“I have the honor of introducing the finest punchcutter”, announced the herald presenting Robert Granjon on September 10, 1579. Granjon himself used to sign contracts with printing houses under the title “bookseller” (Marchand libraire). But unlike other 16th-century punchcutters who worked within type foundries — Granjon owned no studio of his own. Instead he wandered across Europe — Paris, Rome, Antwerp, Geneva, Lyon and more — selling matrices of letters and ornaments of his own design. Over the course of his career he is believed to have created some ninety typefaces: thirty italics, twenty romans, seven in civilités style (imitating Gothic script), nine in Greek, half a dozen musical characters, several Hebrew typefaces, and countless ornaments and arabesques. But his greatest typographic fame rests on the italic style he designed, which became a universal standard.

The year is 2005. Type designer Kai Bernau is a graduate student in the Type and Media program at the Royal Academy of Art, The Hague. In a course on type revival, he develops the Lyon type family — a digital version of letterforms based on several historical models by that same French punchcutter, Granjon. Lyon was designed as a family of two complementary styles: Text and Display. Each Latin style includes five weights: two regular weights, semibold, bold and black — each with a corresponding italic. Lyon engages with the legacy of 16th-century texts, but does not adopt it blindly. As Bernau himself states: “Lyon is not too coldly modern, but neither is it too historically burdened”.

Today, Lyon is distributed through the website of Commercial Type, a type design studio operating simultaneously in New York and London, founded as a joint venture by Paul Barnes and Christian Schwartz. The studio built its reputation through the award-winning Guardian Egyptian typeface designed for The Guardian newspaper (2004–2005). In 2018, the studio approached Yank Yontef and Daniel Grumer of Fontef Studio to design a Hebrew version of Lyon — not their first collaboration. It had been preceded by a successful partnership creating a Hebrew version of the grotesque typeface Graphik.

When adapting an existing typeface to a new language — as in translation — a sensitivity to the nuances and cultural, historical and visual dimensions the original carries is required. “It is not a matter of borrowing forms from one language and sculpting them into another”, explains Grumer. “Issues relating to reading habits and readers' prior familiarity with the forms are at stake”. Alongside the historical weight Lyon carries, the Hebrew version also had to account for the fact that Lyon is a diverse and complex typographic system capable of serving varied content. Furthermore, the Arabic version of Lyon was underway, with its designers — Waël Marcos and Khajag Apelian — basing it on a secondary Persian script known as Nasta'liq, an expressive and flowing style that developed in Iran during the 14th and 15th centuries. The fact that the Arabic version drew on a secondary script prompted Yontef and Grumer to seek out Hebrew writing styles from those same centuries.

The distinctiveness of the Hebrew version begins with the working process Fontef Studio chose. To locate Hebrew script styles characteristic of the 15th and 16th centuries, Yontef and Grumer traveled to the Plantin-Moretus Museum in Antwerp, Belgium. Named after Christophe Plantin and his son-in-law Jan Moretus, the museum was once the home of the Plantin printing house and publishing firm they established in the 16th century. After it closed, the city of Antwerp transformed the premises into a museum with an extraordinary collection of typographic artifacts: antique printing presses, original matrices and punches used to create typographic classics, manuscripts, rare books and Bibles printed in three languages — Latin, Greek and Hebrew.

A punch (patrix) is a rod with the desired letter form engraved at its tip. The tip is struck into soft metal, which becomes the matrix (matrice). Metal is poured into the matrix and the letters are reproduced. Before appearing as a black mark on paper, each letter undergoes a process that takes it far from its original form: it is engraved, stamped, cast and finally inked and printed with a press. For this reason, Yontef and Grumer chose to photograph the original matrices at high resolution using a digital microscope. This allowed them to get close to the original letterforms and reconstruct them with precision. From the photographs they created a first version, which was then processed to be legible for contemporary Hebrew readers and to correspond formally with the Lyon family.

In English, the convention that italic type helps distinguish between kinds of information within a text is well established. In Hebrew, by contrast — in the absence of suitable typefaces — printers had to find alternatives such as underlining, bold letters or a different typeface from the running text. There is also no consensus on the direction of the slant: “Some think the letters should lean to the left, in keeping with the direction of Hebrew writing; others — to the right, like Latin”, explains Yontef. In searching for sources for a secondary typographic structure, they chose Rashi script. In the period when Granjon was active, Rashi script served as the typeface for commentary on source texts. The main text appeared in precise, careful “square script”, while the commentary was set in an intermediate, flowing script.

In the first stage, Yontef and Grumer created a revival — taking letters from a previous technology (metal) and reconstructing them in a current one (digital). They scanned the Rashi script found in the archive and reconstructed it with minimal changes and close fidelity to the source, knowing that for contemporary Hebrew readers the letters would not be legible. In the next stage, they made more significant structural changes to certain letters to bring the forms closer to the eye of the modern reader.

Letters such as alef, tzadi and shin were unrecognizable and required special treatment: the very high junction between the arm and the backbone reduced the letters’ upper white space almost to the point of its disappearance. The letters samekh and final mem also required special attention, and illustrated the close relationship between language, form, material and even economics. It turns out that in the past, for reasons of economy and convenience, punchcutters used the same punch for both samekh and final mem. To distinguish between the two, an additional pit was struck separately into the matrix of the final mem. The same approach was used when a dagesh (a diacritical dot) was needed within a letter.

The continuation of the process involved larger changes that moved the italic version further from its calligraphic source. In the updated version, line endings that originally appeared as drops — the result of hand movement and the properties of the writing instrument — became more absolute and geometric, and the pointed serifs became more angular, as if cut into truncated triangles.

The natural candidates for creating the running text and display letters in Hebrew were examples by 16th-century punchcutters Guillaume Le Bé and Daniel Bomberg. But the moment they placed both models alongside Lyon, they realized that the historical and cultural gap between the two scripts was very wide. Through study, research and work with both models, Yontef and Grumer came to understand that Latin typefaces — such as Lyon — manage to distance themselves from the historical and conservative weight they carry and respond to contemporary everyday content. In fact, 16th-century Latin typography, such as Bodoni and Didot, is widely used today in branding for luxury fashion houses. Hebrew models, by contrast, struggle to detach from that weight. With this insight, they decided to explore a different direction, drawing on canonical Hebrew typographic classics from the 20th century.

As noted, the task was to create a typeface capable of carrying contemporary and varied content as running text. Until now, the Hebrew reader’s eye had grown accustomed to reading running text primarily in Frank Rühl (Raphael Frank and Otto Rühl, 1908) — the most widely used running text typeface in the Hebrew language. An attempt by the newspaper Maariv on March 6, 1987 to replace the typeface was met with such a torrent of criticism that the paper was forced to return to the familiar face. Frank Rühl maintains a moderate relationship between thick and thin strokes, but for this project they recognized it suffers from a lack of versatility. Lyon is a family comprising five weights, whereas Frank Rühl historically includes only two — regular and bold. Yontef and Grumer decided to test the classical typeface’s ability to carry extreme weights, designed several versions, and concluded it would struggle to meet this need.

The next step was an analysis of two additional canonical styles: David (Ismar David, 1954) and Narkiss (Zvi Narkiss, late 1960s). Unlike the contrast in Frank Rühl — based on pointed-nib calligraphy — the calligraphic structure of David and Narkiss is based on broad-nib calligraphy. Using this tool creates a different distribution of black and white, and Yontef and Grumer estimated that this characteristic would allow for a more natural transition and growth across weights, including the very heavy ones. This was the direction chosen. The next version drew on the principles of broad-nib calligraphy.

Designing a letter is a delicate craft requiring sensitivity to the complementary contrasts of black and white and the relationships between them. In the case of the Hebrew version of Lyon, it was a sustained process of distillation, refinement and the removal of curves. At the heart of the abstraction process lay the understanding that a moderate move away from calligraphic elements means a move toward a more contemporary essence. For example, the variation in stroke thickness along the diagonal in alef and tzadi was flattened, and these diagonals are now completely straight. The curvature created beneath the serif of the letter bet as a result of the in-stroke was replaced by a straight line and a sharp angle. Along the way, the distribution of contrast in the final kaf was also changed — from two thick focal points with a thin one between them, to a single thick focal point at the top and thin at the bottom.

Where letter design once required repeated drawings on paper, today automation assists in designing letterforms comfortably and efficiently. Most typefaces today are designed in software called Glyphs. In the first stage, the masters are designed — representing the extreme characteristics of the typeface, such as the light and heavy weights. Based on the visual information in the masters, an automated process produces intermediate weights — interpolation. Using interpolation requires the type designer to ensure that points and curves in the masters are built in correspondence, to review the intermediate weights, and to refine the design decisions the software has made automatically. Lyon was originally designed with three masters: light, regular and black. At a certain point, Yontef and Grumer realized that the bold weight — produced by interpolation between regular and black — looked too light. They therefore turned the master previously called “black” into a lower weight — “bold” — and created a separate black weight. Breaking free of the interpolation constraint released them from limiting technical restrictions and allowed greater formal freedom, producing a black weight that was freer and more refined. This made it possible to enlarge the internal white space of the letter pe, intensify the black mass in the diagonal of the final tzadi, and widen the tip of the right tooth of the letter shin almost to the point of the right arm’s disappearance. A further decision that gave the display style its sense of refinement was to design the diacritical marks as diamond shapes. Unlike the text punctuation marks, which remained round, the diamond-shaped diacritics emphasize the high contrast of the display font. These fine details create graphic interest and attest to the careful attention given to small particulars in this typeface.

The work on the Hebrew version of Lyon was a process of formal improvement and refinement that lasted about two and a half years. To test whether the Hebrew version successfully conveys the qualities of the original typeface — both in essence and graphically — Yontef and Grumer created a Hebrew mockup of an Italian magazine. The Italian magazine originally used the Latin version across its full range of weights, and the comparison highlighted the similarities while also surfacing gaps that were then addressed.

The Hebrew version of Lyon has already made its debut at the Bezalel Graduates Exhibition — the Academy of Arts and Design, Jerusalem (design: Eli Kharoumov and Shai Ben Ari, 2022). It seems that letters, like people, have qualities, a family and relationships. They also carry a historical weight — a burden they remember, engage with, respond to and learn from. Perhaps this is not a story about letters at all, but a story about people who are able to see both the black and the white.