derived its origin from a respectable and not from a mean family. He then printed whole figured pages with the text added. Of this kind I have seen specimens executed in the infancy of the art, being printed only on one side. This was a book composed in our native language by an anonymous author, and entitled Speculum Nostræ Salutis. In this we may observe that in the first productions of the art—for no invention is immediately perfected—the blank pages were pasted together, so that they might not appear as a defect. He afterwards exchanged his beech types for leaden ones, and subsequently he formed his types of tin, as being less flexible and of greater durability. Of the remains of these types certain old wine-vessels were cast, which are still preserved in the house formerly the residence of Lawrence, which, as I have said, looks into the market-place, and which was afterwards inhabited by his great-grandson Gerard Thomas, a citizen of repute, who died an old man a few years ago.

“The new invention being well received, and a new and unheard-of commodity finding on all sides purchasers, to the great profit of the inventor, he became more devoted to the art, his business was increased, and new workmen—the first cause of his misfortune—were employed. Among them was one called John; but whether, as is suspected, he bore the ominous surname of Faust,—infaustusIII.63 and unfaithful to his master—or whether it were some other John, I shall not labour to prove, as I do not wish to disturb the dead already enduring the pangs of conscience for what they had done when living.III.64 This person, who was admitted under an oath to assist in printing, as soon as he thought he had attained 150 the art of joining the letters, a knowledge of the fusile types, and other matters connected with the business, embracing the convenient opportunity of Christmas Eve, when all persons are accustomed to attend to their devotions, stole all the types and conveyed away all the utensils which his master had contrived by his own skill; and then leaving home with the thief, first went to Amsterdam, then to Cologne, and lastly to Mentz, as his altar of refuge, where being safely settled, beyond bowshot as they say, he might commence business, and thence derive a rich profit from the things which he had stolen. Within the space of a year from Christmas, 1442, it is certain that there appeared printed with the types which Lawrence had used at Harlem ‘Alexandri Galli Doctrinale,’ a grammar then in frequent use, with ‘Petri Hispani Tractatus.’

“The above is nearly what I have heard from old men worthy of credit who had received the tradition as a shining torch transferred from hand to hand, and I have heard the same related and affirmed by others. I remember being told by Nicholas Galius, the instructor of my youth,—a man of iron memory, and venerable from his long white hair,—that when a boy he had often heard one Cornelius, a bookbinder, not less than eighty years old (who had been an assistant in the same office), relate with such excited feelings the whole transaction,—the occasion of the invention, its progress, and perfection, as he had heard of them from his master,—that as often as he came to the story of the robbery he would burst into tears; and then the old man’s anger would be so roused on account of the honour that had been lost through the theft, that he appeared as if he could have hanged the thief had he been alive; and then again he would vow perdition on his sacrilegious head, and curse the nights that he had slept in the same bed with him, for the old man had been his bedfellow for some months. This does not differ from the words of Quirinus Talesius, who admitted to me that he had formerly received nearly the same account from the mouth of the same bookseller.”III.65

As Junius died upwards of twelve years before his book was published, it is doubtful whether the above account was actually written by him or not. It may have been an interpolation of an editor or a bookseller anxious for the honour of Harlem, and who might thus expect to gain currency for the story by giving it to the world under the sanction of Junius’s name. There was also another advantage attending this mode of publication; for as the reputed writer was dead, he could not be called on to answer the many objections which remain yet unexplained.

The manner in which Coster, according to the preceding account, first discovered the principle of obtaining impressions from separate 151 letters formed of the bark of the beech-tree requires no remark.III.66 There are, however, other parts of this narrative which more especially force themselves on the attention as being at variance with reason as well as fact.

Coster, we are informed, lived in a large house, and, at the time of his engaging the workman who robbed him, he had brought the art to such perfection that he derived from it a great profit; and in consequence of the demand for the new commodity, which was eagerly sought after by purchasers, he was obliged to increase his establishment and engage assistants. It is therefore evident that the existence of such an art must have been well known, although its details might be kept secret. Coster, we are also informed, was of a respectable family; his grand-children were men of authority in the city, and a great-grandson of his died only a few years before Junius wrote, and yet not one of his friends or descendants made any complaint of the loss which Coster had sustained both in property and fame. Their apathy, however, was compensated by the ardour of old Cornelius, who used to shed involuntary tears whenever the theft was mentioned; and used to heap bitter curses on the head of the thief as often as he thought of the glory of which Coster and Harlem had been so villanously deprived. It is certainly very singular that a person of respectability and authority should be robbed of his materials and deprived of the honour of the invention, and yet neither himself nor any one of his kindred publicly denounce the thief; more especially as the place where he had established himself was known, and where in conjunction with others he had the frontless audacity to claim the honour of the invention.

Of Lawrence Coster, his invention, and his loss, the world knew nothing until he had been nearly a hundred and fifty years in his grave. The presumed writer of the account which had to do justice to his memory had been also twelve years dead when his book was published. His information, which he received when he was a boy, was derived from an old man who when a boy had heard it from another old man who lived with Coster at the time of the robbery, and who had heard the account of the invention from his master. Such is the list of the Harlem witnesses. If Junius had produced any evidence on the authority 152 of Coster’s great-grandson that any of his predecessors—his father or his grandfather—had carried on the business of a printer at Harlem, this might in part have corroborated the narrative of Cornelius; but, though subsequent advocates of the claims of Harlem have asserted that Coster’s grand-children continued the printing business, no book or document has been discovered to establish the fact.

The account of Cornelius involves a contradiction which cannot be easily explained away. If the thief stole the whole or greater part of Coster’s printing materials,—types and press and all, as the narrative seems to imply,—it is difficult to conceive how he could do so without being discovered, even though the time chosen were Christmas Eve; for on an occasion when all or most people were engaged at their devotions, the fact of two persons being employed would in itself be a suspicious circumstance: a tenant with a small stock of furniture who wished to make a “moonlight flitting” would most likely be stopped if he attempted to remove his goods on a Sunday night. As the dishonest workman had an assistant, who is rather unaccountably called “the thief,” it is evident from this circumstance, as well as from the express words of the narrative,III.67 that the quantity of materials stolen must have been considerable. If, on the contrary, the thief only carried away a portion of the types and matrices, with a few other instruments,—“all that could be moved without manifest danger of immediate detection,” to use the words of Mr. Ottley,—what was there to prevent Coster from continuing the business of printing? Did he give up the lucrative trade which he had established, and disappoint his numerous customers, because a dishonest workman had stolen a few of his types? But even if every letter and matrice had been stolen,—though how likely this is to be true I shall leave every one conversant with typography to decide,—was the loss irreparable, and could this “shrewd and ingenious man” not reconstruct the types and other printing materials which he had originally contrived?

