Benton's Vertical Pantograph for Patrices

(Before July 1884)

image link-topic-sf0.jpg

1. Benton's First Pantograph as a Patrix Engraver

It is almost certain that the first vertical pantograph by Linn Boyd Benton was designed and used for engraving patrices for use in electroforming matrices, not for engraving punches (or matrices). However, no evidence for its manner of use contemporary with that use survives. We can be comfortable now with the idea that it was a patrix (not punch) engraving machine because of the analysis done by William Charles Gregan, "the last master engraver at the American Type Founders Company," who spoke with Morris Fuller Benton about Linn Boyd Benton's pantograph engraving machines and "deduced" that they were patrix engraving machines. M. F. Benton would have been about twelve years old at the time the first pantograph was made, but he was both technically competent and deeply involved with his father's work throughout his life and should be considered a reliable source. This information was related by Gregan to Theo Rehak, who in 2006 passed it on to Patricia Cost, who published it in her book, The Bentons (p. 60).

Later in the same chapter, Cost quotes a more extensive e-mail from Rehak discussing Benton's accomplishments. In it, he discusses Gregan's analysis. Rehak's own perspective on Benton and on Gregan's analysis is especially valuable because as a user of Benton's engraving machines his is attuned to the "protocols" of their actual operation; these would differ depending on the material (typemetal patrices vs. steel punches). Rehak says: (1) "Gregan maintained that the first use of the early pantograph prototypes that LBB 'played' with were used to cut letter-punch models for elecctroplating." and (2) "There is also, of course, the reality of exactly how LBB changed his engraving protocols from lead to steel, thus producing the punches which saved Mergenthaler and his machine. So Gregan thought that Morris Benton indicated that the type models for electrotyping came first. ... These changes in engraving protocols likely occurred in rapid succession, thereby preventing any 'paper trail' with enough time to be documented." In other words, Rehak is of the opinion that Gregan's analysis is correct but that we may never know exactly when the transition from cutting in typemetal to cutting in steel occurred.

(As an aside, Rehak also mentions the use of electroforming to copy (often pirate) existing types, and says that "Benton conceived of creating even more accurately-cut, newly-designed models by engraving them in relief on the end of a blank piece of type." (p. 73) This goes a bit too far. The method of engraving patrices by hand for use in electroforming matrices, while never well documented, was common in the Nineteenth Century. Indeed, I suspect that it was responsible for most of the "fancy" display types of the period. Everyone was doing it; Benton didn't invent it.)

Gregan's interpretation of M. F. Benton's recollections as passed down by Rehak to Cost may seem like a thin thread of analysis upon which to base a conclusion, but it is an analysis done by people who actually knew how to use the machines - that counts for a lot. The other references we have to the initial use of this machine for engraving patrices are less certain.

Dr. James Eckman wrote of it in his chapter "The Gifted Bentons, Father and Son" in The Heritage of the Printer (1965) : "The machine which brought Benton fame was his punch-cutting machine, patented in 1885, which, oddly enough, Benton did not use for the cutting of punches. He used it to cut type metal, for the Benton & Waldo foundry at this time was making matrices by the electrotype process." (p. 111) Eckman's account, though, continues by noting that the Mergenthaler company induced Benton to do a trial cutting of steel punches and then leased the machine. It sounds suspiciously like his source is Bullen's 1922 article on Linn Boyd Benton in The Inland Printer . Eckman's book is quite good, and very carefully done, but it would be unwise to consider it an independent source.

Henry Lewis Bullen, in his 1922 biographical article "Linn Boyd Benton - The Man and His Work" in The Inland Printer , does say explicitly that at first the Benton "punch-cutter ... had never cut in steel. It was cutting in type metal, for Benton was using electrotyped matrices." This sounds conclusive, and Bullen was a major figure in the typefounding industry working for the same firm (ATF) as Benton and writing during Benton's lifetime. Yet as I argue elsewhere, Bullen's account of the influence of the Linotype company on the transformation of the Benton engraver from a patrix to a punch engraver cannot be correct. This casts doubt upon his reliability as a source for information on the initial use of the machine. I believe that in this case he is correct that the first use was to cut patrices in typemetal, but he is not in general a trustworthy source.

Other secondary sources have been even less clear. Cost (p. 60) does an excellent job in analysing the errors that Richard Huss made in two of his otherwise remarkable books: The Printer's Composition Matrix (1985) and The Development of Printers' Mechanical Typesetting Methods: 1822-1925 (1973) .

While the identification of Benton's initial use of this machine as a patrix engraver hangs by the slenderest of threads (Gregan's analysis), it makes perfect sense. The method of engraving patrices (by hand) for electrotyping matrices was quite common in the Nineteenth Century (much more so than was admitted then or has been acknowledged subsequently). Moreover, the machine continued in use as a patrix engraver even after it became established as a punch engraver. Cost cites (p. 68) an "undated pamphlet from the Benton-Waldo Type Foundry, probably produced in 1891" which lists several customers leasing the machine. These include the "Electro Matrix Co." of Minneapolis and (another) "Electro Matrix Co." of New York.

