Ottoman Turkish · Image to Text
Six centuries of Ottoman paper, script by script
Ottoman Turkish is a Turkic language written in an Arabic script built for a different sound system, with a vocabulary drawn from Arabic and Persian and a chancery that invented distinct hands for distinct kinds of business. Evidano reads each of those hands on its own terms, from a scribe’s book copy to a treasury clerk’s shorthand.
Choose the hand or document type
Early Ottoman Naskh and Taʿliq Manuscriptsc. 1350–1500Transcribe early Ottoman Turkish in naskh and taʿlik: missing dots, Persian pe and gaf, Arabic quotations and the old spellings of the first Ottoman centuries.
Classical Ottoman Naskh and Nastaʿliq Manuscriptsc. 1500–1800Transcribe Ottoman naskh and nastaʿliq: sloping baselines, izafet chains, marginal glosses and Arabic and Persian passages inside a Turkish text.
Ottoman Dīvānī Imperial and Diplomatic Documentsc. 1500–1900Transcribe fermans, berats and ahidnames in dīvānī: rising interlocked lines, filler dots, tughras, pençe signatures and decade dates.
Ottoman Siyāqat Fiscal and Land Registersc. 1500–1875Transcribe siyāqat accounts and land records: dotless shorthand words, treasury numerals written out in Arabic, column headings and marginal sums.
Printed Ottoman Turkish and Ruqʿah Manuscript Materialc. 1729–1928Transcribe Ottoman books, gazettes and handwritten ruqʿah from the Müteferrika press to the 1928 alphabet reform, with dates in three calendars.
Transcribing Ottoman Turkish across the centuries
The surviving record splits by function rather than by century. Books were copied in naskh and, increasingly, in the sloping nastaʿliq borrowed from Persian. Imperial orders left the chancery in dīvānī, whose rising, interlocked lines and scattered filler dots were designed to make interpolation obvious and forgery difficult. The treasury kept its accounts in siyāqat, a near-shorthand that strips the dots from words and writes its numbers out as stylised Arabic words. Correspondence and, later, the printed press used the plainer ruqʿah.
Two features make the whole corpus harder than its Arabic or Persian neighbours. Turkish vowels have to be carried by Arabic consonantal letters, so a written word can support several readings and the right one depends on knowing the language rather than the letters. And the vocabulary shifts register constantly: a Turkish sentence can contain an Arabic legal formula and a Persian genitive chain, each following its own grammar, inside a single line.
The people working with this material are Ottomanists and social historians reading court registers and petitions, economic historians working through tahrir and timar registers, diplomatic historians comparing the Turkish text of a treaty with its European counterpart, and family researchers across the former empire — from the Balkans to the Levant — whose ancestors appear in a register written in a script no living relative can read.
Not sure which hand you have?
Upload a page, pick the closest domain, and test the prompt on one page before running the rest. You can switch domains at any time.