If the business of Coster was continued uninterruptedly, and after his death carried on by his grand-children, we might naturally expect that some of the works which they printed could be produced, and that some record of their having practised such an art at Harlem would be in existence. The records of Harlem are however silent on the subject; no mention is made by any contemporary author, nor in any contemporary document, of Coster or his descendants as printers in that city; and no book printed by them has been discovered except by persons who decide upon the subject as if they were endowed with the faculty of intuitive discrimination. If Coster’s business had been 153 suspended in consequence of the robbery, his customers, from all parts, who eagerly purchased the “new commodity,” must have been aware of the circumstance; and to suppose that it should not have been mentioned by some old writer, and that the claims of Coster should have lain dormant for a century and a half, exceeds my powers of belief. Where pretended truth can only be perceived by closing the eyes of reason I am content to remain ignorant; nor do I wish to trust myself to the unsafe bridge of conjecture—a rotten plank without a hand-rail,—

“O’er which lame faith leads understanding blind.”

If all Coster’s types had been stolen and he had not supplied himself with new ones, it would be difficult to account for the wine vessels which were cast from the old types; and if he or his heirs continued to print subsequent to the robbery, all that his advocates had to complain of was the theft. For since it must have been well known that he had discovered and practised the art, at least ten years previous to its known establishment at Mentz, and seventeen years before a book appeared with the name of the printers claiming the honour of the invention, the greatest injury which he received must have been from his fellow citizens; who perversely and wilfully would not recollect his previous discovery and do justice to his claims. Even supposing that a thief had stolen the whole of Coster’s printing-materials, types, chases, and presses, it by no means follows that he deprived of their memory not only all the citizens of Harlem, but all Coster’s customers who came from other placesIII.68 to purchase the “new commodity” which his press supplied. Such however must have been the consequences of the robbery, if the narrative of Cornelius were true; for except himself no person seems to have remembered Coster’s invention, or that either he or his immediate descendants had ever printed a single book.

Notwithstanding the internal evidence of the improbability of Cornelius’s account of Coster and his invention, its claims to credibility are still further weakened by those persons who have shown themselves most wishful to establish its truth. Lawrence Janszoon, whom Meerman and others suppose to have been the person described by Cornelius as the inventor of printing, appears to have been custos of the church of St. Bavon at Harlem in the years 1423, 1426, 1432 and 1433. His death is placed by Meerman in 1440; and as, according to the narrative of Cornelius, the types and other printing materials were stolen on Christmas eve 1441, the inventor of typography must have been in his grave at the time the robbery was committed. Cornelius must have known of his master’s death, and yet in his account of the robbery he 154 makes no mention of Coster being dead at the time, nor of the business being carried on by his descendants after his decease. It was at one time supposed that Coster died of grief on the loss of his types, and on account of the thief claiming the honour of the invention. But this it seems is a mistake; he was dead according to Meerman at the time of the robbery, and the business was carried on by his grandchildren.

Koning has discovered that Cornelius the bookbinder died in 1522, aged at least ninety years. Allowing him to have been ninety-two, this assistant in Coster’s printing establishment, and who learnt the account of the invention and improvement of the art from Coster himself, must have been just ten years old when his master died; and yet upon the improbable and uncorroborated testimony of this person are the claims of Coster founded.

Lehne, in his “Chronology of the Harlem fiction,”III.69 thus remarks on the authorities, Galius, and Talesius, referred to by Junius as evidences of its truth. As Cornelius was upwards of eighty when he related the story to Nicholas Galius, who was then a boy, this must have happened about 1510. The boy Galius we will suppose to have been at that time about fifteen years old: Junius was born in 1511, and we will suppose that he was under the care of Nicholas Galius, the instructor of his youth, until he was fifteen; that is, until 1526. In this year Galius, the man venerable from his grey hairs, would be only thirty-six years old, an age at which grey hairs are premature. Grey hairs are only venerable in old age, and it is not usual to praise a young man’s faculty of recollection in the style in which Junius lauds the “iron memory” of his teacher. Talesius, as Koning states, was born in 1505, and consequently six years older than Junius; and on the death of Cornelius, in 1522, he would be seventeen, and Junius eleven years old. Junius might in his eleventh year have heard the whole account from Cornelius himself in the same manner as the latter when only ten must have heard it from Coster; and it is remarkable that Galius who was so well acquainted with Cornelius did not afford his pupil the opportunity. We thus perceive that in the whole of this affair children and old men play the principal parts, and both ages are proverbially addicted to narratives which savour of the marvellous.

Meerman, writing to his correspondent Wagenaar in 1757, expresses his utter disbelief in the story of Coster being the inventor of typography, which, he observes, was daily losing credit: whatever historical evidence Seiz had brought forward in favour of Coster was gratuitously 155 assumed; in short, the whole story of the invention was a fiction.III.70 After the publication of Schœpflin’s Vindiciæ Typographicæ in 1760, giving proofs of Gutemberg having been engaged in 1438 with some invention relating to printing, and in which a press was employed, Meerman appears to have received a new light; for in 1765 he published his own work in support of the very story which he had previously declared to be undeserving of credit. The mere change, however, of a writer’s opinions cannot alter the immutable character of truth; and the guesses and assumptions with which he may endeavour to gloss a fiction can never give to it the solidity of fact. What he has said of the work of Seiz in support of Coster’s claims may with equal truth be applied to his own arguments in the same cause: “Whatever historical evidence he has brought forward in favour of Coster has been gratuitously assumed.” Meerman’s work, like the story which it was written to support, “is daily losing credit.” It is a dangerous book for an advocate of Coster to quote; for he has scarcely advanced an argument in favour of Coster, and in proof of his stolen types being the foundation of typography at Mentz, but what is contradicted by a positive fact.

In order to make the documentary evidence produced by Schœpflin in favour of Gutemberg in some degree correspond with the story of Cornelius, Junius’s authority, he has assumed that Gutemberg had an elder brother also called John; and that he was known as Gænsfleisch the elder, while his younger brother was called by way of distinction Gutemberg. In support of this assumption he refers to Wimpheling,III.71 156 who in one place has called the inventor Gænsfleisch, and in another Gutemberg; and he also supposes that the two epitaphs which have been given at page 144, relate to two different persons. The first, inscribed by Adam Gelthaus to the memory of John Gænsfleisch, he concludes to have been intended for the elder brother. The second, inscribed by Ivo Wittich to the memory of John Gutemberg, he supposes to relate to the younger brother, and to have been erected from a feeling of envy. The fact of Gutemberg being also named Gænsfleisch in several contemporary documents, is not allowed to stand in the way of Meerman’s hypothesis of the two “brother Johns,” which has been supposed to be corroborated by the fact of a John Gænsfleisch the Elder being actually the contemporary of John Gænsfleisch called also Gutemberg.

