Iwona Mazur

Towards common European audio description guidelines: Results of the Pear Tree Project

This article reports on the Pear Tree Project (PTP) conducted as part of the DTV4All project, whose original aim was to develop audio description (AD) guidelines in Europe, in order to ensure consistent high quality AD. However, before streamlining AD standards, a number of issues had to be addressed, the most essential being whether relevant cross-linguistic and cross-cultural differences in Europe are insignificant enough to enable the development of such common European AD guidelines. In order to answer these questions, a methodology proposed by Chafe (1980) concerning the way representatives of various cultures and languages perceive and describe moving images was adopted by a group of AD researchers in the PTP. Participants from various countries were asked to watch a short film and recount what they saw. The data were then subjected to comparative lexical, discourse, and narrative analyses in order to uncover both similarities and differences in the processing of visual information by representatives of the languages and cultures concerned. The results and their analysis will be presented in this paper, on the basis of which the authors will attempt to provide an answer to the question of whether creating common European audio description guidelines is feasible.

Pear Stories and Audio Description: Language, Perception and Cognition across Cultures

In the mid-1970s, professor Wallace Chafe, a specialist in Native American languages at the University of California, Berkeley, set up an experiment to test the reception of a visual story in different languages. He designed a very simple film (the Pear Film) to elicit stories (the so-called ‘pear stories’) from speakers around the world. The idea was to compare how representatives of different cultures and languages talk about universal experiences: a man picking pears, a boy stealing a basket of the pears, then riding a bicycle before colliding with a stone and falling down, and some boys helping him to gather the pears and helping him back on his bike. Some 30 years later, under the umbrella of the European project DTV4ALL, academics in many countries from the field of Audiovisual Translation decided to use the Pear Film to check whether relevant cross-linguistic and cross-cultural differences related to film reception are insignificant enough to enable the development of common European audio description (AD) guidelines. The project was called the Pear Tree Project (PTP) and results related to the project are presented in this special issue. Some of the articles deal directly with the project, whereas others take the film as starting point but extend beyond the project itself in the context of the theory and practice of audio description. In the opening article, Iwona Mazur and Agnieszka Chmiel report on the PTP. In the article, the authors present the project’s rationale, design, participants, and findings, which are based on the comparative lexical, narrative, and discursive analyses of written narratives produced by representatives of various countries. In the final part of the article, the authors point to methodological limitations of the study, but conclude that, despite the limitations and huge variation in the obtained results, we could assume that development of common European AD guidelines should be possible, provided such guidelines take into account the cross-cultural differences in narrative production and the preferences of blind audiences in the respective countries. Christopher Taylor and Giovanni Mauro’s article is an extension of the original PTP analysis presented in the opening article. The authors use the original PTP data and subject it to further geo-statistical and linguistic analyses. In the first part of the article they present a number of mapograms, which are very effective visualisations of the obtained PTP data. The second part of the paper is linguistically oriented: the PTP texts are treated as a small corpus, which is subjected to a number of analyses related to systemic-functional grammar, including theme-rheme progression, information flow, clause-type choice, and appraisal. The authors then link the findings of the geo-statistical analysis with those obtained in the linguistic one, and draw conclusions as to possible applications of the findings to audio description practice in the countries concerned. The overall conclusion in line with that of Mazur and Perspectives: Studies in Translatology Vol. 20, No. 1, March 2012, 1 3

Researching preferences of audio description users — Limitations and solutions

Reception studies are frequently used in audio description research to elicit preferences of the visually impaired about certain aspects and level of acceptance of various solutions. However, this research method is characterised by limitations, which are discussed in this article as regards the participants and the design of reception studies. We then present a study which we think has been successful in overcoming some of these limitations, conducted as part of the European project entitled ADLAB: Lifelong Access for the Blind on 80 visually impaired persons (VIPs) and 77 sighted controls from six project partners’ countries. The respondents were presented with various audio description solutions and answered preference, comprehension and visualisation questions to find out which solutions they preferred, how much they understood following a given description and how easy it was for them to imagine a given description. We conclude that eliciting subjective opinions of respondents might be inconclusive a...

Should Audio Description Reflect the Way Sighted Viewers Look at Films? Combining Eye-Tracking and Reception Study Data

The chapter by Mazur and Chmiel reports on a study combining eye-tracking and reception study data to find out to what extent audio description beneficiaries appreciate eye-tracking based descriptions. In-depth interviews were conducted with 12 visually impaired respondents who saw videos either with audio descriptions based on eye-tracking data from sighted viewers and reflecting film language, or with audio descriptions based on long-established British audio description standards. The results seem to suggest that audio description should take into consideration the perception of sighted viewers. The findings also show that preferences are greatly determined by individual experience (such as prior exposure to audiobooks).