The Canon of John Lydgate Project

The Canon of John Lydgate: Project Descriptions and Reports

John Lydgate's life overlapped with that of Geoffrey Chaucer, whom he repeatedly refers to as his master and model: Chaucer died around 1400, by which time Lydgate is thought to have been about 30 years old, and Lydgate is thought to have survived until about 1450. We do not know whether Lydgate ever met Chaucer personally: it is possible that he did, and he certainly knew Chaucer's son Thomas, who became a very wealthy and influential man in the early part of the fifteenth century.

Lydgate was by far the most prolific of all poets in the history of English literature. While much of his poetic output is in imitation, or even continuation, of works by Chaucer, another large segment of it is of an explicitly religious nature (reflecting Lydgate's vocation, for he spent his life as a monk of the Abbey of Bury St. Edmunds): lives of saints, exempla in verse, and religious allegories. A third part of his poetic output, besides the religious verse and the Chaucerian imitations, is of more "occasional" nature: though a monk, Lydgate maintained relations with influential men (such as Thomas Chaucer) and, indeed, with members of court and even of the royal family, and he received support and patronage from many of the highest in the realm. A significant part of his poetic output, then, could be considered a sort of "laureate" work.

All told, Lydgate is considered the author of about 145,000 lines of verse as well as of one prose tract: according to the most recent list of the works in the Lydgate canon, that published by Alain Renoir and David Benson in the revision of John Wells' Manual of the Writings in Middle English, current opinion ascribes to Lydgate some 195 works (there are, in fact, exactly 200 titles in Renoir and Benson's list, but several of these are merely different versions of single works). Renoir and Benson's list, though, was not based upon a new reconsideration of the evidence for the canon; rather, it differs only in minor details from the list offered in 1911 by Henry MacCracken. Even at the time of MacCracken's essay, however, there were protests from other Lydgate scholars, such as Eleanor Hammond, that MacCracken could not be relied upon: nevertheless, his influence continues, not so much because scholars are happy with his list (there are regularly calls for a new review of the canon), but because a reconsideration of the canon is such a huge undertaking that no one to this point has been willing to take it on.

The vastness of the enterprise is partly a function of Lydgate's prolificity and partly a function of his public visibility: besides his genuine works, a great many other poems came over time to be attached to Lydgate's name. In re-examining the canon, then, I am considering not only at the 200 titles in Renoir and Benson, but I am reopening a number of old debates to reassess the evidence for every work which has ever been considered Lydgate's. Thus, the database in which I am now collecting notes on the arguments for and against ascriptions to Lydgate has 396 separate titles: one could say that the reassessment of the Lydgate canon will involve a significant review of the whole of early fifteenth-century poetry, and the building of a corpus of fifteenth-century poetry will be an important by-product of the project.

I have, in fact, several databases in which I am presently collecting the materials gathered in the first stage of what I have conceived as a four-stage process. The first stage is intended to collect and reassess the "external" evidence for Lydgate's works: that is, the evidence for authorship to be found outside the poems themselves. There are two parts to this process, the first of which is the reassessment of the manuscript evidence. There are extant several manuscripts compiled by John Shirley, a fifteenth-century bookseller who had connections with the family of Thomas Chaucer and who knew Lydgate personally, and Shirley's manuscripts name Lydgate as the author of quite a number of works. With certain exceptions (notably one manuscript compiled late in Shirley's life which is considered by many to be unreliable) Shirley's ascriptions carry a lot of weight. Further, there are other ascriptions in other manuscripts of greater and lesser authority. All told, I have on the computer descriptions of more than 450 early manuscripts which contain one or more of the 396 works which I am studying, and the information which each of these manuscripts provides about the authorship of particular poems is being added to my database of poem titles and ascription arguments. The relative merits of the various manuscripts--the degree of credibility of the various scribes--will need to be assessed and their claims weighted accordingly.

