Interview by students of the University of Tehran

In March 2014 I was contacted by two students of the University of Tehran, Maryam Khaleghi Yazdi and Ramin Pahlavan Hoseini. They were working on their MA thesis under the supervision of Dr. Sedaghat Jabbari. With the working title ‘A Research on Diverse Methods of Designing Farsi/ Arabic typefaces’ Maryam and Ramin wanted to explore current practice and ask professional designers for their perspectives. Against this background, they approached me with questions about my work and more general views on the subject, and over the course of the following months a kind of email-interview emerged. The questions by Maryam and Ramin where interesting and thought-provoking, and I thought might be of interest for others too (eventually the two students did not complete their project together, but Ramin finished it on his own while Maryam submitted a different thesis). With their agreement, I publish the interview here, edited only for consistency of format and small typos.

(Questions in italics, my answers in roman)


– Could you please divide the procedure of your design into some specific phases and explain each part briefly?

This depends on the project, which can generally be divided between commissions and self-initiated designs. The latter are much more idiosyncratic, so there’s more of a general process in the former. There it starts with the brief, client-interview, assessment of needs and establishment of best solutions for the client; Research into the specificities of the project (what use, what medium, what languages/scripts, what voice/style); Drafting of ideas, sketching of letter forms, development of first outline drawings; establishment of crucial letter forms (adhesion), testing in real text; revision and testing; expansion of the glyph set (project-dependent).


– How do you deal with calligraphy? Do you think it is necessary to design the typeface based on a specific Arabic script style?

Arabic typography still retains a strong connection to handwritten forms, so calligraphy plays an important role. Morphology and proportions cannot really be understood without a sound understanding of the handwritten origins, something many contemporary designers seem to lack or ignore. So whilst I don’t think one should imitate calligraphy in type – they are two very different forms – it is essential to have knowledge of the handwritten origins of letterforms.


– How do you decide which Arabic script style is proper to base the design on?

The intended use of the typeface should inform the stylistic choices, whether they pertain to the calligraphic style or other features of the design. Given the status of Naskh as the style for continuous reading, I would only consider different styles for very specific applications. For text typefaces Naskh strikes me as the obvious choice.


– Which Arabic script style do you think to be the best to derive a legible and readable text typeface from?

See the answer above. I get the feeling that currently some designers search for other calligraphic styles because of the novelty and variety of forms, which are certainly attractive. But I would retain that Naskh is far from being fully explored as a source for contemporary typographic forms, and more profound engagement with Naskh and its characteristics may be very fruitful.


– How much do you think that fidelity to calligraphic (script) rules is necessary in each of these factors? a) proportions b) contrast c) placement of thick and thin strokes d) using alternates e) spacing f) (base)lines

I doubt that this can be answered in a general way. It would very much depend on the specific design. For a historical design, obviously stricter adherence to the above points is important, and for some more experimental and free take, the envelope could be pushed further. Having said that, I would maintain that proportions and contrast are decisive for a readable typeface, and can therefore only be played with in a rather restrained way. Besides the readability question (readability requires adherence to convention), it is clearly an aesthetic question. Remember the famous distorted Mona Lisa by Frutiger – in my opinion many new and ‘fresh’ Arabic typefaces do exactly what Frutiger thought should be avoided: distort the canonical proportions with some rather unpleasant results (see attached image).


– In your work, are there any significant differences between the process of designing a text and a display typeface?

No, and I don’t think there should be. I never understand how some people have the expectation that a display typeface should be less work. Only because its use is different, it shouldn’t change the attention to detail, the curve quality, the consistency of design.


– Would you please send us a design brief of one of your typefaces and the final typeface specimen, and explain briefly the characteristics of the design which makes the typeface suitable for the brief requirements?

Well, Nassim was initially conceived as a new take on an Arabic/Persian typeface for news setting. This is why the BBC had chosen it over other typefaces that they tested, as Kutlu Canlioglu, then Senior Creative Director of the BBC’s Global News explained:
‘Our primary criteria in our search for a font for the sites that use Arabic script were:
– readability at body copy sizes
– a good presence at headline sizes to achieve the typographic hierarchy set out in GEL
– a contemporary, fresh aesthetics, again, in fitting with GEL’s philosophy
– and most importantly, to constitute a good base to customise for the different languages. (As we all discovered, this was much more difficult for many reasons, mostly for cultural reasons rather than typographic ones)’


– Do you think simplifying Arabic Letters (for example reducing the guidelines (kursies), simplifying the calligraphic forms of the letters and making them look more geometric, reducing contextual forms, etc.) is a good solution for today’s Arabic typography?

Typography, and therefore type design, always implies a degree of systematisation, which necessarily simplifies things. Movable type is simpler than handwriting in the sense that it is a contained system of finite elements and combinations, it is organised. The degree of this organisation, however, is what your question really is concerned with. In Arabic type, simplification generally went beyond the organisation of the script into recurring elements; it aimed and often succeeded to actually change the way the script works. Rather than finding a way to systematically reproduce all elements and features of the Arabic script, typography most of the time imposed principles of composition which required Arabic type to be different from Arabic manuscript practice – simplified. Now these are two very different kinds of simplification. In my view, Arabic type all too often was shaped according to composition principles which had been conceived for the Latin script without even asking whether this was the most appropriate way of doing things. Of course this cannot be undone today, and reading habits are hugely influential: people who have read simplified typefaces will likely prefer them to those which show more fidelity to manuscript practice. However, I think that a great deal of simplification is not necessary today, and rather betrays a design approach than anything else. To come back to your question, I would say that there is certainly a need for non-calligraphic aesthetics, but I don’t think that this necessarily means simplification. I would argue that structurally, more complexity would be beneficial for many Arabic designs, whereas stylistically there is ample room for all kinds of directions. For example, in the realm of ‘simplifications’, homogeneous alignment along an artificially imposed x-height strikes me as one of the most ill-advised things you can do with Arabic type. It suppresses distinct features of letters, making it harder to distinguish them which can’t be good for legibility. Unfortunately we can still see this being done by some designers.


