Chronicling America
Library of Congress and National Endowment for the Humanities
Millions of digitized pages of historic American newspapers, OCR text included, free to mine.
The National Digital Newspaper Program's collection of digitized U.S. newspaper pages, contributed by state partners. Each item is a newspaper page with a scan and machine-read text, organized by title and issue date. Researchers use it to trace how words, events, and ideas spread through the press.
Access
- Cost
- FreeFree search, API, and bulk OCR downloads.
- License
- Public domainLibrary of Congress states the pages are public domain or have no known copyright restrictions
- Sign-up
- No sign-up
- How you get it
- API and Bulk downloadJSON, OCR text, JP2, PDF
What's in it
- One row is
- One record per newspaper page, grouped by title and issue
- Covers
- United States and its territories
- Years
- 18th century to 1963
- Updated
- Irregular
- Size
- More than 20 million pages
Quality
Large, well-catalogued, and open, but text is uncorrected OCR and title selection depends on which states took part.
- OCR errors are common on older, damaged pages
- Uneven coverage across states and decades
Connects to
- ProQuest Historical Newspapers: Paid archive of the major metro dailies that Chronicling America mostly lacks
- IPUMS USA Full Count Census: Population context for the towns where papers were printed
Questions it could answer
- When did a given phrase or slang term first appear in print, and where?
- How fast did news of major events spread from city to small-town papers?