Söhne design information
Söhne is the memory of Akzidenz-Grotesk framed through the reality of Helvetica. It captures the analogue materiality of “Standard Medium” used in Unimark’s legendary wayfinding system for the NYC Subway.
I first visited New York City in 2010. Despite all local advice, I arrived in the summer. It was a brutish, oppressive heat that I’d not experienced before. Descending from the hot streets through the humid subway stations and into the chilly carriages amongst a press of people was a bodily experience.
Of course, I was excited to finally experience Unimark’s legendary wayfinding system.¹ But it was not as I imagined. My entire understanding of the subway signage was based upon idealised, cherry-picked examples and reproduced snippets of the NYCTA Graphics Standards Manual.² The implemented system is messy, inconsistent and dirty. As Christopher Bonanos writes, “it is imperfectly deployed, even after four-plus decades, and probably always will be”.³ The signage system wasn’t obvious and foregrounded as I expected. It felt recessive, like it had worn into the surrounding environment.
Standards Manual, LLC, “1970 New York City Transit Authority Graphics Standards Manual”, (2014).⤴
The reality was a shock, but illustrates the stark differences between brand guidelines and their real-world implementation. Like seeing a project in a design studio’s portfolio versus the actual thing, or the difference between a photo of the Big Mac on the menu and what you’re actually served. William Foster would understand.⤴
After I became accustomed to the noise, heat and general intensity of the subway I managed to focus on the signs. I knew the letterforms were screen-printed renditions of Standard Medium, originally known as Akzidenz-Grotesk Halbfett in Europe. The letter-spacing was tight and confident. The white letterforms had presence. I remember admiring the angled terminals on the c e s, their anonymous grotesk heritage felt perfect in context.
Several years later I started a half-hearted digitisation of Linotype’s Akzidenz-Grotesk as an unrelated side project, but it was dissatisfying. Possibly because Linotype’s version of Akzidenz-Grotesk was adapted from Berthold’s foundry type to their specific unit-based system. It lacked the spirit and materiality I associated with Akzidenz-Grotesk. Remembering the graphic impression of the NYC Subway signs, I realised that my understanding and love of Akzidenz-Grotesk orbits around the gravity of its Halbfett and Fett weights. I’ve always been drawn to bolder letterforms. I prefer meat on my bones.
Accidenz-Grotesk specimen, Bauer & Co. and Berthold, (ca. 1912).
The base style of Akzidenz-Grotesk — its ‘regular’ weight — was published in fifteen sizes by H. Berthold AG in 1898. The first proper addition to the family was published by Berthold and Bauer & Co. in 1902/03. This lighter-weight design was initially sold under a unique name: Royal-Grotesk. It wasn’t renamed Akzidenz-Grotesk Light until the 1950s. Akzidenz-Grotesk Halbfett and Fett joined the family in 1909. By 1911, Berthold and Bauer & Co. had expanded the Akzidenz-Grotesk family to include a total of six styles with the term Akzidenz-Grotesk in their names. Forty-seven years later, in 1958, the number had grown to thirteen. By 1968, there were twenty-one.⁴
Standard, Berthold type specimen No.539A, (ca. 1956).
Standard Condensed styles.
Standard Extended styles.
Akzidenz-Grotesk became available in the United States around 1957. The fonts of foundry type were sold by a New York company named Amsterdam Continental, a subsidiary of Dutch type foundry N. Tetterode. Amsterdam Continental had an exclusive license to sell the typeface in the USA, but didn’t market the fonts as “Akzidenz-Grotesk” — apparently it was difficult for English speakers to pronounce. They called it “Standard” instead.⁵ This is the name that appears in Unimark’s NYCTA Manual.
“This was much easier to pronounce and almost implies that Akzidenz-Grotesk is the default variety of sans serif type. I think that this was a brilliant marketing move, but I have not yet been able to find out whether this was a decision made at Berthold, Tetterode, or Amsterdam Continental.” — Dan Reynolds.⤴
I was not interested in making a facsimile of Standard Medium or an “authentic” version of Akzidenz-Grotesk.⁶ I don’t believe it’s possible to make an “authentic digital revival” of any analogue typeface. It’s a contradiction in terms. There is no equivalence between physical material (metal type) and a virtual, cartesian vector (digital font). The only way to make an authentic digital revival would be to literally copy and paste. My primary motive was to express the analogue materiality of the letterforms in a digital font. I wanted to capture the memory and sensation I felt from the subway signs. Not the actual physical construction of signs — enamel paint on steel — but their graphic impression.
