Two hundred years of "it cannot be done"

Nastaliq is Urdu's written voice — the slanted, hanging calligraphic style that readers of poetry, newspapers and religious literature have expected for centuries. It is also, by broad agreement, the hardest Arabic-script style to mechanise. Its letters stack diagonally instead of sitting on a baseline, its ligatures cascade, and no two combinations of the same letters look quite alike.

From the metal types of the 1781 Forms of Herkern through the Linotype experiments of the 1960s, every machine attempt had to simplify Nastaliq into something it was not. The typewriter answered with Naskh. Hot-metal linecasting got a simplified "Tamannai" cut. Nothing preserved the cascade.

The family that refused to simplify

The breakthrough came from a calligrapher's household. Mirza Ahmed Jamil — son of the master calligrapher Noor Ahmed Mirchi , after whom the typeface is named — had watched his father's hand set the standard for Lahore's publishing houses. Working with Matlubul Hasan Syed and the British foundry Monotype , the family bet on a radical idea: instead of teaching a machine to draw Nastaliq, they would hand-draw essentially all of it .

The result, digitised for Monotype's Lasercomp imagesetter, was Noori Nastaliq : a library of more than 20,000 individually designed ligatures , with development running through 1979–81. The National Language Authority announced the achievement on 6 December 1980 ; the commercial launch followed in 1981 , and in 1982 it was recognised as an Invention of National Importance.

When we say "hand-drawn", we mean it literally: the ligature set follows the hand of Noor Ahmed Mirchi, whose contribution is chronically under-recognised in the retellings that credit only his son. The elder calligrapher's hand is inside every line of Noori ever since.

The newspaper decade

Daily Jang put Noori to work for what its own tribute pages call "computerised laser composing" in the early 1980s, and Urdu newspaper production was never the same. For roughly two decades, Noori Nastaliq was the production Nastaliq of Pakistan's press — the typeface in which the news literally happened.

There is a myth worth puncturing here: the widely repeated claim that Jang computerised Nastaliq in 1987 could not be verified in any primary source when we checked. Jang's own material points to the early 1980s, consistent with the font's 1981 creation. History gets rewritten one round number at a time.

The compromise that made it work (and its price)

Noori's ligature strategy preserved calligraphic fidelity on 1980s hardware — but it came with structural costs that still shape Urdu computing today:

• Unforeseen words could fail. A ligature library only contains what was drawn; unusual compounds had to be composed from pieces. • No Unicode text flow. Noori lived in proprietary encodings, tied to dedicated composition systems. • The file problem. The approach scales badly: its spiritual descendant Faiz Lahori Nastaliq carries 44,920 glyphs at 37 MB .

When the imagesetter era ended, Concept Software's InPage (1994) licensed Noori and carried it into the desktop-publishing age — which is why generations of Urdu publishers still associate the name with InPage itself. Monotype later proved at ATypI 2005 that OpenType could revive the design; we document that Noori Nastaliq OpenType revival as its own chapter.

Where to see it now

Noori Nastaliq ships inside InPage rather than as a standalone download — its licensing story is part of its history. The cascade it perfected lives on in the fonts it made possible: Jameel Noori Nastaleeq, the internet era's unofficial national typeface, and Jameel Noori Kasheeda, its elongated headline companion.

The full, source-cited record — every date, every claim, every dispute — is open to read in our Nastaliq Research Archive. If you want the engineering sequel (how the industry escaped the ligature trap), continue with Ligature libraries vs character cascades.

Every claim in this article is drawn from the rtlfonts.com Nastaliq Research Archive (database v1.4.0), where each fact carries its graded sources.