Textual Criticism has the primary goal of reconstructing the original text of the New Testament by comparing and contrasting the contents of many individual manuscripts. To do this, scholars need a reliable transcription—a searchable digital file—of each manuscript to be used. Unfortunately, because the transcription process is very laborious, we do not have transcriptions—searchable digital files—for the vast majority of ancient documents.
To transcribe the nearly 19,000 words from a typical Gospel of Luke manuscript, a team of 3 must devote roughly 70 hours of labor. A critical text of a book like Luke requires no less than 200 transcribed manuscripts. That’s 14,000 hours of labor, or 2 scholars plus an editor working 3 years before the “real” work of comparing the manuscripts begins!
Imagine, instead, we had complete transcriptions of all extant New Testament manuscripts, all compiled and sitting in a database ready for analysis. Suddenly, the task of textual criticism becomes far more precise and far quicker, taking mere days or weeks instead of many years.
searchable digital files for the vast majority of ancient documents.
CSNTM proposes to train an AI model to read the text of ancient Greek manuscripts and put that text into digital, searchable form.
Over the past two decades, CSNTM has digitized hundreds of thousands of pages of New Testament Greek manuscripts and made them accessible online. Meanwhile, scholars have published digital transcriptions of 56,000 pages of New Testament Greek manuscripts, all stored in online databases. We will train AI models using the images provided by CSNTM and the transcriptions scholars have produced. The primary output of this project will be a sophisticated tool made available to New Testament scholars to study the text more fully. Your investment in this cutting-edge scholarship will help demonstrate the reliability of the New Testament in unprecedented ways.
Objective One
Assemble approximately 56,000 extant pages of New Testament manuscript transcriptions, pair them with CSNTM high-resolution manuscript images, and use the data to train an AI transcription model to render photographed ancient Greek handwriting into digital text.
Objective Two
Create an intuitive user interface hosted on csntm.org to allow the upload of high-resolution, digital images of ancient Greek manuscript pages—including non-New Testament material—for transcription.
To complete this project, CSNTM is seeking to raise $423,000.
Help PreserveNew Testament Manuscripts