For most languages, imaginative literature forms only a small proportion of extant material. Yet even strictly linguistic research, outside the modern period, concentrates mainly on literary texts wherever they are available. From the outset, the Anglo-Norman Dictionary (AND) went beyond such sources and cited from a substantial number of chronicles and administrative texts.

This project aims to extend the registers covered, by incorporating technical and specialist vocabulary from unpublished National Archives (formerly PRO) documents. Explorations of documents used by Salzman in his Documentary History of Building (1952); of guild documents; and of nautical items in French in Sandahl's Middle English Sea Terms (1951-1982), provide strong evidence that much valuable material awaits attention. HRB-funded work in the 1990s on documents listed by Salzman supplied important attestations. The online AND1 (in letters P-Z) and AND2 (A-G) cite Salzmann 271 times, but only 59 items reference Sandahl. Both these texts contain hundreds of references to National Archives documents, most of which have yet to be combed. No comparable dictionary is engaged in such primary research on unpublished documents, nor is any other able to make the resulting findings so quickly and easily accessible to international scholarship through an in-house electronic editing and publishing system.

The work will involve a targeted investigation of pre-selected documents, carried out by Dr Natasha Romanova, working within the context of the AND project. It has been reliably estimated that, given the chosen approach and techniques, it will be possible to comb over 1,000 documents in the course of a year. The principal selection mechanism (in addition to published references) will be the online National Archives catalogue. This enables identification of pertinent documents by keyword searching (e.g.: tin, wool, fulling, fishing ...). These documents will in many cases be examined as digital images on the National Archives' website: all of SC 8 (Ancient Petitions) is to be made available during the course of 2007.

The project will contribute to Middle English and British Latin (in the trilingual environment within which Anglo-Norman was used, documents of this type often contain multilingual elements) and to medieval French more generally (modern scholarship, partly thanks to the AND, now recognises Anglo-Norman as an important variety of medieval French). It will facilitate access to an increased range of documents and to increased knowledge of the past. It may prompt similar work in French archives, where there are shelf-kilometres of comparable material still awaiting the attention of editors and lexicographers. In due course, a selection of the documents examined might be the basis of an edition, printed or electronic, with the latter linked into the AND; but that would be a separate undertaking. Gleaning for lexicographical purposes, with immediate publication in AND entries thanks to the existing AHRC-funded technical infrastructure of the Dictionary, is to be the focus of this present project.