Having thus provided Gutemberg with an elder brother also named John, Meerman proceeds to find him employment; for at the period of his writing much light had been thrown on the early history of printing, and no person in the least acquainted with the subject could believe that Faust was the thief who stole Coster’s types, as had been insinuated by Junius and affirmed by Boxhorn and Scriverius. Gænsfleisch the Elder is accordingly sent by Meerman to Harlem, and there engaged as a workman in Lawrence Coster’s printing office. It is needless to ask if there be any proof of this: Meerman having introduced a new character into the Harlem farce may claim the right of employing him as he pleases. As there is evidence of Gutemberg, or Gænsfleisch the Younger, being engaged at Strasburg about 1436 in some experiments connected with printing, and mention being made in the same documents of the fair of Aix-la-Chapelle, Meerman sends him there in 1435. From Aix-la-Chapelle, as the distance is not very great, Meerman makes him pay a visit to his elder brother, then working as a printer in Coster’s office at Harlem. He thus has an opportunity of seeing Coster’s printing establishment, and of gaining some information respecting the art, and hence his attempts at printing at Strasburg in 1436. In 1441 he supposes that John Gænsfleisch the Elder stole his master’s types, and printed with them, at Mentz, in 1442, “Alexandri Galli Doctrinale,” and “Petri Hispani Tractatus,” as related by Junius. As this trumpery story rests solely on the conjecture of the writer, it might be briefly dismissed for reconsideration when the proofs should be produced; but as HeinekenIII.72 has afforded the means of showing its utter falsity, it may perhaps be worth while to notice some of the facts produced by him respecting the family and proceedings of Gutemberg.

John Gænsfleisch the Elder, whom Meerman makes Gutemberg’s elder brother, was descended from a branch of the numerous family of 157 Gænsfleisch, which was also known by the local names of zum Jungen, Gutenberg or Gutemberg, and Sorgenloch. This person, whom Meerman engages as a workman with Coster, was a man of property; and at the time that we are given to understand he was residing at Harlem, we have evidence of his being married and having children born to him at Mentz. This objection, however, could easily be answered by the ingenuity of a Dutch commentator, who, as he has made the husband a thief, would find no difficulty in providing him with a suitable wife. He would also be very likely to bring forward the presumed misconduct of the wife in support of his hypothesis of the husband being a thief. John Gænsfleisch the Elder was married to Ketgin, daughter of Nicholas Jostenhofer of Schenkenberg, on the Thursday after St. Agnes’s day, 1437. In 1439 his wife bore him a son named Michael; and in 1442 another son, who died in infancy. In 1441 we have evidence of his residing at Mentz; for in that year his relation Rudiger zum Landeck appeared before a judge to give Gænsfleisch an acknowledgment of his having properly discharged his duties as trustee, and of his having delivered up to the said Rudiger the property left to him by his father and mother.

That John Gænsfleisch the Elder printed “Alexandri Galli Doctrinale,” and “Petri Hispani Tractatus,” at Mentz in 1442 with the types which he had stolen from Coster, is as improbable as every other part of the story. There is, in fact, not the slightest reason to believe that the works in question were printed at Mentz in 1442, or that any book was printed there with types until nearly eight years after that period. In opposition, however, to a host of historical evidence we have the assertion of Cornelius, who told the tale to Galius, who told it to Junius, who told it to the world.

Meerman’s web of sophistry and fiction having been brushed away by Heineken, a modern advocate of Coster’s undertook to spin another, which has also been swept down by a German critic. Jacobus Koning,III.73 town-clerk of Amsterdam, having learnt from a document printed by Fischer, that Gutemberg had a brother named Friele, sends him to Harlem to work with Coster, and makes him the thief who stole the types; thus copying Meerman’s plot, and merely substituting Gutemberg’s known brother for John Gænsfleisch the Elder. On this attempt of Koning’s to make the old sieve hold water by plastering it with his own mud, LehneIII.74 makes the following remarks:—

“He gives up the name of John,—although it might be supposed that old Cornelius would have known the name of his bedfellow better 158 than Koning,—and without hesitation charges Gutemberg’s brother with the theft. In order to flatter the vain-glory of the Harlemers, poor Friele, after he had been nearly four hundred years in his grave, is publicly accused of robbery on no other ground than that Mynheer Koning had occasion for a thief. It is, however, rather unfortunate for the credit of the story that this Friele should have been the founder of one of the first families in Mentz, of the order of knighthood, and possessed of great property both in the city and the neighbourhood. Is it likely that this person should have been engaged as a workman in the employment of the Harlem churchwarden, and that he should have robbed him of his types in order to convey them to his brother, who then lived at Strasburg, and who had been engaged in his own invention at least three years before, as is proved by the process between him and the Drytzehns published by Schœpflin? From this specimen of insulting and unjust accusation on a subject of literary inquiry, we may congratulate the city of Amsterdam that Mynheer Koning is but a law-writer and not a judge, should he be not more just as a man than as an author.”

In a book of old accounts belonging to the city of Harlem, and extending from April 1439 to April 1440, Koning having discovered at least nine entries of expenses incurred on account of messengers despatched to the Justice-Court of Amsterdam, he concludes that there must have been some conference between the judges of Harlem and Amsterdam on the subject of Coster’s robbery. There is not a word mentioned in the entries on what account the messengers were despatched, but he decides that it must have been on some business connected with this robbery, for the first messenger was despatched on the last day of the Christmas holidays; and the thief, according to the account of Junius, made choice of Christmas-eve as the most likely opportunity for effecting his purpose. To this most logical conclusion there happens to be an objection, which however Mynheer Koning readily disposes of. The first messenger was despatched on the last day of the Christmas holidays 1439, and the accounts terminate in April 1440; but according to the narrative of Cornelius the robbery was committed on Christmas-eve 1441. This trifling discrepancy is however easily accounted for by the fact of the Dutch at that period reckoning the commencement of the year from Easter, and by supposing,—as the date is printed in numerals,—that Junius might have written 1442, instead of 1441, as the time when the two books appeared at Mentz printed with the stolen types, and within a year after the robbery. Notwithstanding this satisfactory explanation there still remains a trifling error to be rectified, and it will doubtless give the clear-headed advocate of Coster very little trouble. Admitting that the accounts are for the year commencing at Easter 1440 159 and ending at Easter 1441, it is rather difficult to comprehend how they should contain any notice of an event which happened at the Christmas following. The Harlem scribe possibly might have the gift of seeing into futurity as clearly as Mynheer Koning has the gift of seeing into the past. The arguments derived from paper-marks which Koning has advanced in favour of Coster are not worthy of serious notice.