The second part of the collection of "external" evidence is the examination of claims for and arguments against Lydgate's authorship of various works as published by later critics. A bibliography of editions and studies of these 396 works is being compiled. While based upon the bibliography made by Renoir and Benson, this has been expanded in three ways: by bringing Renoir and Benson's bibliography up to date (a decade has passed since their work), by expanding their 200 titles to my 396, and by fully annotating all of the entries. This process is well advanced, and I hope soon to have a publishable annotated bibliography of Lydgate and Lydgatiana. I am also nearing completion of a handlist of manuscripts of the works of Lydgate and Chaucer which will include summary descriptions of all of the manuscripts involved in my study. I hope, eventually, to exapnd this database to include descriptions of all known late Middle English literary manuscripts.

The bibliography of secondary works will also contribute to the second stage of the project: a review of past opinion on the "internal" evidence of authorship, the stylistic qualities which identify Lydgate's writing. This second stage, though, will not only review the literature on Lydgate's style, but will also examine the literature of stylistics in general and of computer text analysis in particular. And it is in the light of this review that the determination of a specific methodology for the third stage of the project will take place.

The third stage of the project, then, will be an attempt to expand the present state of our knowledge of Lydgate's style by the contribution of new studies of the works with the aid of the computer. I am currently collecting machine-readable versions of Middle English texts, Lydgate and Lydgatiana, primarily, but also works by Chaucer, Hoccleve, Clanvowe, and various fifteenth-century writers to provide materials for a "control" group upon which to test our various hypotheses as they develop. Again, while the precise method is to be determined in the light of the review, yet to come, of the stylistic literature, I am collecting a significant quantity of text upon which to try out various approaches. In general terms, I see this third stage as involving the production of concordances to the various works (lemmatized and including substantive variants) and studies of various stylistic features such as vocabulary distribution, collocations, and rhyme-pairs.

Finally, the fourth stage in this four-stage project will be the actual application of the criteria estalished in stage three to the various texts under consideration--that is, we come now to the production stage. Again, this will involve some methodological problems, since many of the 396 texts are very small, and perhaps too small to provide adequate samples for analysis: we will need to keep in mind throughout that any tests dependent upon large sample sizes will limit their usefulness in parts of this project. At the same time, though, it should be kept in mind that the statistical-stylistic analysis of these texts is only one part of the larger process--none of the traditional approaches of manuscript study and text analysis is being replaced or forgotten, merely supplemented with a new kind of data--and if we are not able to generate significant statistical data for each and every one of our texts, this need not be considered in any way detrimental to the enterprise as a whole: we are simply gathering as many types of evidence as we can to settle these questions of authorship as well as we can.

In sum, then, this project is intended to examine a very large body of texts in order to reassess the evidence in one of the largest canon problems in English literature. I intend to put the current state of "stylometry" to the supreme test, to explore the efficacy of current methods in the solution of a problem of gargantuan proportions: the sheer size of the amount of text involved will involve, literally, tremendous problems. The Lydgate canon of 195 titles, as described by Renoir and Benson, would easily fill three volumes each the size of the Works of Shakespeare; add to these my additional 201 titles, and you have a couple of volumes more; add the "control group" works by Chaucer, Hoccleve, and so on, and you could fill a few bookshelves. And these problems of size will be further complicated by the difficult and relatively fluid nature of these Middle English poems and of Middle English spelling: and then, too, there are the variations in size and in kind among the 396 works to be studied. We may need, in fact, to develop a number of separate procedures to deal with the various genres, sizes, and dates of Lydgate's works, and this stage of the project will probably take the form of a series of smaller, more specific studies. These are, indeed, formidible problems.

But the rewards for success, too, are on a grand scale. And along the way, quite independent of the ultimate success or failure of the canon study, the project will result in a number of usable by-products: a significant corpus of machine-readable late Middle English texts, a bibliography of materials relating to Lydgate and to all the works ever attributed to him, a handlist of all of the manuscripts which contain Lydgatiana, and a set of concordances to these works, all of which should prove to be quite useful to other researchers in fourteenth- and fifteenth-century English literature.

Next page

The Canon of John Lydgate Project

© 1990 Literary & Linguistic Computing.
Reprinted by permission.
All rights reserved.
Last revised: 6 Nov. 1995

email: Stephen.Reimer@UAlberta.Ca
URL: http://www.ualberta.ca/~sreimer/lydgate.htm/