(p.232) Appendix I Research Plan and Selection Criteria
(p.232) Appendix I Research Plan and Selection Criteria
Designing the Project
The goal of the research plan was to find substantial family letter collections written by unknown men and women that spanned the entire long eighteenth century. Ideally, these collections would include a representative sample of writers from different classes, occupations, religions, and geographical areas. Family letters that extended over more than one generation would be used to show continuity and change over time. It was expected that most correspondences would be written by the upper classes, with a sprinkling of middling‐sort and mercantile correspondences. This proved to be a false assumption.
Catalogues of the British Library, the Bodleian Library, the Beinecke Library, Yale University, and a wide range of local archives were sampled to see if there was enough new material for the book. Discussions about sources and methodology were held with the Public Record Office and the Historical Manuscripts Commission, which have since become part of The National Archives. The Internet was searched for new databases, along with traditional bibliographies.
Despite initial assumptions, it appeared that extensive letter collections written by people below the rank of gentry might indeed exist. To test this supposition, the Access to Archives Database (A2A) was searched in detail. Now part of the larger UK Archives Network, A2A contains catalogues describing archives held locally in England and Wales, many of which were previously inaccessible to online users. A list of correspondence that appeared suitable was then compiled for each archive. Next a questionnaire was sent to all local record offices requesting data about letter collections of non‐elites, especially items that had already been found in a broad range of sources. Positive responses were received from archives in the north of England. Since Scottish literacy was high and northern boundaries (p.233) were porous, numerous visits were made to that area. The combination of electronic and archival searches was useful in targeting possible sources. It soon became evident there was more material than one book could encompass. A rigorous selection process would, therefore, have to be developed.
In order to examine each archive in a consistent manner a checklist of required data was constructed. Information would be collected on three levels: (1) for the collections of family papers, (2) for the persons who wrote the letters, and (3) for the letters themselves. Appendix II provides information about selected family archives that were used in this book. It includes the location of the archive, catalogue references, types and dates of documents, and principal locations of writers. Appendix III presents data about persons who wrote the letters including their dates, occupations over time, religion, and geographical location. Much of this information was gleaned from other supporting documents. Appendix IV describes the letters, themselves, including types, principal subjects, social relationships and uses. It is important to note that archives contain many types of letter forms: drafts, copies, those in albums, letter books, diaries, and journals, those post marked, hand delivered, and those never sent. All of these letter forms, which give insights into the writing process, were used in this study.
To obtain the data, notes pertaining to individual letters were indexed by standardized subject headings. The basic literacy of writers was evaluated in a separate data field. Strengths of collections were also recorded, for example if they answered questions about gender, region, or religion. A final summary of the motives, uses, and impact of letters was completed for each collection. These last three summaries of skills, strengths, and uses were crucial in supplying candidates for case studies.
Selecting the Case Studies
Rules were established at the outset for deciding whether to read letters in a particular collection. The basic requirement was a substantial (p.234) unbroken series of letters that spanned at least two consecutive generations. Collections that included replies were highly rated, for this permitted the reconstruction of actual dialogues. To be considered as a case study, however, additional materials that provided background about writers were also necessary. These might include family trees, wills, birth records, marriage settlements, genealogical research, extracts from reading, writing exercises, and annotated instructions about how to maintain the collection. With these references in hand, it was possible to begin to reconstruct the lives of specific letter‐writers. Without these sources, it would be difficult to interpret and place in context the overwhelming minutiae of letters.
The discovery of only a few ‘exceptional’ letter‐writers would have made the task of choosing case studies easy. Instead, disciplined cutting was necessary. One alternative was to use brief quotations from all or most of the collections. Because this would have produced a kaleidoscope of fragments, the decision was made to create fewer, but fuller, case studies that were as in‐depth as space would permit. Finally, a chart of sixty‐three collections that had been located was compiled.
Because of the large number of possible collections, it was decided to eliminate titled and gentry archives and focus on those written by people lower down the social scale. The Evelyn papers were used, but only to present a contrast with those of merchant families. Likewise, a few elite archives contributed to overall themes in general chapters. For example, thirty‐four letter collections, numerous local histories, and a wide range of secondary literature were drawn on to describe uses and attitudes to the post office in Chapter 2. Eventually fifteen core collections were chosen as case studies. I believe they present a representative sample of the collections consulted. Together, they comprise a diverse group of letter‐writers and different types of letters penned at various times and locations.
As the project developed into one that examined levels of literacy, it was clear that manuscript letters should be used whenever possible. Only then would there be a comparable base of texts that showed original spelling, punctuation, and grammar. Editors of printed letters often deleted information that revealed literacy skills, especially those of women, children, and the lower sorts. Manuscripts also showed how language was reworked: words crossed out or added between lines or in margins; dates forgotten or inserted; annotations and endorsements added by others. Most important, editors of printed collections often omitted references to letter‐writing and (p.235) the post office that occurred at the beginning and end of letters. A large database of these references was constructed and used to write this book.
As more case studies were chosen, common patterns emerged. These shared ways of thinking suggest that they are representative collections. Of course, letters of literate writers were ‘self‐selected’ and also biased towards northern areas. Yet without counting all letters, no group of them could ever make the claim to be wholly representative. Those that are included here have the advantage of being carefully selected from a sample of unused collections. The common questions that were addressed to all collections and the use of ‘epistolary literacy’ to measure universal traits of letters (see Chapter 3) have helped to control the random nature of all correspondence. Finally, because this book has dealt with letter‐writing on a popular level, the universe of possible writers is smaller than that of elite individuals. This increases the probability that the letters are representative of their kind.
Obviously, bias in the selection and interpretation of correspondence is inescapable. Yet current awareness of the constructed nature of letters has acted as a safeguard against overly transparent readings. Studies of this kind are rarely possible for periods before the eighteenth century. Indeed works about earlier eras assume that letters were exceptional, written by a tiny portion of the population. The discovery of so numerous and wide a range of eighteenth‐century letters indicate that this stereotype had ceased to be relevant long before 1800.