I determined the sensation (of analogue materiality) was from white letterforms with close spacing, low contrast and angled terminals, activated by a black background.⁷
“We can make them look old, like the original types, or we can make them look fresh. We can’t, however, make them look identical to historical models, for digital type is not metal type. The two are different creatures and they manifest separate identities.” — John Downer, Tribute type specimen, (2003): 9.⤴
I’ve always been attracted to white type on black. I prefer it for display. It looks powerful and ethereal. Instead of light being absorbed by black letterforms, letterforms are reflected by light. My first exposure to digital type on screens was DOS. In the 1980s I managed to break the proto Windows-like GUI on our family’s 286, and it would only boot to DOS. A black CRT screen with pixelated monospaced white type is burned into my psyche.⤴
Akzidenz-Grotesk Halbfett, 36pt.
Söhne Halbfett, −19 track.
Söhne is designed from the Halbfett (semibold) weight outwards. I normally start with a regular weight, because that’s what is most often used. Söhne is different, its entire conception is from the subway signs, and thus centres upon those terms.
Specifically, Akzidenz-Grotesk’s 36pt Halbfett was the starting point. It provided Söhne with family-wide proportions. This also freed me from the usual problem of which specific metal size to use for reference.⁸ Akzidenz-Grotesk is a typeface family born of marketing, not design planning. It was cobbled together from various different cuts made over 70 years. This is quite different to our modern understanding of a planned typeface family, like Univers for example. In practice, this means there is no definitive, true version of Akzidenz-Grotesk Halbfett in metal. There are several slightly different versions of it, all grouped together under the name “Akzidenz-Grotesk Halbfett”. Each letter in each font prints slightly differently, further diluting the defining version.
Accidenz-Grotesk specimen, Bauer & Co. and Berthold (ca. 1912).
Akzidenz-Grotesk Specimen, Berthold, (ca. 1956).
Plakatschrift (poster type) specimen, Berthold, (ca. 1956).
Akzidenz-Grotesk Specimen, Berthold, (ca. 1956).
Akzidenz-Grotesk never really had an “official” launch with dedicated publicity and a definitive specimen. Piecemeal, quotidian specimens of its various cuts were made over the years. It’s not from type specimens that a typeface becomes generally known and understood. Instead, we understand it when designers embed it into the cultural milieu, when it becomes part of the fabric of our daily existence.
However, as a typeface designer, when I think of a typeface it’s usually a specific glyph or specimen. Thinking of Baskerville brings the open loop of its g, Helvetica is the teardrop counter of a, Bodoni makes me visualise the Manuale tipografico and Univers the famous family grid. Akzidenz-Grotesk is one of the few typefaces I visualise in context: Unimark’s subway signage, Marber’s Penguin Crime book jackets, Müller-Brockmann’s posters and Weingart’s letterpress work all jostle for my attention.
“Schiff nach Europa”, designed by Karl Gerstner, (1957).
“die neue Graphik”, designed by Karl Gerstner and Markus Kutter, (1959).
“Neue Grafik, No.3”, designed by Hans Neuburg, (1959).
“Gestaltungsprobleme des Grafikers”, designed by Josef Müller-Brockmann, (1961).
Poster, “Richard P. Lohse, Max Truninger, Kunsthaus Zürich”, designed by Richard P. Lohse, (1962).
Poster, “Lokal - National - International - Nationalzeitung”, designed by Karl Gerstner, (1960).
Poster, “Opernhaus Zürich - Die lustige Witwe”, designed by Josef Müller-Brockmann, (1966).
Large drawing template, designed by Josef Müller-Brockmann, (1969).
Grammo Studio Catalogue No. 2, designed by Siegfried Odermatt, (1960).
Poster, “Schauspiel Stuttgart, Spielzeit 2013/2014”, designed by Spector Bureau: Markus Dressen, Jakob Kirch, Katharina Köhler, Jan Wenzel, (2013).
