a long, passionate kis for you...

Part 2: Some ontogeny and phylogeny

Optical scaling in digital type design – that is, the production of distinct masters at each of several different point sizes – is a subtle and rare thing. When applied, the various point masters are often near clones of one another, spanning a narrow continuum of width and weight. And notwithstanding the pioneering work of Knuth [9], and later of Hoefler [10], Slimbach [11], and Stone et al. [12], as well as a few others (e.g., Burke, Kobayashi) who appear to approach optical scaling as simply part and parcel of constructing a quality typeface, most designers of digital type have adopted a “one size fits all” modus operandi. But this was certainly not the case in the days of metal type. Of pre-digital optical scaling – and with particular regard to the Stempel Janson types – Jack Stauffacher wrote:

When these early letters were hand cut, the punch-makers made their different sizes according to a natural optical scale. Their very limited tools created a style that expressed an intuitive sense for good proportion. The artists clearly retain this kind of spirit in contrast to an incongruous letter that is contrived, having no dignity or beauty. You will notice that each size (font) bears a slight variation in mood for both the romans and italics. These varieties in the fonts are the unique signature of a hand punch-makers skill. This is not found in the modern pantographic methods of adapting the letter design into type. The modern method creates its own kind of simplicity in the exactness for each size [ref. 7 below, p6].

It may well be folly, then, to attempt to compare the Stempel Janson types to their direct descendants in metal, for although each of the Stempel sizes is a familial variant, each is just as much a different cut – different, to underscore Stauffacher, in both point size and mood. And even within the same point size, there are variants; in the 12-point size, for example, there are two distinctly different cuts, labelled 12a and 12b [i, j], the latter being the cut in which Stauffacher’s monograph is set. An examination of the roman reveals that the most apparent differences between these variants may lie in the absolute size and in the x-height, the ‘a’ cut being somewhat smaller and lower than the ‘b’ cut. The lower case also reveals marked differences in the letters f and a. In the 12a cut, the shoulder of the letter f is generous and round, and the terminal extends far to the right of the bar. The shoulder of the f in the 12b cut is much more modest and angular, and the terminal, which is somewhat bulbous, lies closer in. The lower case a in the 12a variant looks nearly out of place; it appears to me to be more Van Dijck than Kis. Not so, however, with the lower case a in the 12b specimen; today’s digital versions of Kis – predominantly, Linotype Janson Text, Monotype Janson, and Monotype Ehrhardt – have retained its more angular shoulder and low-sitting bowl.

Stempel Janson roman in the 14-point size [k] has the look of a stylistic interpolation between the flow of the 12a cut and the stout angularity of the 12b cut. With a few exceptions, perhaps most notably in the tail of the upper case Q, it has all the fatherly markings of Monotype Janson in metal [l]. Its italic progeny, however, are not like Monotype Janson at all in appearance; the overall angularity of the Stempel italic was retained instead in the text sizes of the Linotype issues [m]. Most particularly, the squarish up-over-down shoulders of the lower case m and n in the Stempel 14-point italic [n] are mirrored in its Linotype digital descendant; these constitute a nearly unique hallmark of Linotype Janson Text italic. The relative smoothness of the Monotype italic appears to owe much more, rather, to the Stempel 12a and 12b italics [o, p].

But now I realize that I am getting a bit ahead of myself, for although I’ve mentioned and have shown the Linotype and Monotype descendants of the Stempel types as points of comparison, I’ve left out all specifics; here again, Lawson fills in much of the necessary detail [ref. 2 below, p168]:

Linotype Janson was first produced in the United States in 1937 under the guidance of C.H. Griffith, the typographic director of Mergenthaler; he used the Stempel original as the model for this cutting. In the same year, Sol Hess undertook for the American Monotype firm a modification of the face, using as a pattern a version in a seventeenth-century book. Both of these revivals found immediate favor with the American printers, even though the types lack some of the crispness of the Stempel foundry version, a factor that is most noticeable in the display sizes. [See, for example, the Groot Canon Romein and Clein Canon Romein in this specimen [q], prepared by Stauffacher and used by Heiderhoff [13]].

In the Linotype lineage, then, we proceed from Kis to Griffith, and then from Griffith to Zapf, who, in 1951, retooled the font’s 6, 8, 9, and 10-point cuts (and still later, as shall be mentioned, from Zapf to Frutiger). It should be noted that, retooling and refinement aside, the faithfulness of the Linotype interpretation was necessarily compromised by the Linotype machine itself. Robert Bringhurst puts it succinctly [ref. 1 below, p137]:

Typeface design for the Linotype was restricted by three basic factors. First, kerning is impossible without special compound matrices. (The basic italic f in a Linotype font therefore always has a stunted head and tail.) Second, the em is divided into only 18 units, which discourages subtlety of proportion. Third, the italic and roman matrices are usually paired. In most faces, each italic letter must therefore have the same width as its counterpart in roman.

