The term doculect is sometimes used for a variety of a language that has been described or otherwise documented in a coherent way. The term was originally designed to refer to the lowest level languoid that linguistics can meaningfully refer to. In a more extreme view, even the pure mentioning of a speech variety in any source (possibly without any information about the language itself) can be seen as a doculect. In this view, a reference to a language in a classical source, in a traveler's diary, or in a census are also doculects. The use of the term doculect is meant to remind linguists of the fact that
- the most basic entity of linguistic investigations (i.e. speech varieties) should not be assumed without explicitly mentioning any source;
- languages show high internal variability, which implies that different documentations should not a priori be assumed to describe the same entity.
For instance, two grammars of Egyptian Arabic describe two different doculects, even if they are both about, say, educated middle-class colloquial Cairo Arabic. The mere fact that there are two different descriptions makes the varieties described in them different doculects. Another example would be two different corpora, e.g. a corpus of French based on Le Monde and one based on Agence France Press. Even though one would perhaps not expect any differences in the structure of the language between the two corpora, the two varieties are by definition different doculects. Also, when examples are given in a publication, based on personal knowledge of the author, or on personal communication between the author and a native speaker, these examples themselves form an (idiolectal) doculect.
The term arose in discussions between Michael Cysouw, Jeff Good and Martin Haspelmath in 2006 at the Max Planck Institute for Evolutionary Anthropology. It is based on the term lect and the word documentation.
- "Jeff Good (pers. comm.) has introduced the term ‘doculect’ to refer to the variety of the ‘language’ that ends up in the documentation." (Bowern 2008:8)
Bowern, Claire. 2008. Linguistic fieldwork. New York: Palgrave Macmillan.