“Landscape with dead dons”, designed by Romek Marber, (1963).
Poster, “Picasso - Braque - Chagall (…) - Van Abbemuseum Eindhoven”, designed by Wim Crouwel, (1962).
Poster, “der Film”, designed by Josef Müller-Brockmann, (1960).
Poster, “Theater Nucleo Eclettico”, designed by Lorraine Ferguson, (1982).
Akzidenz-Grotesk didn’t need specimens. It was championed by the best designers at the time and touted as being “the only [type] in spiritual accordance with our time.”⁹ — the best marketing a typeface could ever have. Used in iconic design work, canonised in annuals and design history has cemented it as a classic. And when I think of Akzidenz-Grotesk used by the modernists, it’s only two weights: Halbfett and Fett. These solid, confident poster-like qualities resonate with my personal concept of Akzidenz-Grotesk. Many posters during this time didn’t use type ‘off the shelf’, the fonts simply weren’t large enough. Type on silk-screened posters was often photographed, enlarged and re-drawn by hand. Designers exercised fine manual control to achieve tighter spacing or adjusting certain letterforms to their preference. This analogue translation humanised the sparse, strict grids of Swiss modernism, echoing my feeling of materiality from the subway signage.
Blockschrift, Genzsch & Heyse specimen book, (ca. 1911).
Breite halbfette Grotesk, J.G. Schelter & Giesecke specimen book, (ca. 1900).
Breede Halfvette Antieke, Lettergieterij Amsterdam specimen book, (ca. 1920).
During the modernist period several grotesks were in action, including Breede, Breite and Blockschrift. It’s hard to determine the exact origin for some of these fonts. Dan Reynolds explains, “the same fonts of type, occasionally under different names, could be available from multiple sources.”¹⁰ For example, Dutch type foundry N. Tetterode sold Breede Halfvette Antieke, which they obtained from the German J.G. Schelter & Giesecke foundry. Schelter & Giesecke sold it as Breite halbfette Grotesk, which was probably published in 1890. The Italian foundry Nebiolo sold it as Grotteschi Neri. Sometimes a foundry would further modify certain letterforms after obtaining and renaming a font from a supplier.
Söhne Halbfett, default a.
Söhne Halbfett, alternate a.
Söhne Halbfett, default g.
Breede Halfvette Antieke.
Söhne Halbfett, alternate g.
Regardless of origin, these typefaces have similar qualities to Akzidenz-Grotesk: largely anonymous, robust forms with a satisfying heft. Söhne’s default a and alternate g are taken from these analogous grotesks. The tail on the a is a personal preference.
Akzidenz-Grotesk Schmalfett, 6–12 Cicero, Berthold Plakatschriften Specimen Nr. 468, (ca. 1957)
Akzidenz-Grotesk Schmalfett, 16–28 Cicero, Berthold Plakatschriften Specimen Nr. 468, (ca. 1957)
At first I didn’t pay the small point-size Akzidenz-Grotesk condensed cuts in my specimens much attention. I wasn’t even sure if they were really part of the family. In one specimen they’re called Enge Steinschrift and Halbfette Bücher Grotesk, in later specimens they’re Akzidenz-Grotesk Condensed Heavy and Akzidenz-Grotesk Condensed Bold. Once again, similar but unrelated styles were requisitioned for the family. But when I saw a Berthold Plakatschriften (Poster-Type) specimen, the scale and density of Akzidenz-Grotesk Schmalfett was impressive. Seeing it big changed my mind. It had a similar materiality to the subway signage and confirmed that Söhne needed condensed styles.
Akzidenz-Grotesk’s condensed styles vary in their horizontal and vertical proportions. The two lighter weights have slightly round counters, the middle weight feels much more condensed that the rest, and the two bolder weights are taught and square. Enge Steinschrift provided the proportions for Söhne Schmal, like Akzidenz-Grotesk Halbfett for Söhne.
Akzidenz-Grotesk Skelett, Breitmager, Breit, Breitfett.
Akzidenz-Grotesk Skelett, Breitmager, Breit, Breitfett.
Akzidenz-Grotesk Skelett, Breitmager, Breit, Breitfett.
