Panel Topic

Folkloric communication is a specific mode of social interaction that uses existing knowledge and creatively adapts it to particular situations, audience, and intent. It always includes some repetition and some innovation or, in other words, as Michel de Certeau (1984) expresses it in his study of everyday practices, the main attribute of cultural transmission is the changing nature of everything that is being passed on.

The questions of authorship, stylistics and variation in time and space have been discussed and problematized in folklore studies for over a hundred years. Digital era has brought along mass digitization of cultural heritage documents and the compilation of folklore databases and text corpora (see e.g. Schmitt, 2014). However, the use of computational methods in researching folklore and its specifics has thus far been modest. The intention of this panel is to discuss the possibilities that computational methods have to offer in revealing the inherent qualities of folkloric communication in various text collections and research corpora. The participants of the panel deal with different source data including archival records and publications as well as contemporary social media. They discuss different aspects of folkloric communication and variation, contributing together to a general insight into these processes.

The Panel: speakers and contributions

In this panel, the participants will introduce their projects, data, methods and research results to contextualize and elicit a discussion on the state-of-the-art in digital folkloristics, and on possible ways forward.

Mari Sarv and Risto Järv (Estonian Literary Museum) analyse the essence of folkloric variation relying on the text corpora from the collections of Estonian Folklore Archives. The Estonian Runic Songs’ database (1996-2015) contains ca. 100,000 poetic texts, and the database of folk tales consists of 11,000 tales (both together with metadata). Previous studies of folksongs have shown that the statistical analysis of poetical features of songs as well as their content allows us to locate their geographical origin (i.e. tracing the belonging of songs to the tradition of a local community) quite precisely. At the same time there are clear differences in the geographical distribution of linguistic-poetical features, especially compared to the elements of content. The stylistic analysis of folklore texts enables us to find out how much the personal style of performer is revealed in folkloric recordings of traditional plots (‘types’ in the folkloristic discourse). The potential of computational methods to tackle the dichotomy of stability and variation in the folkloric communication poses a most intriguing challenge.

Kati Kallio (Finnish Literary Society) will discuss the characteristics of oral poetry, including complex patterns of variation and tricky definitions of authorship, and concentrate on the challenges posed by of the SKVR-corpus of Finnic oral poetry (see also Digital Archive of Finnish Folk Tunes). The corpus represents various languages and dialects, orthographies, personal writing styles and traces of different modes of performance, but is held together by similar poetic registers. For this kind of specific poetic register in several related small languages and across a wide variety of genres, no ready-made tools for computational linguistic analysis exists. On the other hand, the corpus already includes a detailed thematic index, and the researchers have applied various manual methods to the corpus for hundred years. What kinds of new questions could be answered with computational methods? The author will present some test analyses and discuss the future possibilities of digital folkloristics on oral poetry.

Greta Franzini and Emily Franzini (Göttingen Centre for Digital Humanities) will elaborate on two research projects focusing on the computational interrogation of folktale collections and corpora. Using the Brothers Grimm’s Kinder- und Hausmärchen as a case study and base reference, one project addresses the popularization of fairy tale motifs triggered by the Brothers, and seeks to algorithmically crawl web corpora to study the global network of motifs. Motifs form the significant set of “key words” of a tale. In order to systematically crawl for parallel texts, we prepare a digital, machine-readable and -citable index in different languages. TRACER as a text reuse framework is used to check if a document contains similar keywords or describes the same tale. It implements a seven-layer approach combining segmentation, preprocessing, featuring, selection, linking, scoring, and postprocessing steps. The other project examines the textual evolution of the Kinder- und Hausmärchen, starting with the first edition published in 1812 to the seventh and last in 1857. The number of fairy tales grew with every edition, and the numerous changes the Brothers made over the decades in terms of both style and content were symptomatic of societal interest and development. These seven editions represent an ideal testbed not only to computationally verify existing research about this progression but also to identify and distinguish the authorial and stylistic fingerprints of Jacob and Wilhelm Grimm. The discussion will demonstrate the possibilities afforded by the Digital Humanities to conduct web-scale and Big Data research.

Liisi Laineste (Estonian Literary Museum) is applying the methodologies of Digital Humanities to the folkloric aspect of social media content in combination with theories of global information flow, participatory journalism and humour theory. Internet has become central in contemporary cultural communication – most of online communication can be treated as folklore, in which shared norms and values are constructed through cultural artifacts. Forums, commentary boards of news sites, blogs, Twitter and other social media applications have become commonplace during the last fifteen years. In the Internet, news, ideas and opinions travel fast. Social media spreads folklore in unforeseen volumes across national and cultural boundaries. Above all, the focus is on producing and consuming texts, images and multimedia. The presentation will attempt to trace such cultural texts as they move across and between cultures. Besides, as social media is often perceived as a catalyst and accelerator of public discussion and citizen movements, cultural texts as valuable agents in citizen engagement will be discussed in the light of the 2015 refugee crisis in Europe.