Ottoman Empire 18th-19th C, Probate Inventories

Background

Associate Professor of History, Ali Yaycioglu has gathered probate inventories of wealthy and powerful individuals who died during the mid-century revolutions from the Balkans to Egypt. The documents record properties, debts and credits linking people to each other. Individuals had debt relationships with entire villages or communities which resulted in complicated negotiations. Debt could be pardoned or sold to another private party. In this period, people were like firms, so the death of an individual can trigger a whole chain of events that can be traced through these documents.

This is just one part of a trove of similar earlier cadastral records that cover provinces or regions (rather than individuals). When land was conquered, the conquerors send scribes to record what they conquered.

(See video interview above with Prof. Yaycioglu for details about the significance of these documents and his research interests.)

Research

This is the kind of material that could be mined extensively and would be of interest to many fields of study. Each inventory holds rich detail about the material culture of the time. Taken as a whole, one can piece together a network of influences that can reveal how political and economic elites are connected: muslims and non-muslims, men and women, politicians and merchants, etc.

Some of the questions that might be asked: It was a time of great inflation, can we track the price changes over time? How were these probate inventories prepared? How were objects/real estate listed, appraised? How do they calculate value?

Data

Scanned document spreads at 400ppi, about 3000 x 2500 pixel images. The codex includes the inventories of about 150 people. Each inventory ranges between 5-30 pages. There are not many pages, but the pages are dense. We have a number of pages that have already been transliterated in Latin script.

Desired output

1. Identify the scribes across the documents. These are manuscript documents, each one of which includes several different hands. The scribes each have a distinct hand. Even without transliteration, it would be valuable to know which scribes contributed to which documents.

2. Transliteration. The scribes wrote Turkish in an Arabic shorthand script. Very few people can read these documents.

3. Entity extraction. Across the 150 inventories, thousands of individuals are mentioned, placenames, objects, etc. We will then want to associate prices with inventoried items.

Experiments

The first step will be to segment the documents into the distinguishable areas of text. We will label the documents (identify the different sections that either have different hands or meaningful content breaks) for which we already have transliterations to train layout analysis and line detection.