This week on the ELAR blog, ELAR’s Senior Archive Assistant and Communications Officer Leonore Lukschy writes about her search for frog story data in ELAR, and the importance of metadata in endangered languages archives.

By Leonore Lukschy

Endangered languages archives provide large amounts of materials on underdescribed languages around the world. In times when linguistic field work is not feasible, archives such as ELAR offer a particularly important resource for linguistics students and researchers. The wealth of data held by ELAR and other endangered languages archives such as AILLA and Paradisec is discoverable through metadata – the richer the metadata, the more likely it is for materials to be found.

Linguists have long used Mercer Mayer’s children’s book ’Frog where are you?’ from 1969 to elicit speech data, more precisely descriptive narratives. The wordless picture book illustrates the story of a boy whose pet frog escapes. The boy and his dog venture out to look for the frog. In the linguistics literature the book is commonly referred to as ‘frog story’. Retellings of the book have been used to analyse how motion is encoded across languages, most prominently in Berman and Slobin’s Frog Story Project.

For a Linguistic Typology assignment, I decided to look at frog story data available in ELAR collections. Many depositors clearly state that their materials contain recordings of retellings of the frog story in the metadata, but others don’t. Some depositors include the title of the book in their collection’s metadata but not ‘frog story’. The issue with this is that there isn’t just one frog story book by Mercer Mayer, but several, including ‘A Boy, a Dog and a Frog’, ‘Frog, Where Are You?’, ‘Frog on His Own’ and ‘One Frog Too Many’. Others might say ‘story about a frog’ or ‘book about a boy and his frog’. And then there are some collections with no English metadata available at all. In order to find as many frog stories as possible, I searched ELAR for ‘frog’, however this lead me not only to frog stories but to all sorts of materials relating to amphibians.

Filtering these materials by genres and topics was also only partially possible, as some frog stories were categorised as narrative instead of elicitation and the topics varied from collection to collection. In a metadata dream world, any recordings of a frog story retelling would contain frog story in the title, have elicitation as a genre, frog story as topic, and include a description detailing the procedure of the elicitation session (were the participants shown the book beforehand? Were they looking at the pictures while describing them or was the description done from memory?). In a students’ or researcher’s dream world, the sessions would furthermore be open access, include annotated, transcribed and translated video, and links or references to publications referring to the data would be included in the session description.

There may be many more frog story recordings in the archive, but without the corresponding metadata they will stay hidden away. Hopefully, as better metadata tools are developed, and depositors’ understanding of the importance of good metadata grows, the metadata helping discover the incredible resources available in endangered languages archives will improve. Remember: Metadata is a love note to the future!