A bazaar without a z
Hindi and Gujarati borrowed the same Persian and Arabic words, and a Hindi speaker will recognize nearly all of them. But Gujarati spelling never took the letters that came with the words, so the bazaar lost its z, saaf lost its f on paper, and paper itself came out with a letter Hindi does not have. How the words arrived, what happened to them at the door, and where the trail runs out.
Walk into a market in Ahmedabad and the word over the gate is બજાર bajaar. A Hindi speaker hears बाज़ार bazaar with something missing. The z is gone, not by accident or by accent: Gujarati spelling has no z of its own, and never had one. The same thing happens to ज़िंदगी zindagi, which is જિંદગી jindagi, to ज़मीन zameen, which is જમીન jameen, and to हज़ार hazaar, which is હજાર hajaar. Same words, same Persian source. Only the clothes differ, and the clothes are the story.
Persian reached Gujarat the way it reached Delhi, as the language of the office. Delhi’s armies took Gujarat around the turn of the fourteenth century, and in the early fifteenth the province’s governor made himself an independent sultan. The Gujarat Sultanate ruled from Ahmedabad for more than a century and a half, until the Mughals annexed the province in the early 1570s. Under both, Persian was the language of the court and the revenue office, and Arabic rode inside it, since Persian had long since filled itself with Arabic words for law, trade and religion. The office left its words behind: સરકાર sarkaar, government, ખર્ચ kharch, expense, જમીન jameen, land, and કાગળ kaagal, paper. After the Mughals the same words kept arriving through Hindustani, which is where much of the modern layer comes from.
The second road was the sea. Khambhat, Bharuch and later Surat traded with Arabia and the Gulf for centuries before any sultan, and Gujarat’s own communities carried the languages of that trade home. The Parsis, Zoroastrians whose ancestors came from Iran at a date that is genuinely disputed, kept an Iranian vocabulary in their religious life and spoke Gujarati for everything else. The Bohras, Khojas and Memons were Gujarati- and Kutchi-speaking trading communities with Arabic and Persian in their religious and commercial lives; the Dawoodi Bohras still write an Arabic-laden Gujarati in the Arabic script. Parsi writers and printers were prominent in the early Gujarati press in Bombay, and Parsi Gujarati has Persian words of its own. But what these communities added went mostly into their own registers, of prayer, law and trade; the everyday layer that Gujarati shares with Hindi came through the office.
The hedge is the general case: for most loans the route cannot be recovered. ખર્ચ kharch was current in Persian offices in Ahmedabad and Delhi alike, in the Hindustani of the Bombay docks and in Marathi next door, and a word that entered by any of those doors ends up spelled the same way. What can be said with confidence is the source: kharch is Arabic kharj, expenditure, dressed in Persian. So the entries in this app name the source and add a route only when one is known.
Now the sounds. The loans brought five consonants that Sanskrit had lacked: the z of zamīn, the f of ṣāf, the fricative kh of khush, the gh of gharīb and the deep q of qalam. Hindi adopted them, at least on the page. The nuqta, a dot under ज, फ, ख, ग and क, turns them into ज़, फ़, ख़, ग़ and क़, and a careful Hindi speaker pronounces the difference. Gujarati never did; each foreign consonant folded into the nearest native letter. The z became જ, so ज़रूर zaroor is જરૂર jaroor. The fricative kh became the plain aspirated ખ, so ख़ुशी khushi is ખુશી khushi. The gh became ગ, so ग़रीब gareeb is ગરીબ gareeb. The q became ક, so क़लम qalam is કલમ kalam.
The f deserves a closer look. ફ is the aspirated p of ફળ phal, fruit, and ફૂલ phool, flower, so on paper साफ़ saaf is સાફ saaf only by courtesy: the letter is properly a puff of breath where Hindi has a fricative. Most Gujarati speakers today say an f in these words, and in native words too, but the alphabet has one letter for both sounds and cannot record which was meant. Hindi’s dot can. Many Hindi speakers also say jindagi and saaph, and much ordinary Hindi writing drops the dot. The difference is that Hindi’s standard keeps the option open and Gujarati’s never had it.
One word went the other way. कागज़ kaagaz became કાગળ kaagal, and ળ is not a sound Gujarati lacked but one it has and Hindi does not, the retroflex l of કાળો kaalo and ફળ phal. Yet it is not the inherited ળ of those words, which answers an old Sanskrit l; here it answers a Persian z, Marathi has कागद kaagad, and how the word landed on ળ is not settled. The loans also obeyed Gujarati’s own gender pattern, Hindi masculine ā to Gujarati o: दरवाज़ा darwaaza is દરવાજો darvaajo, फ़ायदा faayda is ફાયદો faaydo. Nor is the folding perfectly regular. ज़हर zahar, poison, is ઝેર jher, with the aspirated letter rather than the plain one, and an English z goes to ઝ as a matter of course: ઝીરો jhiro for zero.
Side by side, most of the layer matches. किताब kitaab is કિતાબ kitaab, though a Gujarati speaker reaching for the everyday word more often says ચોપડી chopdi or પુસ્તક pustak. दोस्त dost is દોસ્ત dost. रोज़ roz is રોજ roj. Count in either language and you change layers between a hundred and a thousand: सौ sau and સો so are inherited from Sanskrit śata, while हज़ार hazaar and હજાર hajaar are Persian. This kind of sharing is real, but it is not kinship. Two neighbors took the same word from the same lender, which is a different thing from the inherited words that make the two languages sisters, the argument Sisters, not strangers makes in full.
So the bazaar lost nothing on its way into Gujarat except a sound. The word arrived with its meaning intact and has stood over Gujarati shopfronts for centuries. The z was stopped at the edge of the alphabet, and it is still waiting there, on the Hindi side.