2. The Date of Benton's First Pantograph

However, to the best of my knowledge no information survives on what would seem to be a most important detail: the date at which Benton developed this machine. The earliest dated reference to it that I am aware of is from July 1884, by which time it was already being used to cut punches in steel.

We know that Benton patented a machine for casting leads in 1882 (US patent 254,792 of 1882-03-14; it looks very much like a clever but unremarkable adaptation of a conventional stereotype casting box). He was involved with developing his so-called "self-spacing" types in 1883 (US patent 290,201, "Printing Type," filed 1883-05-08, issued 1883-12-18). He did not file any patent for a pantograph until February 29, 1884 (the filing date for US 332,990). We know from a report in The Inland Printer, Vol. 1, No. 10 (July, 1884): 21; cited by {Cost, p. 61}. that by that date he was cutting punches in steel. (At the risk of introducing a red herring, I'll note that patent 332,900 as issued mentions only "Punch-Cutting," not patrix engraving. However, this patent had a difficult path; I do not know what the original application of February 1884 said. So June 1884 is a terminus ante quem for the engraving of punches in steel.)

This does not, however, establish the point at which Benton developed his first vertical pantograph and used it in the engraving of patrices. It would seem likely that it was circa 1883 during the development of his "self-spacing" type, but probably not much earlier. This is of course an unsubstantiated supposition on my part, since I know of no evidence.

There is an interesting question of precedence here, since in 1882 the Central Type Foundry was using a pantograph matrix engraver which had been imported from Germany in 1880 to produce matrices for commercially available type. (See also the Notebook on type-making Beyond (and Before) Benton.)

Although to the best of my present knowledge nothing is known about the technical details of the first version of Benton's machine, I would doubt that it differed substantially from the punch- and patrix-engraving pantograph described in his 1885 patent.

3. Bullen and the Reason for the Pantograph

Henry Lewis Bullen's biographical article on "Linn Boyd Benton - The Man and His Work" ( Inland Printer, Vol. 70, No. 1 (Oct. 1922): frontis, 60-64) is a primary source for information on the development of Benton's pantographs. Unfortunately, it seems to be wrong on at least two of the most important points in this development: the initial reason for Benton's need for a pantograph, and the influence of the Linotype on the development of the pantograph into a punch-cutting machine. This present Notebook will discuss the first of these problems. For the second, see Benton's Vertical Pantograph for Patrices and Punches.

Bullen writes:

"In 1882 Benton's thoughts turned toward the invention of a typesetting machine, in which a near approach to self-justification was to be effected by casting all the characters and spaces and quads of a complete body type font of roman and italic on nine widths, instead of on the more than one hundred widths found in the average font of body types. ... [I omit here a comparison of this with the Lanston Monotype "unit" system and a discussion of the origin of the slightly inaccurate name "self-spacing" for this system] ... Benton was granted a strong pioneer patent for this system of making types. This unexpected merit of unit-width types, first invented by Benton, caused him to defer the creation of the typesetting machine for which they had been made to be used. In any case, fonts of various bodies, old style and modern, had to be made before the machine could be utilized. Punches for every character were required to be engraved and matrices made. There were more than three thousand punches to be cut and not one punch-cutter was available either in America or in Europe. This dilemma was the turning point of Benton's career - it eventually disclosed to himself that he had mechanical genius of the highest order.

"Benton determined to make a machine to cut punches. ...

"Benton had to design thousands of characters to fit his width units. The punch-cutting machine had to have a pattern for each character. Benton ... had to design each letter on a large scale and cut metal patterns to the same scale. ...

Two features of this account merit comment.

First, it attributes to Benton an unfinished project for building a typesetting machine. Now, it may well be that Benton did contemplate just that. But there is no record anywhere else in the literature of it. The only other reference of which I am aware is a single line in Cost's book (p. 47), and this is in a context where she is frequently citing this very article by Bullen.

Bullen implies that it was this typesetting machine which prompted Benton to develop his "self-spacing" type (that is, type cast on a very limited number of set widths, each of which was a simple division of a unit) to permit easy justification. Again, this may be so. At least one previously proposed typesetting machine relied upon similarly rationalized types, the 1854 "Self-Spacing Composer" of Martin Wiberg (GB patent 1548 of 1854; see Huss, p. 46). However, "self-spacing" type is by no means a necessity for a typesetting machine. None of the machines which did achieve some commercial success employed such type. (For example, the Hattersley, which was used commercially from 1857 to 1902, did not ( Huss, pp. 50-51). Neither did the quite successful Thorne (later called the Unitype or Simplex), which originated in 1880 and for a time competed successfully with the Linotype. ( Huss, pp. 97-99.))

Bullen portrays the advantages of this "self-spacing" type for hand composition as an "unexpected" by-product of its development for this typesetting machine. Again, this may be true, but we have no independent confirmation of this.

