[Access article in PDF]
The Brontë Sisters and the Christian Remembrancer:
A Pilot Study in the Use of the "Burrows Method" to Identify the Authorship of Unsigned Articles in the Nineteenth-Century Periodical Press
The nineteenth-century literary scene in Britain, as readers of the Victorian Periodicals Review are well aware, was notable for the number of serious quarterly and monthly periodicals dealing with political, literary, historical, and scientific subjects of the kind that now have their own specialised academic journals. It was, however, the custom for these articles to be unsigned, and this has meant that what the editors of the Wellesley Index to Victorian Periodicals have called the "most basic questions" of the "careful scholar" – "Did So and So write on such and such a topic, and if he did, when and where was it first published?' – have often proved difficult to answer.1
For more than sixty years2 both literary scholars and historians have been trying to provide answers to these questions, and a large number of indexes have been published, including the multi–volume Wellesley Index itself, where the authors of unsigned contributions to 43 nineteenth-cen-tury periodicals have been identified.3 Thus a very substantial database now exists for identifying both the literary journalism of interesting figures and commentary by the less well-remembered on major literary, social, and political issues. The task is, moreover, still ongoing, as the Curran Index and Rosemary T. VanArsdel's Victorian Periodicals: Aids to Research: A Selected Bibliography demonstrate.4 Most of this research has, however, involved the use of what Harold Love calls "external evidence': scholars have concentrated on combing surviving publisher's records, memoirs, and letters from the period for information on contributors [End Page 21] to its various periodicals.5 The more obvious literary and archival sources have, in consequence, been largely wrung dry, while the authorship of many interesting articles still remains unidentified.
The authors of this paper have begun exploring an alternative way of establishing the authorship of such articles: the method devised by Professor John Burrows of the Centre for Literary and Linguistic Computing (CLLC) at the University of Newcastle, Australia. In the early days of computer analysis of texts Professor Burrows made the discovery that the incidence of the very common words of English, the "function" words, varies significantly between texts by different authors while remaining comparatively constant within a single author's work. During the past 25 years he and his colleagues at the CLLC have been devising and refining the statistical procedures that can most effectively isolate these distinctive usages, and they now have a suite of procedures and tests that can be used to identify the authorial "signature" embedded within a text or a group of texts.6 This "Burrows method" as it is coming to be called, has to date yielded interesting results when applied to novels, poems, and plays, but, until the present authors began their investigations, it has not been used to investigate unsigned periodical literature.
We believe that our initial application of the "Burrows method" to nineteenth-century periodical literature suggests that this technique can open a new avenue within the ongoing project of attribution. Whereas most methods currently in use rely on searching for existing hard evidence of authorship, this method relies on testing hypotheses that must be devised by the scholars concerned. Most scholars using periodical literature develop hunches about who was or was not the author of particular pieces, and there are also frequent fragmentary references in letters and diaries that hint at other identifications. The "Burrows method" provides a way of testing such hypotheses, and, although unable to "prove" authorship in the way an editor's letter or a notation in a diary does, it can transform a hunch into a statistically assessable possibility.
In this article we report on the...