Reexamine the Linotype sample [m] from above; it is indeed the lower case italic f that stands dramatically apart from the rest of the lower case italics. You will see the result of the handicapped kern in the lower case roman ff ligature as well, in the form of a truncated terminal in the second f (spot, as an example, the word “indifferent” in this [r] sample).

If you pick up, at random, an American book printed in the mid-twentieth century, you may well find that it is printed in Linotype Janson; for a couple of decades at least, this type appeared to be used with nearly the frequency of Caledonia. As it is today in its digital incarnation, Monotype Janson in metal was used far less frequently than its Linotype counterpart. With regard to the genesis and subsequent history of the Monotype cut, I unfortunately know relatively little. Nevertheless, something of an aside to the very early history of Monotype Janson is recounted, in part, in Bruce Rogers’s Paragraphs on Printing [14]:

In planning the Limited Editions Club Shakespeare [s, t] the first consideration was of the type, which needed to be bold and vigorous enough to convey to the reader’s eye something of the rugged Elizabethan quality of the text. A large format was necessary to allow for the illustrations and therefore a correspondingly large type was indicated. The first experiments were made at the Oxford University Press with great-primer Fell types; but when for various reasons it became necessary to print the volumes on this side of the Atlantic, the use of Fell types had to be abandoned and search made for something else comparable in effect.

For so extensive an undertaking hand-setting in this country was out of the question. After experiments with several of the type faces made by the machine companies it was felt that none of them was as suitable as the reproduction of a type cut by a Hollander, Anton Janson, between 1660 and 1687 – less than a hundred years after Shakespeare’s time.

The Lanston Monotype Machine Company undertook to cut the 18-point size in close facsimile of the original, preserving all the slight irregularities of design and alignment which help to give it life and vigor. The 18-point italic of the original Janson appears distinctly inferior to the other sizes; perhaps an odd fount that has been introduced into the series.

James Hendrickson, the interviewer of Rogers and the editor of Paragraphs on Printing went on to add, “The brief description of this partial re-cutting of the Janson type sounds quite simple here, but it was anything but that. To accomplish it B.R. had to conduct a detailed and almost daily correspondence with Mr. Sol. Hess, Art Director for the Monotype Company, for some weeks, a reading of which would be an eye-opener for anyone who may fancy that fine printing is principally a pleasant pastime.”

Now up to this point, I have focused principally on the Linotype and American Monotype interpretations of Kis; what of that of the British Monotype firm? Well, in the mid-to-late 1930’s, Stanley Morison was, as ever, doing double duty as both historian and businessman, at once attempting to determine the origin of the Kis types and directing the production of a Kis of his own. Nearly concomitant, then, with the Linotype and American Monotype releases of Kis revivals, Morison issued a Kis of a slightly different kind, not so closely based on the Stempel types but rather more similar in spirit to the original Groot Canon Romein type shown earlier [q], yet retaining Morison’s indelible imprint – most notably, perhaps, the color and compactness he considered necessary for good book types. And so Monotype Ehrhardt [u, v] would become a sort of bastard child whose parentage appears to owe as much to Morison as it does to Kis. Particularly illuminating in this regard is Morison’s early direction of the Ehrhardt face. From Barker [ref 4 below, pp 345–346]:

...Morison took up the idea of recutting [the Janson], and – uniquely – his instructions for it survive. On 26 January 1937, armed with photographs of Mori’s original Ehrhardt specimens, he wrote to Mr Steltzer in the Type Drawing Office.

Last Spring we put you in hand with you the initial work of re-creating the old Dutch Face for which more than one of our customers were asking.

Through paucity of material however, we were unable to satisfy them and the several characters in Series 453 were abandoned.

We are now sending you:–

A. A photograph of the Ehrhardt Series, which amongst other sizes shows a portion in Tertia, both Roman and Italic.

B. Five photographic prints showing the same fount with a small amount of Italic.

We now ask you to put in hand the work of a few Roman and italic characters.

The ultimate completion of the Roman alphabet would not seem to be a matter of much difficulty but in the matter of the Italic, some obstacles may be encountered as in the capitals several are missing.

We think, however, that for the present the accompanying photographs will supply you with sufficient data for the making of the specimen characters.

We should like the Face to be as big on the body as the 14pt. in Series 101 [Imprint].

We imagine that the weight will be between Series 101 and Series 110 [Plantin].

Finally, we shall be glad if you will regularise the fount as regards alignment.

For the time being, the Face should be called ‘Ehrhardt’.

By 20 February, he already had proofs.