Akzidenz-Grotesk Skelett, Breitmager, Breit, Breitfett.
Akzidenz-Grotesk had five breit (wide) styles by 1961: Akzidenz-Grotesk Skelett, 1911 or 1914; Breitmager, 1911; Breit, 1908; Breithalbfett, 1961; Breitfett, 1957. Of all the Akzidenz-Grotesk family members, these cuts are the most disharmonious. The proportions, skeleton and details vary wildly from weight to weight. The Skelett and breitfett cut are the worst of them all. Skelett feels like a weird cousin of Venus;¹¹ Breitfett feels artificially stretched rather than sensitively drawn. They are only unified by width.
Breite Accidenz-Grotesk, 28pt.
Söhne Breit Buch.
Wide fonts are hard to get right, especially if they’re modelled on an extended oval rather than a rectangle, like Eurostile. If the oval forms are too wide, it starts to feel goofy and spacing becomes very tricky. Breite Akzidenz-Grotesk had the most promise, it felt like it was actually related to the normal widths. It’s slightly too wide and gets into squashed oval territory. Söhne Breit tempers the width and keeps the deep crossbar on G, tight bite of e, and amplifies the serpentine curves of s. Again, Söhne Breit works from the middle weight out, keeping consistent proportions and details across 8 weights.
Söhne Mono was the final family added to the Söhne Collection. I drew it in deference to contemporary typography and the ever evolving concept of a “super family”. Aesthetically, monospaced fonts are always handy for a typographic palette. They have a lovely, brutish quality. Outside the realm of programming fonts, there’s not really any technical need for monos.¹² Few typefaces adapt well to the regularised structures of the mono system. Söhne posed a good challenge to adapt to mono space. Its overall proportions are narrow with tight spacing; most monos are wide with loose spacing.¹³ Adding slabs to i j l r retains the tight spacing and lends a more technical aesthetic. Black weights are very hard to mono space, especially if the proportions are narrow. The challenge was to see if Söhne would convincingly adapt — and it does.
Adapting monospace typefaces has recently become de rigueur. Oliver Reichenstein made IBM Plex Mono “duo spaced”. Dinamo and Stephen Nixon made variable fonts with a “mono” axis for Monument Grotesk and Recursive. It seems the aesthetic of monospacing is more desirable than the actuality of a single width.⤴
“South side, Queens Blvd” is Standard Medium. “North side, Queens Blvd” is Helvetica Medium.
In 2017, when Söhne was almost finished, I returned to NYC.¹⁴ Dropping down into the subway I expected Standard but was shocked to see it was mostly Helvetica.¹⁵ How did I not know this? Had I mistaken my earlier memories of Standard for Helvetica? Had I simply seen what I wanted to see — Helvetica as Standard? To be fair, without the distinguishing letters like a c e g s they are pretty tricky to disambiguate at a distance… but that’s no excuse for even a greenhorn type nerd.
Reading A General Theory of Love I learned memory is fallible and subject to change. Memory is “not carved into the rock” like we imagine it to be.¹⁶
“For starters, memory within the brain is fluid, […] the brain scatters its memory treasures across a number of individual connections. The boon of distribution is security; the disadvantage, infidelity. A brain can lose a neuron here and there and the stored data suffers relatively little, a property of neural network memory termed “graceful degradation”. But as new facts rain down upon and trickle through the network, some older links are dissolved. Dissimilar information patterns can co-habitate in the brain without much mutual disturbance, because they rely on largely different sets of neurons. But like patterns jostle and coincide and overlie, and cannot avoid erasing and emending each other’s contours.”
I realised that I was remembering Akzidenz-Grotesk, but Helvetica was guiding my hand. They are similar typefaces, and in my mind they had erased and emended each other’s contours.
Akzidenz-Grotesk feels noble, humble and authentic. Nobody knows who designed it. Someone cut its forms, decided on spacing and christened it the most quotidian of names.¹⁷ Its anonymity cements its authenticity. Swiss Modernism, the wellspring of our contemporary graphic design practice, solidified and re-framed it into the typographic canon. Typeset within newly discovered grids and post-war aesthetic optimism, Akzidenz-Grotesk was reborn.
