21 March, 2008

Big brass matrices again: the Enschedé ‘Chalcographia’ type

In the post of 19 March 2007 it was suggested that existing brass matrices for big types could not have been struck with steel punches in so hard a hard metal as brass, but were probably reproductions cast in brass of strikes that had been made in lead with steel punches. Examples that were cited were the ‘large capitals’ of Garamond at the Museum Plantin-Moretus, titling capitals among the ‘Fell types’ at the University Press, Oxford, a titling from the French foundry of Claude Mozet that was acquired by Benjamin Franklin, and the series of two-line capitals of the romain du roi at the Imprimerie nationale in France. The brass matrices for the Garamond titling are accompanied by strikes in lead. In the case of the titling letters of the romain du roi there is a claim from the punchcutter that strikes were made with the steel punches in lead, and that these were used as patterns to cast replicas in brass.

A passing reference was made to a titling type in the museum of Joh. Enschedé en Zonen in Haarlem of which the height of the face measures 16 mm and for which there are ‘matrices’ in brass and lead, and relief ‘punches’, also of brass. Johannes Enschedé acquired them with other materials from the foundry of Jan Roman in 1767, and showed some characters under the heading ‘Chalcographia’ in his specimen of 1768, of which a facsimile was published in 1993. In a note below them Enschedé says, ‘The punches of these types are cut in brass, and struck and cast in leaden matrices, following the practice of the first typefounders.’ He does not refer to the matrices in brass, but there are reasons, given below, for thinking that the types that he shows were not cast from the strikes in lead.

The type can be dated to the middle of the 16th century, when its use in Lyon is documented. It was later used in Frankfurt am Main, and it is shown (with some altered and additional characters) in a specimen of titling capitals from the typefoundry of Johann Erasmus Luther dated 1665. For further details see the introduction and notes by John Lane that accompany the facsimile of 1993. The conclusion of Harry Carter, endorsed by Lane, was that the alphabet can be attributed to Jacques Sabon, originally of Lyon, the former owner of the Luther foundry, who had a German privilege for his method of casting large letters.

In his history of typefounding in the Low Countries published in 1908, Charles Enschedé printed an alphabet cast from the brass matrices. The face of these types was defective, like that of the types shown in 1768, which, despite the note by Johannes Enschedé, were pretty certainly also cast from the same brass matrices. Charles Enschedé added that he had not dared to try to cast from the thin strikes in lead. These were in excellent condition, as they still are, with an unblemished face. He had them copied by electrotyping, which produced good matrices, and he also showed an alphabet cast from these.

The strikes in lead are only 4 mm thick, and the depth of strike of 2 to 3 mm reduces the thickness of the metal still further at the face of the letter: if an attempt had been made to use them as matrices to cast type they would have suffered damage, since the overall thickness of the lead matrix is not much greater than that of the pattern. In the words of a note written by Stan Nelson, ‘the bottom of the lead matrix is very, very thin and impossible to use for casting type. There isn’t enough metal to absorb the heat of molten alloy being poured into the matrix.’ Given the good condition of the lead ‘matrices’, one must conclude that an attempt to cast type in them has never been made.

What then is the purpose of the little brass letters that accompany the matrices. They fit the lead strikes snugly. Are they – as Johannes and Charles Enschedé believed – the original ‘punches’ that were used to drive the impressions in the lead plates? Their thinness and flimsy construction, with completely open counters, makes them seem less than ideally suited to the purpose. Harry Carter thought that they were ‘castings reproducing punches of steel’ (A View of Early Typography, p. 15) . The lead plates were almost certainly made with some kind of punch, for there are impressions showing at the back of some of them.

This is what Charles Enschedé wrote in 1908 (as translated by Harry Carter for his English edition of 1978):

‘We have always shrunk from using the [lead] matrices. … I doubt whether any casts were made from them before their acquisition by Johannes Enschedé; the more so because we found brass reproductions with the originals. It may be that Enschedé or an earlier owner had found means to avoid the risk of damaging these precious relics by casting in them. The brass matrices appear to be castings from clay moulds. The moulding, however, is unskilful; [the illustration] is set in letters cast in the brass facsimiles, and it shows that these matrices are too poor to produce clean casts …’

If brass punches – the present letters or others related to them – were used to make the impressions in lead, the limitations of this method become clear. They could not be used afterwards to ‘clean up’ the cast replicas in brass, as was probably done with the steel punches that accompany the brass matrices at Antwerp, Oxford and Paris. So despite the fact that (notwithstanding the criticism of Charles Enschedé) the casting seems to have been done skilfully enough, it was inherently impossible to achieve a perfect reproduction of the face by this method alone, and the type cast from them suffered accordingly.

The notion that the original steel punches were used to clean up the face of matrices cast in brass has hitherto been guesswork – although in the last post I cited an 18th-century account which indicated that this was the practice when medals with a high relief were struck. Now a source has been found which confirms that this practice was also used by typefounders. In his extensive and well-informed discussion of early typefounding, all of which is well worth reading (Essai sur la typographie, Paris, 1851, col. 607, note 3), Ambroise Firmin-Didot has this passage (the English translation is mine):

‘Fournier le jeune is mistaken when he asserts in his work on the origin and progress of printing, page 20, that ‘matrices have never been cast: they are struck with a steel punch’. In order to assist the striking of very delicate punches, like the capitals of the large ‘ornamented gothic’ cut for our own typefoundry with such remarkable skill by Monsieur de Cornouailles, I had matrices cast in brass after matrices in lead that had been struck with the steel punches. After cleaning them out with care, to accommodate the effect of shrinkage in cooling, I drove the steel punches again into the cast brass matrices obtained in this way.’

If no steel punches were available to clean up the cast brass matrices in the Enschedé collection at Haarlem, this may explain the roughness of their face, and the signs they appear to show of a rather crude attempt to smooth it out. Their visible graininess suggests that, as in the case of the big matrices at the Imprimerie nationale, sand, rather than clay, was used for the moulding.

I am much indebted to the Stichting Museum Enschedé, Haarlem, for permission to make the images shown in this post, and to its curator Johan de Zoete and to Stan Nelson for their advice and help.