University of Surrey Institute of Advanced Studies

University of Surrey

Audio Description for Visually Impaired People

Towards an Interdisciplinary Research Agenda

28 June 2007 - 29 June 2007

Workshop Report

Audio Description is a means of helping visually impaired people to follow TV programmes, films and theatre performances by using quiet moments in the original to provide a description of the actions, the scenery, body language, and other relevant details. AD is the key to improving access to audiovisual material for VIP.

In AD, visual images are 'translated' into text i.e. transferred between 'modes', raising challenging new questions for translation studies, a discipline which has traditionally been concerned with translating messages between languages. This seminar aimed to map out a research agenda for this emergent area that identified key research questions and methodologies, drawing on multimodal discourse analysis, film and media studies, and creative writing as well as analysis of visual content.

Our plans to host this seminar were driven by our firm conviction that there is not only a pressing need to foster the spread of audio description practice (as a means of media access for blind and partially sighted people) but that there is also an important role for research to raise awareness for the challenges associated with the creation of audio description (AD) and its reception. AD is an emerging research discipline with a nascent research agenda which to date is often practice-oriented, e.g. relating to the development of standards and guidelines in different countries. We felt it was timely (or perhaps high time) to bring together researchers and practitioners from different contexts and backgrounds to discuss questions at hand and to work towards extending the research agenda for this area. We were convinced that broadening the range of disciplines contributing to the research agenda would support this socially important activity, add fresh perspectives to its investigation and in turn also provide new research opportunities to these disciplines. Hence, interdisciplinarity was at the very heart of this seminar from its inception.

The Seminar


The 2-day event was held at the University of Surrey on 28-29th June 2007. It turned out to be not only highly interdisciplinary, but also truly international. It was attended by a total of 30 participants from eight countries (Belgium, Brazil, France, Germany, Norway, Spain, Sweden and the UK), including 2 PhD students among a total of 16 presenters. In addition, staff members of the University of Surrey (Languages and Translation Studies (LTS), Culture, Media and Communication (CMC), and Dance and Theatre Studies) attended the seminar as well as postgraduate students from the MA in Monolingual Subtitling and Audio Description offered by the Centre for Translation Studies at Surrey. Especially welcome were two blind users of audio description services from Norway and the UK.

The Programme: Day 1

The first session on day 1 — entitled "from practice to research" — was opened by Veronika Hyks (IMS Media and CTS Associate Tutor, Surrey) and Andrew Holland (VocalEyes), two pioneers of AD for film and theatre respectively. Their presentations clearly demonstrated AD is not simply a matter of describing what ones sees on the screen or stage, but that more depth of description needs to be provided to contextualise the visual information for the target audience. Taking this idea further and applying concepts from narratology to AD, Andrew Salway (Burton Bradstock Research Labs) and Alan Palmer (London) investigated how linguistic means of expression can convey information about characters' mental states when re-telling stories. Focussing on cross-cultural differences in story telling, Pilar Orero (Barcelona) presented findings from an experiment replicating Wallace Chafe's Pear Stories project and explored the most salient characteristics of three AD scripts for the same film as well as differences in the narrative structures across different cultures. In another comparative study, Eliana Franco (Bahia) examined amateur and academic AD practices in Brazil, stressing the need for local varieties adapted to the target audience's preferences and showing how this approach had guided her research group in the development of AD standards for the Brazilian audiovisual media.

The second session of day 1 was devoted to training. It was opened by Peter Lilliecrona (Gothenburg), who reported on a project for training audio describers in Swedish theatres, drawing the seminar participants' attention to the specific challenges of this form of 'live' AD and also acknowledging the enormous amount of voluntary work done in this area. Regine Hampel (Open University) then focussed on the different communicative affordances of visual and verbal modes of communication, using the example of language learning, and emphasising how it had changed through the availability of an increasing variety of communication modes. This opened up a discussion of future applications of AD and the need for inclusion of such aspects in AD training.

The third part of day 1 was taken up by a workshop on AD in dance. Kate Lawrence (Surrey) enchanted the participants with a marvellous self-choreographed dance performance and subsequently discussed different approaches to describing and thus reconstructing live dance performances. This was complemented by the participants' on-the-spot attempts to create different versions of AD for Kate's the dance performance, which led to a most insightful discussion among the sighted and blind participants of the seminar. The multiple layers of interpretation raised once again questions relating to what should be the basis for the audio describer's selection of aspects of (in this case) the performance.

Margaret Rogers (Surrey) wrapped up day 1 by formulating a number of research questions ensuing from the presentations, and emphasising the emerging parallels between the study of AD and Translation Studies in particular. These had become especially apparent in the application of Eugene Nida's concept of 'dynamic equivalence' to characterise the need to achieve a similar effect on source and target audiences, and in raising in the question of the 'visibility' of the audio describer. They were also well reflected by a commonly used metaphor, juxtaposing the audio describer to a 'visual interpreter'.