He has found, as Meerman did before him, that one Lawrence Janszoon was living in Harlem between 1420 and 1436, and that his name occurs within that period as custos or warden of St. Bavon’s church. As he is never called “Coster,” a name acquired by the family, according to Junius, in consequence of the office which they enjoyed by hereditary right, the identity of Lawrence Janszoon and Lawrence Coster is by no means clearly established; and even if it were, the sole evidence of his having been a printer rests on the testimony of Cornelius, who was scarcely ten years old when Lawrence Janszoon died. The correctness of Cornelius’s narrative is questioned both by Meerman and Koning whenever his statements do not accord with their theory, and yet they require others to believe the most incredible of his assertions. They themselves throw doubts on the evidence of their own witness, and yet require their opponents to receive as true his deposition on the most important point in dispute—-that Coster invented typography previous to 1441,—a point on which he is positively contradicted by more than twenty authors who wrote previous to 1500; and negatively by the silence of Coster’s contemporaries. Supposing that the account of Cornelius had been published in 1488 instead of 1588, it would be of very little weight unless corroborated by the testimony of others who must have been as well aware of Coster’s invention as himself; for the silence of contemporary writers on the subject of an important invention or memorable event, will always be of greater negative authority than the unsupported assertion of an individual who when an old man professes to relate what he had heard and seen when a boy. If therefore the uncorroborated testimony of Cornelius would be so little worth, even if published in 1488, of what value can it be printed in 1588, in the name of a person who was then dead, and who could not be called on to explain the discrepancies of his part of the narrative? Whatever might be the original value of Cornelius’s testimony, it is deteriorated by the channel through which it descends to us. He told it to a boy, who, when an old man, told it to another boy, who when nearly sixty years old inserts it in a book which he is writing, but which is not printed until twelve years after his death.

It is singular how Mr. Ottley, who contends for the truth of Papillon’s story of the Cunio, and who maintains that the art of 160 engraving figures and text upon wood was well known and practised previous to 1285, should believe the account given by Cornelius of the origin of Coster’s invention. If he does not believe this part of the account, with what consistency can he require other people to give credit to the rest? With respect to the origin and progress of the invention, Cornelius was as likely to be correctly informed as he was with regard to the theft and the establishment of printing at Mentz; if therefore Coster’s advocates themselves establish the incorrectness of his testimony in the first part of the story, they destroy the general credibility of his evidence.

With respect to the fragments of “Alexandri Galli Doctrinale” and “Catonis Disticha” which have been discovered, printed with the same, or similar types as the Speculum Salvationis, no good argument can be founded on them in support of Coster’s claims, although the facts which they establish are decisive of the fallacy of Meerman’s assumptions. In order to suit his own theory, he was pleased to assert that the first edition of the Speculum was the only one of that book printed by Coster, and that it was printed with wooden types. Mr. Ottley has, however, shown that the edition which Meerman and others supposed to be the first was in reality the second; and that the presumed second was unquestionably the first, and that the text was throughout printed with metal types by means of a press. It is thus the fate of all Coster’s advocates that the last should always produce some fact directly contradicting his predecessors’ speculations, but not one confirmatory of the truth of the story on which all their arguments are based. Meerman questions the accuracy of Cornelius as reported by Junius; Meerman’s arguments are rejected by Koning; and Mr. Ottley, who espouses the same cause, has from his diligent collation of two different editions of the Speculum afforded a convincing proof that on a most material point all his predecessors are wrong. His inquiries have established beyond a doubt, that the text of the first edition of the Speculum was printed wholly with metal types; and that in the second the text was printed partly from metal types by means of a press, and partly from wood-blocks by means of friction. The assertion that Coster printed the first edition with wooden types, and that his grandsons and successors printed the second edition with types of metal, is thus most clearly refuted. As no printer’s name has been discovered in any of the fragments referred to, it is uncertain where or when they were printed. It however seems more likely that they were printed in Holland or the Low Countries than in Germany. The presumption of their antiquity in consequence of their rarity is not a good ground of argument. Of an edition of a “Donatus,” printed by Sweinheim 161 and Pannartz, between 1465 and 1470, and consisting of three hundred copies, not one is known to exist. From sundry fragments of a “Donatus,” embellished with the same ornamented small capitals as are used in Faust and Scheffer’s Psalter, Fischer was pleased to conjecture that the book had been printed by Gutemberg and Faust previous to 1455. A copy, however, has been discovered bearing the imprint of Scheffer, and printed, in all probability, subsequent to 1467, as it is in this year that Scheffer’s name first appears alone. The “Historia Alexandri Magni,” pretendedly printed with wooden types, and ascribed by Meerman to Coster, was printed by Ketelar and Leempt, who first established a printing-office at Utrecht in 1473.

John Enschedius, a letter-founder and printer of Harlem, and a strenuous assertor of Coster’s pretensions, discovered a very curious specimen of typography which he and others have supposed to be the identical “short sentences” mentioned by Junius as having been printed by Coster for the instruction of his grand-children. This unique specimen of typography consists of eight small pages, each being about one inch and six-eighths high, by one and five-eighths wide, printed on parchment and on both sides. The contents are an alphabet; the Lord’s Prayer; the Creed; the Ave Mary; and two short prayers, all in Latin. Meerman has given a fac-simile of all the eight pages in the second volume of his “Origines Typographicæ;”III.75 and if this be correct, I am strongly inclined to suspect that this singular “Horarium” is a modern forgery. The letters are rudely formed, and the shape of some of the pages is irregular; but the whole appears to me rather as an imitation of rudeness and a studied irregularity, than as the first essay of an inventor. There are very few contractions in the words; and though the letters are rudely formed, and there are no points, yet I have seen no early specimen of typography which is so easy to read. It is apparent that the printer, whoever he might be, did not forget that the little manual was intended for children. The letters I am positive could not be thus printed with types formed of beech-bark; and I am further of opinion that they were not, and could not be, printed with moveable types of wood. I am also certain that, whatever might be the material of which the types were formed, those letters could only be printed on 162 parchment on both sides by means of a press. The most strenuous of Coster’s advocates have not ventured to assert that he was acquainted with the use of metal types in 1423, the pretended date of his first printing short sentences for the use of his grand-children, nor have any of them suggested that he used a press for the purpose of obtaining impressions from his letters of beech-bark; how then can it be pretended with any degree of consistency that this “Horarium” agrees exactly with the description of Cornelius? It is said that Enschedius discovered this singular specimen of typography pasted in the cover of an old book. It is certainly such a one as he was most wishful to find, and which he in his capacity of typefounder and printer would find little difficulty in producing. I am firmly convinced that it is neither printed with wooden types nor a specimen of early typography; on the contrary, I suspect it to be a Dutch typographic essay on popular credulity.