– What are the advantages and disadvantages of defining 4 contextual forms for Arabic letters? How much, do you think, should a typeface simulate the calligraphic complexities (especially in system of letter shape alternation)?

The point that I’ve already tried to raise in my previous answers is that features which come from manuscript practice should not be called calligraphic complexities. They are features of the script, and many of them might actually help legibility, rather than create obstacles. The prevalent common-sense idea that more complex equals more difficult strikes me as nonsense, as already Noordzij argued in relation to the teaching of writing to children. A single-storey g is more complex to draw, but is it more complex to read? Noordzij would say no, and I think in Arabic are many more parallel cases. Tooth-height alteration is more difficult to draw and implement, but is it more difficult to read? I would say on the contrary, it helps to disambiguate letters and create unique word images, rather than some design in which all elements are equalised, creating the impression of a garden fence.


– How much, do you think, using ligatures in a typeface help simulating the dynamic characteristic of the Arabic script?

In principle ligatures are an inadequate means. For Arabic they are a crutch which originates from foundry type, used to replicate a narrow selection of Arabic script morphology. Ligatures have many limitations, not least the exponential multiplication of glyphs they necessitate. But of course there are situations in which they might improve specific letter combinations. Because of economic and time considerations most Arabic typefaces won’t implement all features of the script – which would be a daunting task – and so a few selected ligatures will often be better than none at all. Indeed, many clients would not even want a full reproduction of the Arabic script morphology because of their ideas of ‘modernity’. I suppose as always, compromises have to be made between best practice and best possible practice.


– Do you use specific tools and materials as you begin your design? Do you believe that tools play a key role in this step?

The use of specific tools will very much depend on the specific design and the desired aesthetic. The more closely aligned a typeface should be to a calligraphic style or model, the more one would need to use the tool the design aspires to reflect. Use of a specific manual writing tool like a qalam also adds idiosyncrasies, which may make a chirographic type more interesting and unique. For more synthetic design languages, however, an imaginary tool may prove more successful than an actual manual tool. It may provide flexibility and features that a writing instrument cannot provide. In either way, an understanding of the original writing tools and practises of a script are paramount for the understanding of formal features that should be reflected in type design.


– In your work and in the hand- sketch phase, do you prefer rough sketches  or precise hand drawings? do you create a highly finished drawing before getting to computer or do you think that rough sketches are enough to be scanned and worked on?

In general, I much prefer rough sketches over precise hand drawings. In my experience, the precision available with vector drawing tools and the ease and speed of manipulation make precise hand drawings – despite their aesthetic appeal – redundant for most type design work. Often I don’t even think a rough sketch needs to be scanned; for me it mainly serves to quickly develop an idea, which I then implement with the precision the computer drawing provides.


– Do you start by designing a specific letter, word or sentence? If so, please explain which letter, word or sentence and why?

For Arabic script designs, I start out with a handful of letters which will provide key dimensions and features. Commonly I would begin with the various shapes of Alif, Bah and Ayn (ten glyphs), to get the fundamental x and y axis proportions. They are all frequent letters, and the Ayn provides one of the more difficult elements: the deep curved bowl. Then I gradually add further letters in order of frequency and elements that are important for the overall design: counters, specific joins, various bowls and teeth.


– In your work, is there any sequence or order to design different characters?

Yes, I design according to my own analysis of letter frequency and key features of letterforms that need to be coordinated throughout the type. It helps me to make the design process efficient and the outcome more consistent.


– Do you advance the procedure of designing different weights and styles of a type family simultaneously or separately one after another? If you design them separately, Which style or case (uppercase, lowercase, regular or bold, etc.) do you usually start with? Why?

I tend to start with the lowercase of the regular weight as it is the most important and most used element – at least in text type. It’s here that I want to establish the main characteristics, which I then extend to the other elements and members of the family.


– In designing multilingual typefaces do you design different scripts simultaneously alongside of each other or separately one after another? Why?

A scenario in which multiple scripts are designed in parallel from the start is great, but one of the rarer situations. Most of the time an existing design gets extended with another script, and therefore it is particularly important to make the right decisions in such a case. If you have the opportunity to design multiple scripts together from scratch, I think it ideal to plan them together from the outset. Here much depends on the brief, but apart from the obvious stylistic considerations, I would think that the different proportions and use of cartesian space should receive most attention early on. Especially the vertical metrics can pose considerable challenges in multi-script designs, and planning them together can help later down the line.


– Do you follow a special technique to design the different weights for a type family?

I haven’t found a particular technique that would suit equally well to different designs yet, so most of the time it will vary from project to project. I find that higher contrast designs lend themselves better to systematic ‘techniques’ of weight increase, which are less applicable in low contrast designs, where the eye has to be the ultimate judge.


– Do you use any specific formula to perform spacing? Do you use any “key” letters for this purpose? (like H&O and n&o in Latin)

Spacing in Arabic script type works very differently than in Latin designs. Because most letters join, the key spacing glyphs are those which follow each other without joining. Thus, I am primarily looking at combinations of glyphs with letters like Alif, Reh and Waw, as they are all very frequent, differently shaped and crucial to the word images.


– Do you use any “key” letters to set the proportions of the characters? (specially in character heights)

Yes, as I mentioned above I begin by looking at Alif, Bah, Ayn and Waw to establish the principal proportions of letters and their individual features. With these I get a pretty good idea of the direction of the entire type, which I can then expand to related characters. Obviously, there will be glyphs which require entirely new solutions, but it provides a starting point.