I am very interested in the proof of the Ehrhardt series 453 14pt. roman and italic, but I am prevented from making up my mind definitely in favour of these specimen characters owing to the fact that the set appears to me to be so wide. I should like to see a specimen on a narrower set if you could have this made.

At first glance the fount appears to me to be a little more condensed than I expected. Also I do not recognize the capitals. Having no copy of the photographs of the Ehrhardt specimen, I do not dogmatise, by my distinct impression is that the capital G. though correct in colour, proportions and general form, is not correctly seriffed at the top or at the bottom. There is a serif, I think, at the bottom of the capital G. where the thin curve joins the thick perpendicular. I should have said, too, that the horizontal at the foot is a trifle too short. I am speaking now only from memory, but I feel sufficiently confident to ask you to look into it.

The lower case roman letters appear to me to be sound.

With the italic I am not quite so pleased. I should have imagined, for example (and here again I am now only speaking from memory) that the italic lower-case g. would have had much more sweep to it.

After that, all went smoothly; the design was approved in August, and matrices were made available in November 1937. There was a full showing in the Summer 1938 Monotype Recorder, and Morison himself set out the history of the type in the March 1939 issue of Signature...

More than thirty years after Ehrhardt’s release, Harry Carter described the face in the Appendix to Morison’s A Tally of Types [15]:

The letters of Monotype Ehrhardt are like those of the Janson, but the appearance of a page set in it is different. The Janson is more rotund and has greater contrast of thick and thin. It exemplifies the qualities that Joseph Moxon so admired in the ‘late made Dutch letters’ when he wrote in 1683, ‘the commodious Fatness they have beyond other letters,...As also the true placing their Fats and their Leans, with the sweet driving them into one another.’ In ‘Ehrhardt’ these qualities have been to some extent sacrificed to economy of space: it belongs to a late phase in Morison’s thinking where he was less interested in the reproduction of an old type than in the production of one that gave good value in legibility.

We might conclude, then, that Morison was even more businessman at that point than historian. Indeed, he wrote in Signature shortly after Ehrhardt was produced [ref. 5 below, as reprinted in Carter’s article in A Tally of Types]:

But the justification for printing is not primarily stylistic; first and foremost its justification is economic. It follows, therefore that appreciation of the art of lettering as applied by punch-cutters to the service of the printing trade must reckon with the element of economy. Indeed it seems to the present writer not too much to say, regarding book typography in the period after Garamond and Granjon, that it is in the exploitation of the available space, rather than in conspicuous details such as serifs and stresses, there lies the secret of successful type design.

What was good for business was good for Ehrhardt. Carter concluded “It is a successful type-face. I think Kis would have liked this series. He would certainly have liked the smooth finish given to it by the industrial process. The grace of his alphabets is not lost by any means.” Smooth, yes; but somewhat sterile, too. Reconsider the sample [v] above. It is strange to see a pre-digital typeface that is optically scaled with the constancy of stroke and proportion that we are accustomed to seeing in our digital age. There is very little variation, among the seven point sizes, in what Stauffacher described as mood; and within each size, there is remarkable inertia with little hint of ostentation or eccentricity. A perfect crystalline goblet face, then – never quite in or out of fashion – but always deserving to be put to good use.

Today’s digital version of Monotype Ehrhardt is a largely faithful interpretation of its metal forebear. Can the same be said for Monotype Janson and Linotype Janson Text? How have the three main Kis revivals evolved; what characteristics do they share, and how do they differ? In what aspects do I feel they excel, and where are they lacking? In the last part of this series, I’ll compare and contrast the three main digital versions of Kis to one another. And over the rest of the summer, I plan to post a little on the journal Grand Street, on various pieces in the short-lived The Imprint, on some relatively recent books on design, and on semiotics and typography. Thank you for reading!

9. Knuth, Donald E. (1999) Digital Typography. Stanford: CSLI.

10. Hoefler, Jonathan (1992) HTF Didot. See Muse No. 1 and Hoefler’s article in Serif (5), pp 58–61.

11. Slimbach, Robert (1990–present) The Brioso, Cronos, Adobe Jenson, Kepler, Minion, Sanvito, Utopia, and Warnock type families. See the Adobe Type Library Reference Book.

12. Stone, Sumner et al. (1994) ITC Bodoni. See, inter alia, “You say Bodoni, I say Bodoni...”, by Dave Farey in the online journal, x-height.

13. Heiderhoff, Horst (1989) “The Rediscovery of a Type Designer: Miklós Kis” in Fine Print on Type. San Francisco: Bedford Arts.

14. Rogers, Bruce (1979) Paragraphs on Printing. New York: Dover.

15. Carter, Harry (1999) “Ehrhardt” in A Tally of Types, by Stanley Morison, pp 118, 120–121. Jaffrey, NH: Godine.

26-June 2004