Of all the works which have been claimed for Coster, his advocates have not succeeded in making out his title to a single one; and the best evidence of the fallacy of his claims is to be found in the writings of those persons by whom they have been most confidently asserted. Having no theory of my own to support, and having no predilection in favour of Gutemberg, I was long inclined to think that there might be some rational foundation for the claims which have been so confidently advanced in favour of Harlem. An examination, however, of the presumed proofs and arguments adduced by Coster’s advocates has convinced me that the claims put forward on his behalf, as the inventor of typography, are untenable. They have certainly discovered that a person of the name of Lawrence Janszoon was living at Harlem between the years 1420 and 1440, but they have not been able to show anything in proof of this person ever having printed any book either from wood-blocks or with moveable types. There is indeed reason to believe that at the period referred to there were three persons of the name of Lawrence Janszoon,—or Fitz-John, as the surname may be rendered;—but to which of them the pretended invention is to be ascribed is a matter of doubt. At one time we find the inventor described as an illegitimate scion of the noble family of Brederode, which was descended from the ancient sovereigns of Holland; at another he is said to have been called Coster in consequence of the office of custos or warden of St. Bavon’s church being hereditary in his family; and in a third account we find Lawrence Janszoon figuring as a promoter of sedition and one of the leaders of a body of rioters. The advocates for the claims of Harlem have brought forward every Lawrence that they could find at that period whose father’s name was John; as if the more they could produce the more conclusive would be 163 the proof of one of them at least being the inventor of printing. As the books which are ascribed to Coster furnish positive evidence of the incorrectness of the story of Cornelius and of the comments of Meerman; and as records, which are now matters of history, prove that neither Gutemberg nor Faust stole any types from Coster or his descendants, the next supporter of the claims of Harlem will have to begin de novo; and lest the palm should be awarded to the wrong Lawrence Janszoon, he ought first to ascertain which of them is really the hero of the old bookbinder’s tale.

  see text

III.1 Nouvelles Recherches sur l’origine de l’Imprimerie, dans lesquelles on fait voir que la première idée est due aux Brabançons. Par M. Desroches. Lu à la séance du 8 Janvier, 1777.—Mémoires de l’Academie Impériale des Sciences et Belles Lettres, tom. i. pp. 523-547. Edit 1780.

III.2 The following is the French translation of Monsieur Desroches: “En ces temps mourut de la mort commune à tous les hommes, Louis cet excellent faiseur d’instrumens de musique, le meilleur artist qu’on eut vû jusques-là dans l’univers, en fait d’ouvrages mechaniques. Il étoit de Vaelbeke en Brabant, et il en porta le nom. Il fut le premier qui inventa la manière d’imprimer, qui est presentement en usage.” The reason of Monsieur Desroches for his periphrasis of the simple word “vedelare”—fidler—is as follows: “J’ai rendu Vedelare par ‘faiseur d’instrumens de musique.’ Le mot radical est vedel, violin: par consequent, Vedelare doit signifier celui qui en joue, ou qui en fait. Je me suis determiné pour le dernier à cause des vers suivans, où il n’est point question de jouer mais de faire. Si l’on préfère le premier, je ne m’y opposerai pas; rien empêche que ce habile homme n’ait été musicien.”—Mem. de l’Acad. de Brux. tom. i. p. 536.

III.3 Lettre de M. J. G[hesquiere] à M. l’Abbé Turberville Needham, directeur de l’Academie Impériale et Royale de Bruxelles.—Printed in l’Esprit des Journaux for June 1779, pp. 232-260.

III.4 Lichtenberger, Initia Typographica, § De Prenteris ante inventam Typographiam, p. 140.—Lambinet, Recherches sur l’Origine de l’Imprimerie, p. 115.

III.5 Reflexions sur deux pièces relatives à l’Hist. de l’Imprimerie. Nivelles, 1780.—Lambinet, Recherches, p. 394.

III.6 Friedrich Lehne, Einige Bemerkungen über das Unternehmen der gelehrten Gesellschaft zu Harlem, ihrer Stadt die Ehre der Erfindung der Buchdruckerkunst zu ertrotzen, S. 24-26. Zweite Ausgabe, Mainz. 1825.

III.7 This is a mistake into which the compiler of the chronicle printed at Rome, 1474, by Philippus de Lignamine, has also fallen. Gutemberg was not a native of Strasburg, but of Mentz.

III.8 Mallinkrot appears to have been the first who gave a translation of the entire passage in the Cologne Chronicle which relates to the invention of printing. His version of the last sentence is as follows: “Reperiuntur Scioli aliquot qui dicant, dudum ante hæc tempora typorum ope libros excusos esse, qui tamen et se et alios decipiunt; nullibi enim terrarum libri eo tempore impressi reperiuntur.”—De Ortu et Progressu Artis Typographicæ, p. 38. Colon. Agrippinæ, 1640.

III.9 Angelus Rocca mentions having seen a “Donatus” on parchment, at the commencement of which was written in the hand of Mariangelus Accursius, who flourished about 1530: “Impressus est autem hic Donatus et Confessionalia primùm omnium anno MCCCCL. Admonitus certè fuit ex Donato Hollandiæ, prius impresso in tabula incisa.”—Bibliotheca Vaticana commentario illustrata, 1591, cited by Prosper Marchand in his Hist. de l’Imprimerie, 2nde Partie, p. 35. It is likely that Accursius derived his information about a Donatus being printed in Holland from the Cologne Chronicle.

III.10 Schwartz observes that in the instrument drawn up by the notary Ulric Helmasperger, Gutemberg is styled “Juncker,” an honourable addition which was at that period expressive of nobility.—Primaria quædam Documenta de Origine Typographiæ, p. 20, 4to. Altorfii, 1740.

