You trained as a graphic designer, but later became a full-time type designer. That was not an obvious step at the time — type design being such a specific skill. How did one thing lead to another?

I began studying at a private design school in Mannheim when I was very young. Writing and lettering were part of the graphic design program. We wrote with the broad-nibbed pen and designed lettering for posters. All of this was very decorative and had nothing to do with text typography. I even had classes with the famous Herbert Post, who lectured there as a favor to the school’s director. But as the students lacked any typographic background or orientation, we weren’t very interesting material for Mr. Post. I suppose he wasn’t impressed by our level.

I went on to study graphic design at Nürnberg Art Academy. Typesetting and typography were taught there as well, but not very intensively. After graduation I worked at the Historical Museum in Frankfurt, where I designed exhibitions. I lived with friends in a theater workshop, and had a lot of free time on my hands. And out of the blue I decided to try to design a typeface. My first uppercase alphabet looked kind of nice — no more, no less. But I continued and created my first text typeface. I wasn’t happy with it, but I couldn’t put my finger on what was wrong with it. I didn’t have any criteria yet as to what a text face should look like. A text face is totally different from the scripts we wrote in art school: blackletters and uncials, artsy-fartsy serif faces that had nothing to do with the history of printing.

And so, in the early 1980s, I began an independent study of about five years. I looked into printing types in detail. How were they made? A Garamond, for example: Why are the italics so completely different from the uprights? I created several typefaces in a rather curious way. I painted alphabets on large boards. At first I drew them really big, but I quickly realized how impractical that was. So I switched to a letter size of about 8.5 cm (just over 3"). I had a very good 6×6 camera that I used to reproduce those letters, which I then cut out with scissors. On photo paper, I reduced the characters to about an inch in height. From that I made something that looked like the old Letraset sheets — I even used the same letter frequency: many ‘e’s and ‘n’s, but fewer ‘a’s, and so forth. I drew a baseline on each character for building lines of text. The reproduced lines were then composed into longer lines and blocks of dummy text to give me a good impression of the final font.

When I was finally happy with my first type designs, I put them into a portfolio and sent them to the H. Berthold AG typefoundry.

And it worked out? Just like that?

It did! Günter Gerhard Lange — the artistic director of H. Berthold AG — wrote back to me personally. To my surprise, they’d selected a typeface which I regarded as a kind of youthful folly, even then. I had put it into the portfolio merely as a filler. So although I thought there were much better designs in there, that was the one they wanted. It was released as part of the Berthold program under the name of Vergil. Anyway, I had a foot in the door, which was the main thing.

Some time later I officially applied for a vacant position at Berthold, and that worked out well, too.

What was it like to work with Günter Gerhard Lange?

My first job at H. Berthold AG went like this: Günter Gerhard Lange came in and showed me an oldstyle family he had designed in the 1950s. This was Arena, a typeface for hand-setting. Originally there was a Regular weight and a Bold, perhaps also a Heavy version. But these cuts were all very different in shape: the Regular was rather broad and round — a very classic typeface — and the bolder weights were really narrow. They did not fit together terribly well.

Since Arena was relatively successful, Berthold AG had the idea to do a complete revamp. The plan was to develop a family in which all members were perfectly balanced — from Light to Bold. I drew them, Lange looked at them and made corrections. This went back and forth a few times. I did not have to remain strictly faithful to the original model. The font family was then published as Neue Arena, or Arena New. It had actually become a very different typeface from the original Arena. This approach became our working method.

For some reason, Lange held me in high regard. At the beginning he said he wanted only three characters when starting on a new font — ‘n’, ‘a’ and ‘g’. But I preferred showing him half an alphabet, which he was not accustomed to. I would also present alternative shapes for each character. I think he liked the fact that he simply had to choose. In this respect, Lange’s career has been remarkable: his early typefaces were his own designs — in the 1950s he made several fonts in the style of the times, mostly calligraphic in character. At some point he stopped doing original designs and went on to produce only new editions of classic faces, or to revise — and usually improve — designs by others. In fact, Lange was not very interested in new designs. He simply wanted the fonts that were released to be very well made — in particular the “classics”. That was his main ambition: any fonts that were submitted were improved under his scrutiny.

Did you discuss the works-in-progress in person? Or did the drafts go back and forth through the mail?

By mail, most of the time, since Lange did not work in-house anymore, and only came by once a week. He never said much, but he went into detail in the corrections he made at his home studio. Character sets were printed on photographic paper and sent to him on A4 sheets, with the type printed in the center. Lange sent back the corrected sheets, which always looked terrific because he had such a clear handwriting. Around the text setting the whole page was filled with notes, with lots of arrows, pointers, and extra sheets attached — he was extremely detailed. I learned a lot from this — not so much about taste or style, but about functionality. That was his main concern — all the rest was vanity to him.

Lange wanted to receive the typefaces as photographic prints at 18pt type size — the entire character set. But those 18pt prints only served to make corrections in writing. The final evaluation was made on the basis of a reduction at 8pt. To him, that was the absolute criterion for a text font. He held the prints against the light, and sometimes studied them with a magnifying glass. “8pt is the moment of truth” — that was his motto. It makes sense, because at that size you can judge the proportions, see whether the character width is sufficient for a text font, etc.

Functionality is ultimately more important than some flamboyant shapes that look great at display sizes, but are superfluous or annoying at text sizes. Lange was really strict on that point. So on the one hand he gave me a lot of freedom, and on the other, he was relentless when it came to functionality.

How were fonts produced at the time? Was the process digital, or still analog?

When I worked in-house at Berthold’s type design studio between 1989-1993, the fonts were all digital. But in the design phase they still used technologies originally developed for photo-typesetting. My task was limited to making drawings and basic character spacing. I actually worked on paper or cardboard, with ink and opaque white. For each character, a working drawing was made at 12 cm [4.7"] cap height. I then handed in my drawings for digitization. I had nothing to do with the technical side of font production. Lange said to me: “You just make your drawings, don’t worry about the people over there.” My colleagues sat in front of their Tektronix monitors — which only displayed a dim outline — using Ikarus font development software. It was a very cumbersome affair in the beginning. Ikarus allowed you to create and edit outlines, but without a mouse. All input was done on the keyboard, you had to know every shortcut to edit the outline. In order to test the design, you made a reproduction of the drawing and reduced it to reading size in the darkroom. After that it was back to your ink and opaque white.

At some point this process became too tedious for me; I thought I would go crazy! So I simply began doing it myself, and learned to use Ikarus. Things went a lot faster. I made photographic prints of my own designs, or cut them with a plotter from Ulano film and made further reductions. Thank God I had the courage to take that step — it gave me a very good foundation to learn Ikarus M (the Mac version written by Petr van Blokland) and work on them at home later. From 1993 onward, I worked independently.

Did you continue your collaboration with Günter Gerhard Lange after your time at H. Berthold AG?

Yes, we began several new fonts, some of which were never quite completed. The idea was to develop designs to a basic level which would be finalized later. I only have prints of some of these fonts. After the Berthold library was taken over by Berthold Types Ltd. in Chicago, the first family we produced was Whittingham, an English typeface from the mid-19th century. Charles Whittingham was a printer; he probably did not produce this font himself, but bought it from William Thorowgood. I think it has an important place in the Berthold library — it is the Protestant counterpart of the Catholic Bodoni. It is a very austere typeface. You sense the improvement of production tools of the period — those sharp and fine outlines, engraved in steel.

In early 2011 you presented your first Hoftype fonts. To launch one’s own digital foundry and start selling fonts via MyFonts is a common move among young designers, but less so among people of your generation. What motivated you to start up Hoftype?

I’ve begun so many typefaces over the years. I accumulated a lot of ideas and worked on many of them, without ever completing any of it. I simply didn’t find the time. Also, I lacked a technical background in font production. So I started learning these things two-and-a-half years ago. I acquainted myself with kerning, OpenType and all those things. Then I got going.

After its debut on MyFonts, Hoftype was almost immediately successful. Were you in any way overcome by the sudden popularity of your fonts, or surprised that it worked out so well?

Not at all, it was what I expected, haha! I was even a bit disappointed at the start. Impara was the first one, and for months nobody bought it. The next few sold quite slowly as well. But I gradually realized that you have to establish a brand; people aren’t very interested in a single family from a new foundry. But after the first bit of success, the earlier fonts, too, started picking up. The whole marketing thing is very new to me. You have to experiment a little, see how others do it: they give away one or two fonts for free and still realize good sales, so you try something similar. I don’t think I’m losing much when thousands download the one free font in the family. For customers who are seriously interested it’s an advantage because they can try out a complete font, including the OpenType features. In any case, I’m happy with the result: about eighty percent of my sales are now complete families.

Have the new Hoftype fonts all been designed recently, or do you have a stack of old concepts that you gradually prepare for publication?

I don’t have a particular system. Several of my fonts are revisions of typefaces that are in fact quite old. Cassia, for instance, was started fifteen years ago but I did not publish it until 2011. It originally looked quite different: it began as a neoclassical text face with relatively fine hairlines and thin serifs. Then I realized it wasn’t very original, there are already a lot of those faces around. I wanted something that looked contemporary — powerful, sturdy and more linear. So I simply reinforced the hairlines and serifs; that was the most important change. I find it more original now.

Each of my typefaces has its own history. The most recent one is Foro. It was produced in a relatively short time. Someone proclaimed 2012 to be the Year of Slab Serifs on MyFonts, so I figured it was time I do one of those! I must admit Foro was finished in just a few weeks. It has 16 fonts altogether: eight weights with italics.

Many designers need a lot more time to produce a new font family. How do you manage to be so productive?

I can handle the tools relatively quickly, because I’ve been doing this kind of work for many years. It’s true that I haven’t made finished fonts for a long time, but the process — producing and interpolating the characters, controlling and correcting — I can do all these things quite efficiently. Before I sit down, I have formed quite a precise idea in my head of the shape I want to see. And with digital tools you get a result so much faster! I haven’t drawn fonts by hand for a long time. Sometimes I do regret that a bit.

Your typefaces, whether with or without serifs, are all proper text fonts. Is type for body text your definitive niche, or can we expect display and script fonts from Hoftype in due course?

Of course you can make a broad array of fonts, in every direction. But that’s not my thing. I do text fonts. I like looking at other kinds of faces, but since I work alone, I must somehow limit myself. This work is enough for me. It is not so easy to make good text fonts, and it is a problem I like to delve into. I once thought “OK, I need to go and make something more popular,” and then did Corda. It has sharp serifs, and pointed, drop-like forms — so it is rather decorative. Corda is selling well, but rarely as a family. It is usually bought in individual weights, indicating that it is mostly used as a display font.

In the field of text fonts, I’m not particularly interested in pure replicas. There have been many revivals based on, say, a printed page by Manutius or Jenson, and people have discussed whether this or that version is most faithful to the original. All this is speculation, because you cannot know how the forms were actually meant to be. Even if I could reconstruct the original lead type… but the type is worn and the outlines on the page are deformed. Did the punchcutter take the modifications into account that are produced during multiple print runs on rough paper, or is it a random result? Those are pointless questions.

When I make typefaces that relate to historical forms — such as Cala or Sina — it is not my intention to make revivals. I just want to use what worked well in the old types, and apply that to today’s sense of form.

All your fonts on MyFonts are also available as web fonts. Your designs are very contemporary, and web fonts are the most recent important development in type. Does the issue of on-screen representation play a role in your design decisions?

I have received some good feedback on my web fonts, but I think that’s mainly because my fonts look relatively stable. As I’ve said, I always have in mind that a font should look good at 8pt. So, yes, a sturdy text font is likely to be a good web font. On the whole, my fonts seem to be well suited for the web. Apparently, it’s almost automatic — a happy by-product that I had not counted on. Of course, I don’t mind at all!

Thank you for sharing these fascinating stories with us!