Folio 03 · about 4 minutes

The Scripts of Trade

Most bahis were not written in the Devanagari of printed books but in mercantile scripts such as Mahajani and Landa. These dropped vowels and spaces and were hard for outsiders to read. Whether that was by design or by habit is still an open question.

What is Mahajani script?

Mahajani is a Brahmi-based script that was in common use across northern India until the middle of the twentieth century. It was a specialised commercial script for accounts and financial records, used to write Hindi, Marwari and Punjabi.¹ Its name comes from महाजन (mahājan), a banker or moneylender, because bankers and moneylenders were its main users. It belongs to a family of accounting scripts that includes Sarrafi, the “banker's script”, and Kothival, the “merchant's”.¹ Most surviving Mahajani documents are bahi-khatas. The rest are merchant diaries (roznāmā), letters (chiṭṭhī) and hundis.¹

The word “Mahajani” could also mean a style rather than a script. Pandey notes that it was applied to a merchant's form of Dogra, to the Gujarati script, and to a form of Devanagari used to write Marwari and other languages of Rajasthan.²

What made these scripts “clipped”?

Mahajani is structurally close to the Landa scripts of Punjab and Sind. Laṇḍā means “tailless” or “clipped”. The name describes their shared habits: a reduced set of letters, no vowel signs and little attention to spaces between words.² In Mahajani, Pandey explains, a single letter can stand for ka, kā, ki, kī or ke, and the reader has to infer which from context.³ Nasal sounds were often simply dropped. The word हुंडी (huṇḍī) could be written with just two letters, h-ḍ.⁴ Ledgers used ruled lines to title and structure a page, signs for fractions of the rupee, and a tick-like mark over figures whose exact use is still not understood.⁵

For a writer, the result was speed: fewer strokes per word and figures that stood out. For anyone outside the trade, a page could be close to unreadable.

Were merchant scripts meant to be secret?

This is where tradition and evidence part ways. The best-known statement of the secrecy view is a Victorian linguist's guess. In 1872 John Beames wrote that the Mahajani character was “quite unintelligible to any but commercial men”, and went on:

“Perhaps this is intentional. Secresy has always been an important consideration with native merchants, and it is probable that they purposely made their peculiar alphabet as unlike anything else as possible, in order that they alone might have the key to it.”

Note Beames's own hedges: “perhaps” and “probable”. The structural explanation, that a clipped script is quick to write and good enough for people who already know the names and sums involved, fits the evidence equally well. The two are not exclusive. A script built for speed among insiders would also keep outsiders out, whether or not anyone intended it. What can be said with confidence is that the ledgers were legible to the community that kept them and opaque to most others, including, later, to colonial courts. That matters for the hundi chapter.

How did merchants learn to write the books?

Mahajani was taught as a medium of education in Punjab, Rajasthan, Uttar Pradesh, Bihar and Madhya Pradesh, in schools called catashalas. Boys from merchant and trading families learned the script there along with the other writing skills a business needed.¹ Printed primers survive. In 1875 Śrīlāla published a Mahājanī-sār at Allahabad. A Delhi primer by Lala Gangadas taught the script together with accounting and letter writing, and included instructions for writing a hundi.⁶ Colonial scholars collected specimens: Samuel Kellogg in 1876, G. W. Leitner in 1882 and George Grierson in 1899.¹

What survives of these scripts?

Very little in daily use. In 2004 The Tribune of Chandigarh reported that “Langdi Hindi”, apparently a Haryanvi form used for bookkeeping and written in Mahajani, was nearly extinct.⁷ The scripts have, however, entered the digital record. Mahajani, Modi, Khojki and Khudawadi were all encoded in Unicode version 7.0, Multani in 8.0 and Kaithi in 5.2, all from proposals written by Anshuman Pandey for the Script Encoding Initiative.⁸ That matters for one practical reason. Pandey called Mahajani a key to the historical financial records of northern India.⁹ Records that can be typed can be transcribed, searched and studied.

A note on images. This chapter would be better with photographs of real Mahajani ledger pages. None are reproduced here because their copyright and licensing could not be confirmed in time. Plates in the nineteenth-century primers and in Leitner's 1882 collection are out of copyright and are good candidates for a later edition.