III.11 “Morabatur autem prædictus Joannes Gutenberg Moguntiæ in domo zum Jungen, quæ domus usque in præsentem diem [1513] illius novæ Artis nomine noscitur insignita.”—Trithemii Chronicum Spanhemiense, ad annum 1450.

III.12 In the release which he grants to the town-clerk of Mentz, in 1434, he describes himself as, “Johann Gensefleisch der Junge, genant Gutemberg.”

III.13 In “Het derde Jubeljaer der uitgevondene Boekdrukkonst door Laurens Jansz Koster,” p. 71. Harlem, 1740.—Oberlin, Essai d’Annales.

III.14 The release is given in Schœpflin’s Vindiciæ Typographicæ, Documentum I.

III.15Ennelin zu der Iserin Thure.” She was then living at Strasburg, and was of an honourable family, originally of Alsace.—Schœpflin. Vind. Typ. p. 17.

III.16 When Andrew Heilman was proposed as a partner, Gutemberg observed that his friends would perhaps treat the business into which he was about to embark as mere jugglery [göckel werck], and object to his having anything to do with it.—Schœpflin, Vind. Typ. Document. p. 10.

III.17 This decision is dated “On the Eve of St. Lucia and St. Otilia, [12th December,] 1439.”

III.18 Traité de l’origine et des productions de l’Imprimerie primitive en taille de bois, Paris, 1758; et Remarques sur un Ouvrage, &c. pour servir de suite au Traité, Paris, 1762.

III.19 “Andres Dritzehn uwer bruder selige hat iiij stücke undenan inn einer pressen ligen, da hat uch Hanns Gutemberg gebetten das ir die darusz nement ünd uff die presse legent von einander so kan man nit gesehen was das ist.”—Schœpflin, Vind. Typ. Document. p. 6.

III.20 “Nym die stücke usz der pressen und zerlege sü von einander so weis nyemand was es ist:” literally: “Take the pieces out of the press and distribute [or separate] them, so that no man may know what it is.”—Schœpflin, Vind. Typ. Document. p. 6. “The word zerlegen,” says Lichtenberger, Initia Typograph. p. 11, “is used at the present day by printers to denote the distribution of the types which the compositor has set up.” The original word “stücke”—pieces—is always translated “paginæ”—pages—by Schelhorn. Dr. Dibdin calls them “forms kept together by two screws or press-spindles.”—Life of Caxton, in his edition of Ames’s and Herbert’s Typ. Antiq. p. lxxxvii. note.

III.21 St. Stephen’s Day is on 26th December. Andrew Drytzehn, being very ill, confessed himself to Peter Eckhart on Christmas-day, 1438, and it would seem that he died on the 27th.

III.22 “Dirre gezuge hat ouch geseit das er wol wisse das Gutenberg unlange vor Wihnahten sinen kneht sante zu den beden Andresen, alle formen zu holen, und würdent zur lossen das er ess sehe, un jn joch ettliche formen ruwete.”—Schœpflin, Vind. Typ. Document. p. 12. The separate letters, which are now called “types,” were frequently called “formæ” by the early printers and writers of the fifteenth century. They are thus named by Joh. and Vindelin de Spire in 1469; by Franciscus Philelphus in 1470; by Ludovicus Carbo in 1471; and by Phil. de Lignamine in 1474.—Lichtenberger, Init. Typ. p. 11.

III.23 “Hanns Dünne der goltsmyt hat gesait, das er vor dryen jaren oder daby Gutemberg by den hundert guldin verdienet habe, alleine das zu dem trucken gehöret”—Schœpflin, Vind. Typ. Document. p. 13.

III.24 The words of Bär, who was almoner of the Swedish chapel at Paris in 1761, are these: “Tout le monde sait que dans ce temps les orfèvres exerçoient aussi l’art de la gravûre; et nous concluons de-là que Guttemberg a commencé par des caractères de bois, que de-là il a passé aux caractères de plomb.” On this passage Fournier makes the following observations: “Tout le monde sait au contraire que dans ce temps il n’y avoit pas un seul graveur dans le genre dont vous parlez, et cela par une raison bien simple: c’est que cet art de la gravûre n’a été inventé que vingt-trois ans après ce que vous citez, c’est-à-dire en 1460, par Masso Piniguera.”—Remarques, &c. p. 20. Bär mentioned no particular kind of engraving; and the name of the Italian goldsmith who is supposed to have been the first who discovered the art of taking impressions from a plate on paper, was Finiguerra, not Piniguera, as Fournier, with his usual inaccuracy, spells it.

III.25 Essai d’Annales de la Vie de Jean Gutenberg, par Jer. J. Oberlin. 8vo. Strasbourg, An ix. [1802.]

III.26 Trithemii Annales Hirsaugienses, tom. ii. ad annum 1450. The original passage is printed in Prosper Marchand’s Histoire de l’Imprimerie, 2nde Partie, p. 7.

III.27 Vindiciæ Typographicæ, pp. 77, 78.

III.28 In the first work which issued from Faust and Scheffer’s press, with a date and the printer’s names,—the Psalter of 1457,—and in several others, Scheffer appears on an equal footing with Faust. In the colophon of an edition of Cicero de Officiis, 1465, Faust has inserted the following degrading words: “Presens opus Joh. Fust Moguntinus civis . . . . arte quadam perpulcra Petri manu pueri mei feliciter effeci.” His partner, to whose ingenuity he is chiefly indebted for his fame, is here represented in the character of a menial. Peter Scheffer, of Gernsheim, clerk, who perfected the art of printing, is now degraded to “Peter, my boy” by whose hand—not by his ingenuity—John Faust exercises a certain beautiful art.

III.29 Henry Keffer was employed in Gutemberg and Faust’s printing-office. He afterwards went to Nuremberg, where his name appears as a printer, in 1473, in conjunction with John Sensenschmid.—Primaria quædam Documenta de origine Typographiæ, edente C. G. Schwartzio. 8vo. Altorfii, 1740.

III.30 “Er [Johan Fust] denselben solt fürter under Christen und Iudden hab müssen ussnemen, und davor sess und dreyssig Gulden ungevärlich zu guter Rechnung zu Gesuch geben, das sich also zusamen mit dem Hauptgeld ungevärlich trifft an zvvytusend und zvvanzig Gulden.” Schwartz in an observation upon this passage conceives the sum of 2,020 florins to be thus made up: capital advanced, in two sums of 800 each, 1,600 florins: interest 390; on account of compound interest, incurred by Faust, 36; making in all 2,026. He thinks that 2,020 florins only were claimed as a round sum; and that the second sum of 800 florins was advanced in October 1452.—Primaria quædam Documenta, pp. 9-14.

