Novel Constructions
the making of a typeface

A sudden bolt of inspiration would makes for an enticing story of a typeface’s beginnings, one that would perhaps be helpful when marketing it. However, in reality, not all typefaces come into the world that way. Sometimes, as was the case for Novel, the idea slowly percolates. Even the somewhat unspectacular name I chose for this family reflects that process.

Just like many of my fellow Type and Media graduates I was adamant about continuing to work on designing typefaces after graduation. But unlike most of them I never had to face the question of whether to continue on the project developed over the course of my study. My efforts were less than stellar, so I couldn’t wait to start from scratch.

Though there was no sudden moment of insight, I believe that with most type designers it’s very obvious which typefaces could have had an influence. As for me I admire Jan van Krimpen’s Romulus and Bram de Does’s typefaces Trinité and Lexicon. In addition, I can’t imagine ever getting tired of looking at Peter Verheul’s Versa. As I studied both graphic design and type design at the Royal Academy of Art, my affection for reading typefaces with broad-nib contrast and calligraphic details isn’t surprising.

For me the start of working on a typeface involves strictly being away from computer. The design process is rather simple. I spend a lot of time on sketching different ideas — at first quite roughly, but as soon as I find something I like I develop it further in greater detail. My favorite part of drawing letters is sitting down with a rather soft pencil and my sketch book. Most glyphs I draw have an x-height of about 6 cm, which enables me to redraw them quickly when I make mistakes, or consider alternative shapes. In this phase the design and characteristics of letters take precedence, so I don’t worry about inaccuracy in the overral rhythm. When I feel that I’ve found some ideas worth developing further, I strive to develop these letters as far as possible on paper. When I reach a phase where I discover the shapes I was searching for, I tend to switch to a Rapidograph as it permits greater precision. Those drawings have no gray scale and they sometimes brutally reveal weaknesses in the concept.

Designing typefaces this way takes a little longer at the beginning, but I feel that working this way provides me with a much better and clearer understanding of the shapes. I am also convinced that this way I am able to implement changes much quicker than if I had to think about nodes and path directions. Immediate contact with shapes that drawing with a pencil provides liberates me from thinking about font production and lets me concentrate on what matters most — designing the alphabet. It often feels that adjusting anchor points, nodes, and extrema are a distraction at this point in the process.

When drawing type I tend to focus on the darker book weight. That way I can imagine or envisage how the shapes will alter when made lighter or bolder. Precise drawings take a lot of time and effort. I try to work as efficiently as possible, even though I find drawing type very enjoyable.

Once I feel I have designed enough glyphs, I scan the drawings and begin to vectorize them. Here I begin to integrate the future interpolations in the design process. I use the glyphs I just digitized as a basis for the light and the bold weights. My preference to develop italics and roman at the same time makes this phase of the process very time consuming. A lot of the issues only become apparent when test prints at reading sizes are made. That is why in this phase some major design ideas might be altered or jettisoned all together.

There appears to be a growing trend in typefaces for italics whose design is fairly close to the roman, and I find this disconcerting. I think that highlighting some passages of text is more elegant when done with distinctly designed italics. While naturally roman and italic should have similarities, slant angle should not be the only means to differentiate them.

As lowercase constructions are more complex than the uppercase I felt it was necessary to have them a little more dynamic. For that reason lowercase letters are slightly more slanted than the uppercase letters.

Flexible f-terminal
In my early drawings I fell in love with the long f-terminal of Novel. This feature looked nice in many letter combinations, but not all. Letters followed by b for example, or when an accented glyph follows f looked very unattractive. The solution was a flexible f-terminal that would be wide when space permitted, show up as a ligature, or shrink if it were followed by an accented letter. All typefaces in the Novel family have this feature, except Novel Mono.

Swash-like Terminals
Novel is supposed to have a friendly appearance. Swash-like endings or terminals on round shapes like a, c, f, r, and y contribute to a natural feel. Also the calligraphic details on A, V, W, v, and w help to extend the warmth of design elements that have their origin in handwriting. As italics are closer to handwritten words, developing them came easier than the roman. Naturally these characteristics were not adopted to the sans serif version, lending it a more neutral personality.

Novel Sans
Creating a matching sans serif typeface was the plan from the outset. Only a few weeks after I digitized the first drawings of Novel, I began designing Novel Sans. I decided to make the two typefaces in parallel to make sure they were complementary. This way I could ensure that elements I designed for Novel would also work for Novel Sans. Obviously many shapes would not be problematic, but some of my favorite details proved to be difficult to adopt. For example, I spent considerable time on letters such as lowercase v, w, x, y, z before I was confident that I could go for a particular design feature without fear of later regretting it.

Novel Mono
When Novel Sans was nearing completion, friends suggested that I should try to make a monospaced version. I must admit that at first I wasn’t taken by this idea. I thought that too many of Novel’s design features, such as its Classic proportions, a generous roman combined with a pretty narrow italic, and the warmth would get lost. Despite these reservations, I decided to give it a go.

Of course there are differences to be seen, especially between the italics of Novel Sans and Novel Mono, but when comparing both designs side by side it’s more obvious what they have in common. The final outcome surprised me. It works much better than I ever thought it would. In fact, of all the styles, I use Novel Mono most.

Novel Sans Condensed
The classic proportions of Novel and Novel Sans might be considered distinct characteristics of the family. The Sans Condensed version, however, required a much more balanced rhythm. The width contrast between the romans and the italics had to be decreased, as the italics of Novel Sans were already very narrow.

There are many ways to measure legibility, with some more legitimate than others. But with all of them I have difficulties when it comes to the most interesting aspect: What is the reader accustomed to looking at? I don’t think that actually measuring legibility is nonsense, but it’s only one of the many aspects or ingredients of a typeface design. I believe a good type designer instinctively makes the right decisions when faced with the question: Should I go for the distinctive detail or for reading quality? Reducing Typeface design simply to considerations of legibility seems to be unfair and perhaps misguided.

When I was designing Novel, I printed high resolution proofs of different type sizes to judge quality and legibility. To me it helps much better discussing legibility with colleagues like Albert-Jan Pool, rather than spending time on eye tracking or mathematic grayscale calculations. 

Christoph Dunst studied at the Royal Academy of Fine Arts in The Hague, The Netherlands, where he graduated with a degree in graphic and typographic design and a masters in type design. The design studio he founded in The Hague in 2006 moved to Berlin in 2009 and was renamed Büro Dunst. In 2012 he established the Atlas Font Foundry.