Helvetica, however, nowadays has the whiff of corporate slickness, a digital default that both sustains and saps the spirit. When it was released in the 1950s it was a revelation. Miedinger took Akzidenz-Grotesk Halbfett as the starting point — already 60 years old by then — and modernised it superbly.¹⁸ The physicality and materiality of tightly spaced, well printed (foundry type) Helvetica is still sublime. Since then, Helvetica has transitioned through machine and photo composition, and is currently embalmed as a myriad of digital corpses. None of them exactly capture the exact original atmosphere, but collectively they continue the idea of Helvetica.¹⁹ It’s the idea of Helvetica that’s important now. I don’t want to re-hash tired debates about it, we should let straw-men lie. Suffice to say that what it was and is now are different things.
Helvetica’s power lies in its obviously designed nature. It feels like it was made for the maturing profession of graphic design and corporate identity. I imagine designers in the 1960s connected immediately and intuitively with it. It represents the ascendency of graphic design over book typography, which was the de-facto standard bearer for typographic standards and discourse.
As a typeface designer it’s hard not to engage with the concept and reality of Helvetica. It provokes strong feelings from within the industry, ranging from outright scorn, through to jealously and adoration. It seems like there are no neutral feelings towards a supposedly neutral typeface. Despite this, I strongly suspect that any one of us would secretly love to profit from drawing “the new Helvetica”. A smash hit typeface is impossible to predict, but we can take some design strategies from it. Namely adapting existing typefaces to our time, systematic family planning, and tailoring the aesthetic precisely to the needs of contemporary designers.
Once a typeface is in the canon, it’s in the canon. Garamond, Jannon, Fournier, Bodoni, Akzidenz-Grotesk, Futura, Times, Helvetica, all of them — they cannot be ignored or even forgotten. We have to maintain our awareness and be mindful of their presence in our working lives. No matter how hard a modern typeface designer wants to push back or pretend they don’t exist, they are very much present in the daily working lives of designers. And of course, without designers using our fonts our work means nothing.
The term ‘Akzidenz’ comes from the German ‘Akzidenzen’, a catch-all word meaning ‘jobbing printing’. This referred to jobs that weren’t ‘proper’ book work, or body text in magazines and newspapers. Printers kept their legible serif types for ‘proper’ work – or their Frakturs, which they would have found just as easily legible – but when it came to setting things like ads, business cards, forms and such, they needed something else. ‘Display type’ usually refers to large-sized printing, but the results of ‘jobbing printing’ were often small pieces of paper you could hold in your hand.⤴
The fourth advantage I see in the word “design” (in addition to its modesty, its attention to detail and the semiotic skills it always carries with it), is that it is never a process that begins from scratch: to design is always to redesign. There is always something that exists first as a given, as an issue, as a problem. — Bruno Latour, “A cautious Prometheus? A few steps towards a philosophy of design”, The Design Philosophy Reader, (2019): 20.⤴
“There are no longer, as there were with metal type or early photocomposition, a few well-characterized renderings of particular designs whose appearance is fully defined by their manufacturers’ specimens. Instead, multiple versions of the same design cluster around each other in a space that embraces the whole range of possible appearances of a particular mechanically written script.” — Richard Southall, “Printers type in the twentieth century”, (2005): 137.⤴
Let me repeat our principle: not to design new faces but to improve (where possible) and to develop the best, perfecting them to the utmost and making them conform to a governing principle as far as possible.
Why is this conformation to a governing principle so important? Because we set such a store by being able to combine the letters harmoniously and without restriction. But is the typographer not free to combine as he likes? No, all he can do is combine the material available. And in our opinion that is too little.
— Karl Gerstner, “The old Berthold sans-serif on a new basis”, Designing programmes, (1964): 34.
My sincere thanks to Tim Kelleher for his encouragement and early adoption of Söhne, to Indra Kupferschmid & Dan Reynolds for their patience and expertise with German font naming conventions and history, DIA studio for the campaign films, and Dave Scaringe for his subway photography. I also thank Jesse Reed & Hamish Smyth at Order for permission to use the standards manual photograph and Letterform Archive for various specimen photography. All trademarks are the property of their respective owners.