But even if one assumes that Benton's "self-spacing" type did come into being as a part of his typesetting machine project, Bullen's account is misleading in its assertion that this was a unique set of circumstances of such overwhelming difficulty that it necessitated the invention of a pantograph engraving machine.

A font of "self-spacing" type is no different from any other font of type, save only that it is cast to a limited number of set widths. You could take any set of matrices, come up with some scheme of a limited number of set widths, and cast a font of type to that scheme. It probably wouldn't look very good, though, because the width of each sort's face would only approximately match the width of the body it was cast on. In order to make a font of "self-spacing" type that is at all presentable, you would have to design the typeface so that the characters "worked" visually in a fixed number of widths.

So if you want to make a reasonably decent looking self-spacing font of type, yes, you do have to design a new typeface for it and make matrices for it. But this is no different than the design of any new typeface. The number of matrices that a typefounder needs for casting a font of "self-spacing" type is exactly the same as the number required for casting any font of type: one matrix per sort.

Let's say that each set of matrices for a particular type in a particular size contains 72 mats (which is a fairly common number of mats for casting display type with uppercase, lowercase, figures, and punctuation). Say you want six body sizes. Now you're up to 432 matrices for just one variation (ordinary roman, for example). If you want italic, add another 432 mats. If you want roman condensed, add another 432. Indeed the number of mats required rises very quickly. If Benton wished to re-equip his foundry to offer a wide range of "self-spacing" types, Bullen's estimate of "more than three thousand" punches (he probably meant patrices) to make the matrices is probably an underestimate.

But - and this is the important point - this situation is no different than that of any new typefoundry at any time during the history of metal type. If you wanted to set up a new typefoundry at any point before the 1840s, you had no other option; you were going to cut a lot of punches. Some of the best known names in classical typefounding (Bodoni, Caslon, Fournier) cut a prodigious number of punches. After the introduction of electroforming in 1840 you had two more options. The easy way out was simply to pirate existing type by electroforming new matrices out of it - many typefoundries, both new and established, did just that. You could also make new type legitimately by engraving patrices for electroforming rather than cutting punches. Working in typemetal originals is easier than working in steel, and an unknown (but very large) percentage of 19th century type was made in this way.

In other words, Benton's position was no different than that of any other typefoundry seeking to introduce a new product. His solution might easily have been the same. Punchcutting and patrix engraving were established crafts - skilled crafts to be sure, but not rare ones.

Bullen's statement that "There were more than three thousand punches to be cut and not one punch-cutter was available either in America or in Europe" is curious. If by "punch-cutter" he meant punch-cutting machine, then his statement is false but he may perhaps be excused the error. He is overlooking the Central Type Foundry pantograph matrix engraver , which had been in commercial operation since 1882. It isn't yet clear, however, how well-known this fact was when Bullen was writing in 1922. But if by "punch-cutter" he meant "punch-cutter" (a person who cuts punches and by extension patrices), then his error is not excusable. Any typefounder would have known that punch-cutting was an established profession in the late 19th century. Indeed, at the turn of the 20th century William Loy published in The Inland Printer an extensive series of articles over a three year period with biographical sketches of 27 contemporary punch and patrix engravers working for American typefoundries.

It is entirely possible that everything else Bullen says in this passage is literally true. Benton may have been working on a typesetting machine that has otherwise vanished from the historical record, and its particular design may have required unit-system types. The utility of these types in hand composition may have occurred to him only after the first font was cast for the (unfinished) machine. This may have been what prompted him to reposition his foundry as the purveyor of his (patented) "self-spacing" types. We have no independent confirmation of any of this, but, still, it all may be true. Yet it remains a misleading story because, as Bullen presents it, he implies that in this Benton faced a crisis of overwhelming magnitude which could only be solved by his genius - and a pantograph. Benton was a genius, true. But the "crisis" he faced here was no different than that of any non-piratical typefounder seeking to give his foundry's offerings a new dress. Nothing in this situation forced upon Benton the invention of a pantographic engraving machine as the only solution. Typefounders worldwide had been dealing with the same situation by hand, successfully, since 1450.

Benton developed his pantograph engraver because he wanted to, not because he had to.

4. Bibliography

Bullen, Henry Lewis. "Linn Boyd Benton - The Man and His Work" Inland Printer, Vol. 70, No. 1 (Oct. 1922): frontis, 60-64

Cost, Patricia. The Bentons: How an American Father and Son Changed the Printing Industry Rochester, NY: RIT Cary Graphic Arts Press, 2011.

Eckman, James. The Heritage of the Printer. Volume 1. Philadelphia, PA: North American Publishing Company, 1965.

Huss, Richard. The Development of Printers' Mechanical Typesetting Methods: 1822-1925 (Charlottesville, VA: By the University of Virginia Press for the Bibliographical Society of the University of Virginia, 1973.)


Select Resolution: 0 [other resolutions temporarily disabled due to lack of disk space]