Urdu’s Last Calligraphers?

June 6, 2007

Bill Poser pointed to a four month old piece about an Urdu newspaper in Madras:

The age of calligraphy died when British soldiers toppled the Mughal courts. It’s hard to remember that there was a time before the age of computers when penmanship was considered one of the highest art forms. Outside of a some particularly ornate wedding invitations and hand-written copies of the Koran there is little need for formally trained Urdu calligraphers. That is, except for one small ink-stained corner of Chennai where the world’s last hand written newspaper still churns out 20,000 broad sheets a day.

This is just flat-out wrong. The vast majority of “printed” Urdu works are hand written. That includes books, newspapers, magazines, posters, and so on. It’s true that there is at the moment an incredibly fast pace of change towards computerization amongst the major newspapers, and a slower pace for books, but this has been only over the last five years. In fact, one of the only things which are almost always computer composed are wedding invitations, because it’s cost effective for these small batches, and because the consumer bears the higher labor costs of DTP vs. caligraphy. If that sound backwards, think about this: hand copying of manuscripts at the Salar Jung library is cheaper per page than photocopies from those archives which will let you copy (which SJ won’t.)

As for the Qur’an, a trained Urdu calligrapher would be completely unqualified for that. Qur’an’s are written by teams of huffaz (people who’ve memorized the Qur’an) trained in Qur’anic calligraphy. They are repeatedly checked and certified by government bodies in multiple countries. Generally speaking, a publisher who prints a Qur’an with an error of a single diacritical mark can expect to be held responsible for recall and destruction of the entire issue. 95% of the Qur’ans you’ll run across in South Asia are reprints (mostly unauthorized) of the Taj Company edition of the 1930s. These are resized and recut to fit different editions. I watched someone spend two years cutting and pasting by hand to produce a new, large print multi-lingual interlinear edition which he hoped would become a marketplace sensation. Essentially, there are two visual styles in Qur’ans today, the Taj version, and that handed out by the Saudis. Everything else is a rarity. Many South Asians have trouble reading the Saudi version as it lacks visual cues they grew up with.

Going back to Urdu, to understand the slow adoption of computer typesetting, we need a bit of historical perspective. First off, Urdu had hardly begun life as a respectable language for prose literature and written communication “when British soldiers toppled the Mughal courts.” It was actually the British who made “Hindustani” the language of administration and who were the first promoters of printed Urdu. Publishing of Arabic and to a lesser extent Persian began in Europe, and by the beginning of the 19th century, there were presses in Cairo and elsewhere and the reading public slowly became accustomed to the ‘flatness’ of block type. (Remember that all Arabic-derived scripts except Divehi are scripts, and thus real – wood or metal – type will always mean a mutilation.)
The earliest Urdu – and Bengali, Hindi, Telugu, etc. – books to be printed were printed at Fort William College. These are in a flat naskh style similar to printed Persian of the 19th century. I have a nice photocopy of a book printed in this style. From a modern perspective, it takes time to realize you are looking at Urdu rather than Farsi. The flatness of naskh is used as a visual clue even in handwritten Urdu books of a switch to Farsi or Arabic.
Shortly after this, though, cheap lithographic offset printing arrived. This significantly lowered the barrier to entry. There were far more skilled calligraphers available than there were typesetters – just as with computer operators today. More generally, labor intensive processes have always been more economical in India than technology intensive ones. The book market – and with it the reading public – exploded. This was the era that gave us the insipid Taqwiyat ul-Iman, and which gave birth to every major trend and controversy in South Asian Islam and Muslim society.
The style of handwriting used in these and all subsequent works was (a version of) nasta`liq, which grew out of short hand styles used in courts and government offices. While it was once one Persian style of handwriting amongst others, nasta`liq is now totally identified with Urdu. The style is very compressed, with letters stacked on top of each other. Any combination of two or more successive characters will require a particular shaped ligature and proportion – many more than are available in Unicode’s Arabic Extended section.
Lithography was easier and cheaper than block type, and there was no particular reason to look back. That is, until the invention of computers and the various stages of mechanical and computerized typesetting. Still, developing the algorithm for proper rendering of nasta`liq was just not attracting the necessary investment. One factor may have been the fact that those kids from Urdu speaking families who went into computers in India – and even, to some extent Pakistan – were those least likely to be able to read Urdu. Even now, if you are Muslim and go to an English medium school, you take Arabic as your second language, and if you go to an Urdu medium school, the chances of you pursuing engineering are slim.
Various attempts at Urdu word processing started showing up and multiplied in the ’90s. But the people who finally did get electronic nasta`liq right were Concept Software of Delhi, with their desktop publishing package InPage. They have a complete and total monopoly. If you have seen an Urdu wedding invitation or a typed booklet or newspaper, or looked at any of the major newspapers’ websites, you have seen their font. If you walk into the Urdu printing neighborhood in any South Asian city, or the area around the courts or other government offices, and you see signs in the typing or invitation shops saying “Urdu/Arabic DTP”, there will be a Windows XP or 2000 box inside and some guy under 30. (Or maybe they’ll call him and he’ll show up on a Yamaha in a few minutes.)
The problem for Concept (and for Microsoft) is that none of the software on that machine will be licensed copies. Concept even went to the extreme step of requiring a physical usb dongle to activate their software. It didn’t help. (If you’re in the US and want to use a legitimate version of the software – think ahead. You won’t need the dongle, but every time you install or re-install it you will have to go through a multi-stage reactivation process which requires personal communication with the developers from the email address used when purchasing the software. A huge pain if you’re buying it for multiple users.)
The problem for users is that InPage does not use – and could not achieve the effect which has made them successful – fonts or encoding recognized by any software except their own. This means it can not be used as plain text for all of the things for which plain text is perfect. It can’t be indexed or shared or manipulated in any of the ways the web relies upon. Go back to those newspaper sites and you’ll see that all of the Urdu is in image files. That means I can’t google a news story – even though Google deals with Urdu fine. That means that if you ever want to research anything you’ll have to go through all of the past articles by hand as if computers had never existed.
That brings us to Unicode. The first implementations of Urdu in Unicode fonts were abysmal. Slowly we have seen improvement. There’s an upper limit to the performance Unicode can give us for nasta`liq (in the number of available glyphs) but the basic standard has now been set by CRULP’s fonts, and by those used by the BBC. BBC Urdu is the only major Unicode driven Urdu site on the web. Perhaps they’re the only ones who understand the importance.
And what about the print world? Even though most of the big papers now have an InPage fueled web presence for the NRI set, the printed versions are still done in the same way as the Musulman. Munsif was the first paper I saw to use computer typeseting. And that’s probably precisely because of how new it is. It’s easy to set up a desk and a few operators to type up all of the stories. It’s another thing altogether to retool your entire production process at a time of heavy financial pressures.
Most readers consider InPage nasta`liq barely adequate, and Unicode fonts only acceptable in emergencies. i know this because I’ve been scolded for producing unreadable materials when using CRULP’s fonts. I have yet to see a book over about 200 pages which was typed, except for one (a biography) which was written in America, typed in Canada, printed in Lahore, and distributed in Hyderabad, india. Generally it is only short booklets, mostly what we would call “vanity printed”, which are done on DTP. (You must realize that most of the non-English publishing in India is done at the expense of the author or the author’s friends or students or disciples. This actually leads to very vibrant literary scenes about which people actually care quite a bit.) I think most people would find a long book unpleasant to read if it was not hand calligraphed.
I would be interested to hear about which papers do their print editions in InPage these days. About half of the weeklys in New York, Chicago, and Toronto seem to still be hand written. (I remember one summer seeing ads in the New York Times classifieds for Urdu/Punjabi calligraphers.)
As a last note, this is not entirely exclusive to India. The first time I saw a copy of Mawlid Barzanji from Yemen, I assumed it was from India, because it was hand-calligraphed in a very nasta`liq-like font. In North and West Africa they use a script which differs significantly from standard printed Arabic, and along 116th Street you can find dozens of lithographed mawlids, lives of the awlia, and other tracts in this script.

