Large Language Models and Ottoman Court Records: Long-Term Change and Regional Comparison

Metin Cosgel ()

No 2026-04, Working papers from University of Connecticut, Department of Economics

Abstract: Large language models offer new ways to turn historical texts into quantitative evidence, but their value depends on whether the resulting data support reliable comparisons. I examine that potential through Ottoman court registers, a source for studying long-term change and regional variation after regular tax surveys ceased around 1600. I combine rule-based extraction with large language models to convert 385 transcribed volumes from twelve districts into a dataset of 160,395 entries. Rules extract formulaic information, while models classify documents, identify participants’ roles, and interpret legal outcomes. Validation uses hand-coded samples and comparisons with three published Istanbul studies of intergenerational persistence of social standing, foreign-coin use, and the decisive oath in litigation. The new extraction recovers their principal patterns and adds later observations to the standing and oath series. Applying common definitions across districts reveals different relationships between the capital and the provinces. The oath’s early seventeenth-century peak and subsequent decline appear broadly across the observed courts. Foreign silver accounted for substantial shares of recorded transaction value earlier in Antep, Amasya, and Kayseri than in Istanbul and several western and central districts. The early seventeenth-century decline in father–son status persistence in Istanbul was shared by Bursa but was not a general provincial pattern. Uneven transcription coverage, selection into court, and extraction error limit these comparisons. Within those limits, the method makes published court records usable for comparative quantitative history across several centuries and regions.

Keywords: Ottoman Empire; court records; sicil; tahrir; quantitative history; long-run change; regional comparison; large language models (search for similar items in EconPapers)
JEL-codes: C81 K40 N01 N35 N45 N95 (search for similar items in EconPapers)
Pages: 54 pages
Date: 2026-11
References: Add references at CitEc
Citations:

Downloads: (external link)
https://media.economics.uconn.edu/working/2026-04.pdf Full text (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:uct:uconnp:2026-04

Access Statistics for this paper

More papers in Working papers from University of Connecticut, Department of Economics University of Connecticut 365 Fairfield Way, Unit 1063 Storrs, CT 06269-1063. Contact information at EDIRC.
Bibliographic data for series maintained by Mark McConnel ().