III.31 “. . . . und das Johannes [Fust] ym ierlichen 300 Gulden vor Kosten geben, und auch Gesinde Lone, Huss Zinss, Vermet, Papier, Tinte, &c. verlegen solte.” Primaria qæedam Doc. p. 10.

III.32 “. . . . von den ubrigen 800 Gulden vvegen begert er ym ein rechnung zu thun, so gestett er auch ym keins Soldes noch Wuchers, und hofft ym im rechten darum nit pflichtigk sin.” Primaria quædam Doc. p. 11.

III.33 Mercier, who is frequently referred to as an authority on subjects connected with Bibliography, has, in his supplement to Prosper Marchand’s Histoire de l’Imprimerie, confounded this document with that containing an account of the process between the Drytzehns and Gutemberg at Strasburg in 1439; and Heineken, at p. 255 of his Idée Générale, has committed the same mistake.

III.34 “Je crois, que ces tables [deux planches de bois autrefois chez le Duc de la Valliere] sont du livre que le Chroniqueur de Cologne appelle un Donat et que Trithem nomme un Catholicon, (livre universel,) ce qu’on a confondu ensuite avec le grand ouvrage intitulé Catholicon Januensis.”—Idée Générale, p. 258.

III.35 Oberlin, Essai d’Annales de la Vie de Gutenberg.

III.36 “. . . . ligneos typos, ex buxi frutice, perforatos in medio, ut zona colligari una jungique commode possint, ex Fausti officina reliquos, Moguntiæ aliquando me conspexisse memini.”—Paulus Pater, in Dissertatione de Typis Literarum, &c. p, 10. 4to. Lipsiæ, 1710. Heineken, at p. 254 of his Idée Gén., declares himself to be convinced that Gutemberg had cut separate letters on wood, but he thinks that no person would be able to cut a quantity sufficient to print whole sheets, and, still less, large volumes as many pretend.

III.37 Oberlin, Essai d’Annales de la Vie de Gutenberg.

III.38 Dr. Dibdin, Bibliog. Tour, vol iii. p. 135, second edition.

III.39 Gotthelf Fischer, Notice du premier livre imprimé avec date. 4to. Mayence, An xi. Typographisch. Seltenheit. 6te. Lieferung, S. 25. 8vo. Nürnberg, 1804. When Fischer published his account of the Calendar, Aretin had not discovered the tract entitled “Eyn Manung der Cristenheit widder die durken.”

III.40 It is called the Mazarine Bible in consequence of the first known copy being discovered in the library formed by Cardinal Mazarine. Dr. Dibdin, in his Bibliographical Tour, vol. ii. p. 191, mentions having seen not fewer than ten or twelve copies of this edition, which he says must not be designated as “of the very first degree of rarity.” An edition of the Bible, supposed to have been printed at Bamberg by Albert Pfister about 1461, is much more scarce.

III.41 In most of the early printed books the capitals were left to be inserted in red ink by the pen or pencil of the “rubricator.”

III.42 There are fac-simile tracings of those memorandums, on separate slips of paper, in the copy of the Mazarine Bible in the King’s Library at the British Museum; and fac-simile engravings of them are given in the M’Carthy Catalogue.

III.43 Typograph. Seltenheit. S. 20, 3te. Lieferung.

III.44 Recherches sur l’Origine de l’Imprimerie, p. 135.

III.45 Oberlin says that “Ville-Ostein” lies near Erfurth, and is in the diocese of Mentz.

III.46 Index librorum sub incunabula typograph. impressorum. 1739; cited by Fischer, Typograph. Seltenheit. S. 21, 3te. Lieferung.

III.47 Philippi de Lignamine Chronica Summorum Pontificum Imperatorumque, anno 1474, Romæ impressa. A second edition of this chronicle was printed at Rome in 1476 by “Schurener de Bopardia.” In both editions Gutemberg is called “Jacobus,”—James, and is said to be a native of Strasburg. Under the same year John Mentelin is mentioned as a printer at Strasburg.

III.48 Fischer, Typograph. Seltenheit. S. 44, 1ste. Lieferung. In this instrument Gutemberg describes himself as “Henne Genssfleisch von Sulgeloch, genennt Gudinberg.”

III.49 Primaria quædam Document. pp. 26-34.

III.50 “. . . . per henricum bechtermuncze pie memorie in altavilla est inchoatum. et demū sub anno dñi M.CCCCLXII. ipō die Leonardi confessoris qui fuit quarta die mensis novembris p. nycolaum bechtermūcze fratrem dicti Henrici et Wygandū Spyess de orthenberg ē consummatū.” There is a copy of this edition in the Royal Library at Paris.

III.51 Typographisch. Seltenheit. S. 101, 5te. Lieferung.

III.52 The two following works, without date or printer’s name, are printed with the same types as the Catholicon, and it is doubtful whether they were printed by Gutemberg, or by other persons with his types.

1. Matthei de Cracovia tractatus, seu dialogus racionis et consciencie de sumpcione pabuli salutiferi corporis domini nostri ihesu christi. 4to. foliis 22.

2. Thome de Aquino summa de articulis fidei et ecclesie sacramentis. 4to. foliis 13.

A declaration of Thierry von Isenburg, archbishop of Mayence, offering to resign in favour of his opponent, Adolphus of Nassau, printed in German and Latin in 1462, is ascribed to Gutemberg: it is of quarto size and consists of four leaves.—Oberlin, Annales de la Vie de Gutenberg.

III.53 St. Matthias’s Day is on 24th February.

III.54 In the instrument dated 1434, wherein Gutemberg agrees to release the town-clerk of Mentz, whom he had arrested, mention is made of a relation of his, Ort Gelthus, living at Oppenheim. Schœpflin, mistaking the word, has printed in his Documenta, p. 4, “Artgeld huss,” which he translates “Artgeld domo,” the house of Artgeld.

III.55 Serarii Historia Mogunt. lib. 1. cap. xxxvii. p. 159. Heineken, Nachrichten von Kunstlern und Kunst-Sachen, 2te. Theil, S. 299.

III.56 In the colophon to “Trithemii Breviarium historiarum de origine Regum et Gentis Francorum,” printed at Mentz in 1515 by John Scheffer, son of Peter Scheffer and Christina, the daughter of Faust, it is stated that the art of printing was perfected in 1452, through the labour and ingenious contrivances of Peter Scheffer of Gernsheim, and that Faust gave him his daughter Christina in marriage as a reward.

