From the field to the web: implementing best-practice recommendations in documentary linguistics

The imminent death of thousands of the world’s languages over the next fifty years has driven linguists to search for solutions to the problems caused by this impending loss of linguistic diversity. A series of works published in the 1990s declared a state of crisis among the world’s languages (e.g., Robins and Uhlenbeck (1991), Crystal (2000), Nettle and Romaine (2000)) and initiated a focus on what has become known as documentary linguistics: “a fairly independent field of linguistic inquiry and practice that is no longer linked exclusively to the descriptive framework.” (Himmelmann 1998, p. 161) The crucial difference between documentary and descriptive linguistics is that the former concentrates on the collection of primary data (e.g., elicitation, recording, transcribing, translation) whereas the latter is concerned with the analysis of the primary data (see Himmelmann 1998, p. 162). Although in many cases it is impossible to stop a language from dying, refining the methodologies underlying documentary linguistics is essential to a good documentation of the language while speakers still exist. This documentation in turn provides present and future generations of linguists with empirical data for research, historians and anthropologists with information on a speech community’s unique history and cultural heritage, and speakers themselves with essential material for their heritage preservation efforts (see, e.g., Nettle and Romaine (2000), Blythe and Wightman (2003)). One major trend emerging in documentary linguistics over the last decade is the widespread use of computers for digital instead of paperand tape-based language documentation. Bird and Simons (2003a) are concerned that “digital language documentation and description becomes inaccessible within a decade of its creation” because digital records “are often tied to software versions, file formats, and system configurations having a lifespan of three to five years.” (p. 557). Following discussions in the academic community, notably among members of the Open Language Archives Community (OLAC) (Bird and Simons (2003b)), the International Standards for Language Engineering Metadata Initiative (ISLE-IMDI) (Wittenburg and Broeder (2002)) and the Electronic Metastructures for Endangered Languages Data initiative (EMELD) (Dry (2002)), Bird and Simons (2003a) construct detailed best-practice recommendations for the creation of digital language documentations and descriptions. The authors envision that once the academic community embraces a coherent set of best-practice recommendations, it will become possible to overcome the “unparalleled confusion in the management of digital language documentation and description.” (p. 579). At the end of their in-depth article, Bird and Simons point out that