Written documents

This paper explains why large historical sociolinguistic corpora are needed to interpret traces of spoken features through the written medium. To support this, the eighteenth-century personal diary of a small merchant is compared with other documents to show that the diary displays a number of vernacular and formal features and is therefore considered hybrid in nature. It is also shown that even a homogeneous collection of family letters can constitute a microcosm of linguistic communities and can reveal sociolinguistic changes. Through a study of the relations between close and extended family members living in Detroit in the nineteenth century, the author examines how to interpret the linguistic variation found in documents written by less-skilled authors.