The Book of the Tales of Canterbury

Mural of the Canterbury Tales created by Ezra Winter 1939, west wall of the North Reading Room in the Library of Congress John Adams Building in Washington, D.C, photo by Carol M. Highsmith, via Wikimedia Commons (public domain).

Geoffrey Chaucer

A Critical Edition

General Editors: Barbara Bordalejo and Peter Robinson

Saskatoon, 2026-.

Available in this online edition

The General Prologue. Edited by Barbara Bordalejo and Peter Robinson. (21 July 2026)

The Miller’s Tale. Edited by Barbara Bordalejo and Peter Robinson. (21 July 2026)

The Tales of the Reeve and Cook. Edited by Thomas J. Farrell. (21 July 2026)

What we have discovered

For nearly 500 years, we have been calling it the wrong name. It is The Book of the Tales of Canterbury. This is the title given in the first lines of the oldest manuscript, and in every manuscript and edition before 1500. The modern title The Canterbury Tales first appears in William Thynne's 1532 edition.
For over 150 years, editions have been dominated by just two manuscripts: the Hengwrt manuscript, now in the National Library of Wales (Peniarth 392D) and written around 1400, perhaps during Chaucer's lifetime, and the Ellesmere manuscript, now in the Huntington Library (MS 26.c.9) in San Marino, California, written shortly after Hengwrt. Our edition argues that a third manuscript, Christ Church 152, now in Christ Church College Oxford, though written around 50 years later is of equivalent authority to Hengwrt and Ellesmere.
Many previous scholars have thought that the Tales was originally copied and distributed in many parts, with complete copies only appearing later. This led scholars to think that the history of the text is so complicated that it cannot be understood. Our research demonstrates that this is the wrong way around. The first copies were complete copies of the Tales as it was then available to scribes. This greatly simplifies our understanding of the textual history of the Tales
Our work has broken new ground in its application of new technologies to complex textual problems. We use bioinformatics (evolutionary biology) software to discover family groupings among the manuscripts. We use a custom-built search engine (VBase) to ask highly-specific questions about the distribution of readings across the tradition, for example how many and which readings are shared by a particular set of manuscripts. We offer ways to explore the manuscripts, in our transcription and in thousands of page images, which no print edition could ever do.
Our edition presents a new text of the Tales, closely based on three key manuscripts (Christ Church, Hengwrt, Ellesmere). All three manuscripts show extraordinary agreement and care in fine details of wording and spelling. Basing our edition on these wordings and spellings gives us a Tales which is more dramatic, more varied, and closer to the rhythms of everyday speech. Reading the manuscripts this way brings the text to life. We wish our readers the same experience.

In the last decades, many digital editions have appeared and disappeared. Indeed, our earlier publications on CD-ROM and DVD met this fate within just a few years of publication: thus Robinson's Wife of Bath's Prologue on CD-ROM and Bordalejo's Caxton's Canterbury Tales: the British Library Copies, and others. We have made this online edition in such a way (following the suggestions of the Endings Project) that we hope that this edition will be available for many years to come.