III.57 On the Pleasures and Advantages of Science, p. 160. Edit. 1831.

III.58 Ludovico Guicciardini, Descrittione di tutti i Paesi Bassi: folio, Anversa, 1581. The original passage is given by Meerman. The original words altre memorie—translated in the above extract “other memoirs”—are rendered by Mr. Ottley “other records.” This may pass; but it scarcely can be believed that Guicciardini consulted or personally knew of the existence of any such records. Mr. Ottley also, to match his “records,” refers to the relations of Coornhert, Zuyren, Guicciardini, and Junius as “documents.”

III.59 Junius was a physician, and unquestionably a learned man. He is the author of a nomenclator in Latin, Greek, Dutch, and French. An edition, with the English synonyms, by John Higins and Abraham Fleming, was printed at London in 1585. The following passage concerning Junius occurs in Southey’s Biographical Sketch of the Earl of Surrey in the “Select Works of the British Poets from Chaucer to Jonson:” “Surrey is next found distinguishing himself at the siege of Landrecy. At that siege Bonner, who was afterwards so eminently infamous, invited Hadrian Junius to England. When that distinguished scholar arrived, Bonner wanted either the means, or more probably the heart, to assist him; but Surrey took him into his family in the capacity of physician, and gave him a pension of fifty angels.”

III.60 Koning’s Dissertation on the Invention of Printing, which was crowned by the Society of Sciences of Harlem, was first printed at Harlem in the Dutch language in 1816. It was afterwards abridged and translated into French with the approbation, and under the revision, of the author. In 1817 he published a first supplement; and a second appeared in 1820.

III.61 Reckoning from 1568, the period referred to would be 1440.

III.62 “Ædituus Custosve.” The word “Koster” in modern Dutch is synonymous with the English “Sexton.”

III.63 “Sive is (ut fert suspicio) Faustus fuerit ominoso cognomine, hero suo infidus et infaustus.” The author here indulges in an ominous pun. The Latinised name “Faustus,” signifies lucky; the word “infaustus,” unlucky. The German name Füst may be literally translated “Fist.” A clenched hand is the crest of the family of Faust.

III.64 This is an admirable instance of candour. A charge is insinuated, and presumed to be a fact, and yet the writer kindly forbears to bring forward proof, that he may not disturb the dead. History has long since given the lie to the insinuation of the thief having been Faust.

III.65 Hadriani Junii Batavia, p. 253, et sequent. Edit. Ludg. Batavor. 1588.

III.66 Scriverius—whose book was printed in 1628—thinking that there might be some objection raised to the letters of beech-bark, thus, according to his own fancy, amends the account of Cornelius as given by Junius: “Coster walking in the wood picked up a small bough of a beech, or rather of an oak-tree blown off by the wind; and after amusing himself with cutting some letters on it, wrapped it up in paper, and afterwards laid himself down to sleep. When he awoke, he perceived that the paper, by a shower of rain or some accident having got moist, had received an impression from these letters; which induced him to pursue the accidental discovery.” This is more imaginative than the account of Cornelius, but scarcely more probable.

III.67 “Choragium omne typorum involat, instrumentorum herilium ei artificio comparatorum suppelectilem convasat, deinde cum fure domo se proripit.”—H. Junii Batavia, p. 255.

III.68 “. . . . . quum nova merx, nunquam antea visa, emptores undique exciret cum huberrimo questu.”—Junii Batavia.

III.69 In “Einige Bemerkungen über das Unternehmen der gelehrten Gesellschaft zu Harlem,” &c. S. 31.

III.70 Santander has published a French translation of this letter in his Dictionnaire Bibliographique, tom. i. pp. 14-18.

III.71 Wimpheling, who was born at Sletstadt in 1451, thus addresses the inventor of printing,—whose name, Gænsfleisch, he Latinises “Ansicarus,”—in an epigram printed at the end of “Memoriæ Marsilii ab Inghen,” 4to. 1499.

“Felix Ansicare, per te Germania felix

Omnibus in terris præmia laudis habet.

Urbe Moguntina, divino fulte Joannes

Ingenio, primus imprimis ære notas.

Multum Relligio, multum tibi Græca sophia,

Et multum debet lingua Latina.”

In his “Epitome Rerum Germanicarum,” 1502, he says that the art of printing was discovered at Strasburg in 1440 by a native of that city, who afterwards removing to Mentz there perfected the art. In his “Episcoporum Argentinensium Catalogus,” 1508, he says that printing was invented by a native of Strasburg, and that when the inventor had joined some other persons engaged on the same invention at Mentz, the art was there perfected by one John Gænsfleisch, who was blind through age, in the house called Gutemberg, in which, in 1508, the College of Justice held its sittings. Wimpheling does not seem to have known that Gænsfleisch was also called Gutemberg, and that his first attempts at printing were made in Strasburg.

III.72 Nachrichten von Kunstlern und Kunst-Sachen, 1te. Theil, S. 286-293.

III.73 In a Memoir on the Invention of Printing, which was crowned by the Academy of Sciences at Harlem in 1816.

III.74 Einige Bemerkungen, &c. S. 18, 19.

III.75 Enschedius published a fac-simile himself, with the following title: “Afbeelding van ’t A. B. C. ’t Pater Noster, Ave Maria, ’t Credo, en Ave Salus Mundi, door Laurens Janszoon, te Haarlem, ten behoeven van zyne dochters Kinderen, met beweegbaare Letteren gedrukt, en teffens aangeweesen de groote der Stukjes pergament, zekerlyk ’t oudste overblyfsel der eerste Boekdrukkery, ’t welk als zulk een eersteling der Konst bewaard word en berust in de Boekery van Joannes Enschedé, Lettergieter en Boekdrukker te Haarlem, 1768.—A. J. Polak sculps. ex originali.

Errors in Chapter III

(Displaying thus his meikle skill,)
closing parenthesis missing

for in no country are books to be found printed
foe in

[III.19]
footnote tag missing: best guess

einen spätern tag
spatern

was printed by Ketelar and Leempt
spelling unchanged

Footnote III.2

“J’ai rendu Vedelare
rendn

Introduction (separate file)
List of Illustrations (separate file)

Chapter I
Chapter II
Chapter III
Chapter IV (separate file)
Chapter V (separate file)
Chapter VI (separate file)
Chapter VII (separate file)
Chapter VIII (separate file)
Chapter IX (separate file)

Index (separate file)