ABOUT
Why damaged & distinctive types?
Recent estimates suggest that the printers of more than half of all books, pamphlets, and broadsides printed in Restoration London remain unknown. The CDT is a resource to help identify unknown printers. “Type,” Harry Carter once remarked, “is something you can pick up and hold in your hand.” Made of a pliant lead alloy, letterpress letters were often damaged during presswork, broken or bent by the heavy action of the press or spoiled by debris. Printers kept these disfigured letters in use while damages and idiosyncrasies accrued, providing a useful signal for contemporary scholars. Since no two pieces of type will degrade in precisely the same way, damaged and distinctive type offers a kind of typographical fingerprint that can reveal the identities of the individuals(s) responsible for a book’s making.
(Known) Limitations in CDT Data
While the CDT is a first-of-its-kind resource for the analysis of distinctive and damaged types at scale, it is not comprehensive for its period of coverage (1660-1700). Damaged type data for many printers remains limited for a variety of reasons. In many cases, a printer’s output survives only through a handful of editions, some of which may survive in only a few copies, few or none of which have been digitized. Moreover, many printers undoubtedly owned more fonts than appear here. What’s included in CDT in most cases simply reflects what’s been digitized and is publicly available. The CDT is also limited to uppercase, non-ligature characters. Extending beyond that set would impair our intention to create a tractable and searchable body of evidence.
Still, the CDT represents a carefully curated set of more than 15,000 characters extracted from digital copies of more than 1900 editions printed between 1660 and 1700.
A Note on Printers
Some Restoration printers worked exclusively in collaboration with a second printer. The respective types of these printers — if they had separate type cases at all — are impossible to reliably assign or differentiate. In these cases, the names of both collaborators are given as a single entry in the CDT. An index of printers included in the CDT is linked here.
Print & Probability
The type impressions gathered on the CDT were identified using machine learning and computer vision tools developed by the Print & Probability project — an interdisciplinary team of literary historians, bibliographers, librarians, and computer scientists at Carnegie Mellon University and the University of California, San Diego.
To learn more about the Print & Probability project and our methods — what we call “computational bibliography” — see our 2021 lecture at the Grolier Club or the list of publications available at https://printprobability.org/.
The CDT Team
Editor
Christopher N. Warren, Carnegie Mellon University, Department of English
Editorial & Technical Team
Taylor Berg-Kirkpatrick, UC San Diego, Computer Science & Engineering
Laura DeLuca, Carnegie Mellon University, Department of English
Baron Glanvill, Carnegie Mellon University, Department of English
Kartik Goyal, Georgia Tech, College of Computing
John Ladd, Washington & Jefferson College
Sam Lemley, Carnegie Mellon University Libraries
DJ Schuldt, Burke Library, Hamilton College
Kari Thomas, Carnegie Mellon University, Department of History
Nikolai Vogler, UC San Diego, Computer Science & Engineering
Henry Pham, Bachelor of Science in Information Technology, Software Development
Jonathan Armoza
Deployment & Publishing
Jonathan Kiritharan, Carnegie Mellon University Libraries
Sam Lemley, Carnegie Mellon University Libraries
Talia Perry, Carnegie Mellon University Libraries
Henry Pham, Bachelor of Science in Information Technology, Software Development
The CDT gratefully acknowledges support from the National Endowment for the Humanities.