(I just noticed that Scott Carney is one of the guys who writes Bodyhack. I’m sure he knows his stuff – and the rest of his blog is great, he just seems to have been given some wrong info on Urdu. Also, if I’m reading his bio right, then it would seem our paths have crossed at least once.)

UPDATE: I seem to have unintentionally libeled poor Unicode. I love Unicode, and I wish every Urdu site on the internet was using it. But even the most beautiful Unicode fonts aren’t satisfactory for the public at large – who don’t care about the things which make me love plain text, and don’t mind if everything is in JPEGs. THE FAULT IS NOT UNICODE’S. It’s in what can practically be done with Urdu Unicode in currently available browsers on contemporary operating systems. With a bigger user base, and more active and interested developers we could probably have perfectly beautiful solutions on every platform and browser. But it’s not going to happen. While the number of Urdu-as-preferred-language internet users may have slightly grown, it will plateau in the near future. The number of developers and ‘power users’ who care particularly about an accessible archive of Urdu literature (in the broadest sense), and have the knowledge of both Urdu and computing to pull it off is minuscule and not going anywhere. Urdu is not Arabic or Farsi. Maybe we will see a surge of interest which will change all of that, but I am not optimistic.

UPDATE 2: Here is a January article on somewhat the same subject. They pointed out something I wished I had mentioned before – the difference between khattati or khush-nawisi and kitabat, i.e. calligraphy and lettering. There are lots of institutions offering training in Urdu DTP these days. Last time I was at the Idara Adabiyyat-e Urdu the calligraphy classes were more popular, but no one in those classes expected a job, it was a hobby for them. Most of the people in the DTP classes had already taken some sort of Computer Applications training, as well as typing classes. They were definitely all looking to add an employable skill. So this is definitely the future, for better or worse. One difference we may see is that, while lots of women have learned calligraphy, few were employed as calligraphers. You will see many more women employed in computer typesetting of Urdu. I don’t think the situation will be quite analagous to that in Iran – there are two many varying factors – but it will be (may already be?) a serious option for women from traditional families with an Urdu education.


One Response to “Urdu’s Last Calligraphers?”

  1. Helen DeWitt Says:

    This is fabulous. I especially love the idea of a vibrant literary scene based on self-published work.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: