The design of Skolar Sans Arabic

This text was first published on 15 November 2016 in a slightly modified version on Rosetta’s website at the occasion of the launch of Skolar Sans Arabic.


When I received the brief for the design of Skolar Sans Arabic, David made it sound fairly straightforward. It was an extension to an existing design, wherein a significant part of the conceptual work had been done. David described the whole family as a system with ‘universal’ aspirations, and as a typeface that had been designed to look good on the web. It is a sans-serif design conceived for continuous reading and developed to cater for complex typography, as suggested by the name. According to David ‘nothing deep really’.

While one may differ about this assessment for the Skolar Sans family as released in 2014 – by most standards a typeface of exceptional complexity – translating these characteristics to an Arabic design proved to be challenging. Here, the notion of a sans-serif, for example, is everything but straightforward. In the Latin script, the sans-serif genre has been solidly established for more than a hundred years, and had ample time for evolution, design movements, counter-movements and even plentiful revivals. In Arabic typography, however, sans-serifs are an utmost novelty, seriously pursued only since the last ten years (if one is generous). While the model of the Latin sans-serif towers above any Arabic attempts at the genre and is bound to influence its inception, the definition of an Arabic sans-serif is a work in progress with open outcomes. Recently many Arabic typefaces conceived as companions or extensions to established Latin sans-serif designs have fallen into the so-called Kufi strand, defined by geometric, strongly modular, and often static typeforms. To me, however, it is all but certain that this approach is most suitable for this genre, especially if the typeface is intended to be used for text, rather than headlines. In my view a contemporary Arabic design might as well be based on a (typographic) Naskh structure, and for Skolar Sans with its explicit text orientation it was my direction of choice. The fundamental nature of the decisions and design approaches that are only emerging for contemporary Arabic typeface design is also apparent in the terminology: in the absence of serifs a ‘sans’ companion makes little sense, and it is more useful to refer to such Arabic type as ‘low-contrast’ designs until a better term has been found.

It comes as no surprise then, that finding the right amount of typographic contrast was one of the first issues that arose in the development of Skolar Sans Arabic, and finding an answer was not self-evident. Many of the most frequent and important Arabic letterforms – alif, lam and bah, for example – could easily have a completely monolinear design because of their simple skeletons. Yet, some of the most frequent and character-defining Latin letters – think of n, e, and a – by their very nature require a significant amount of contrast. If the goal is to match the Arabic extension to the Latin model, should then the key letterforms be drawn with (unnecessarily) pronounced contrast? Or in other words, should contrast be consistent within a script or between multiple scripts? Or should individual typeforms be allowed to diverge from general principles if it contributes to the overall appearance of words and sentences? In the design of Skolar Sans Arabic I began with fairly geometric letterforms. Alif and lam, for example, were initially based on straight rectangles – following the simple stems of the Latin letters I and l –  with minimal adjustments for the joining glyphs. Yet, the results in my mock-up text did not convince me: the frequent pairing of alif and lam looked too rigid in comparison to the joined horizontal letterforms, breaking the flow of the line.

Similarly, the design of the diacritical dots required extensive testing and frequent revisions. Since it was going to be a companion to a sans-serif, I queried if the traditional rhombic dot would be suitable and assumed that a different configuration would work better. As the dots in Skolar Sans are round, I began with rounded, but slightly more squat dots for the Arabic too. This seemed to work reasonably well, but neither David nor I were fully convinced: because of the high frequency of Arabic diacritics the round design felt somewhat too jovial. Thus, I made a range of variations, half-rounded, at different angles, rectangular. Yet, even if a drawing was successful as a single dot, its repetition in double and tripple dots was usually not. Especially if the angle was half-way between the conventional rhombic dot and the square, the nesting of the three dot combinations did not work well. We carried on with a slightly rounded-off rectangle at a very shallow inclination, but I was not yet satisfied. Eventually, after weeks of development, testing and revision, I resorted to a much more conventional, classical configuration, more akin to the rhombic dot. Seeing its effect across the typeface, to me it became instantly clear that this was the way to go, and it demonstrated why wheels rarely need to be reinvented.

Many more questions and design challenges such as these made this a particularly exciting and interesting project to work on. In this process David was an ideal partner as his confidence in my work never descended into uncritical reliance, asking pertinent questions and raising relevant observations. In all the design decisions that were eventually taken, the ultimate goal was always to arrive at an authentic, original and functional interpretation of Skolar’s design for the Arabic script, extending the reach of its voice to many more countries and languages.