The Programme: Day 2

The presentations on day 2 were guided by the motto "interdisciplinary contributions and future prospects" and were aimed at approaching AD research from a wealth of different perspectives and neighbouring disciplines. Sabine Braun (Surrey) adopted a discourse analysis framework to model the comprehension process of the audio describer, focussing on the interaction of verbal, visual and auditory cues with each other and with the audio describer's knowledge, and highlighting the ensuing implications and challenges for 'objectivity' in AD. Fiona Doloughan (Surrey), in turn, explored the process of creating AD narrative and in particular investigated the possibilities and limitations of creating visual effects through words, drawing on research on translation, multimodal communication and on narrative in multimodal and multimedia contexts. Gert Vercauteren (Antwerp) used a pragmatics-led approach, based on Relevance Theory, to provide guidance for the selection of filmic elements for AD and, equally importantly, for how to describe them, thus taking up the question — which ran as a major theme through many presentations — whether the audio describer should provide a mere 'reproduction' or some degree of interpretation of the visual stimuli presented in the film (and whether there is such a choice at all).

Starting from the often overlooked point that sound effects can easily become meaningless without the visual image and are therefore likely candidates for AD, Aline Remael (Antwerp) used insights from film studies to distinguish different types of filmic sounds, and presented first findings of her research into the challenges that these create for AD. In a similar vein, Joaquim Pujol (Barcelona) presented a model of film analysis for audio description, arguing that there is a need for audio describers to understand the basic structure and mechanisms of films in order to produce AD narrative which fully takes the artistic intent of films into account. Last but certainly not least, Ana Ballester (Granada) and Andoni Eguíluz Morán (Deusto) presented two forward-looking research projects (MUSAI and TRACCE), which are aimed at the development of AD production and analysis software and the creation of an AD corpus as a data base for future research. This triggered a highly interesting discussion of the value of different types of data and research methodologies in AD.

The Seminar: Panel Discussion

The seminar was concluded with a panel discussion, chaired by Margaret Rogers and joined by some of the presenters as well as Joan Greening from the British Royal National Institute for the Blind, and aimed at highlighting the issues that should figure on an interdisciplinary research agenda for AD.

To recap, at the beginning of the seminar, Veronika Hyks had reminded the participants of the need to deliver an intelligent and appropriate AD which satisfies the audience's varying preferences, and of the difficulty of maintaining a constant high standard in both the writing of an AD script and its delivery. This was later corroborated by other presenters' findings on cultural differences and local preferences for AD solutions. Andrew Holland had drawn the seminar's attention to the fact that every work of art — whether a painting, film, theatre or dance performance — is an expression of ideas and that an audio description therefore needs to "lead through the physicality [of the performance] to those ideas". The ensuing challenges, such as the 'problem' of subjectivity in the descriptions provided a recurring theme of the seminar. The issue of methodologies arose a number of times: corpus-based studies emerged as a particularly fruitful path, supported by customised AD tools and allowing both quantitative and qualitative research. The presentations and the lively discussions drew on a wide variety of insights from disciplines such as narratology, discourse analysis, pragmatics, semiotics, translation, film and dance studies, all with the aim of modelling the creative processes involved in AD and providing AD solutions which, as Andrew Salway put it, are capable of taking the audience on an "emotional journey" through a work of art. This point was considered to be crucial for making a work of art not simply 'accessible' for a blind and partially sighted audience but turning it into a socially meaningful and — last but not least — enjoyable experience.


The seminar has provided ample evidence that AD is a complex cognitive-linguistic and intermodal mediation activity and that an interdisciplinary approach to the study of AD is therefore more than appropriate. It has also shown that the study of AD is about to establish itself as an exciting new research discipline. At the same time, it was felt that the further study of AD will be capable of fertilising the many related fields on which it draws by providing new sources of data in intermodal communication and that it will produce results which are transferable to other areas of audiovisual translation, especially subtitling for the deaf and hard-of-hearing.

While acknowledging that there are clearly many areas of multimodal and multimedia communication which need to be addressed in terms of media access and that much work lies ahead for all of these, there was also a strong sense of agreement among the participants that an extended seminar format dedicated to the study to AD was an extremely useful (and hitherto rare) opportunity. There was great satisfaction with the outcomes and the stimulus which the seminar had provided to practitioners and academics alike.

In view of this and given the overwhelming interest in the event (including a large number of enquirers, e.g. from Austria, Denmark and Australia, who, for a plethora of reasons, were not able to join us), a follow-up event on the study of AD, possibly on a larger scale, is planned in due course. Selected papers from the event in June will be published in a volume with an international publisher. Plans are already underway (to date, a proposal for an edited volume has been prepared and is due to be submitted by the end of October) with a projected completion date of Autumn 2008.

What remains to be said is that the organisers and participants are grateful to the Institute of Advanced Studies at the University of Surrey for their financial support and helping us to create this unique opportunity. The organisers would also like to thank Kim Stacey at the conference office for her support throughout, and the university's catering teams for lunches, refreshments and a wonderful conference dinner at Oak House. Special thanks to Heather Norman for her invaluable support with all administrative matters and to Gillian James for her passionate assistance throughout the event.

Margaret Rogers and Sabine Braun
Centre for Translation Studies, University of Surrey
29 October 2007