Audio description is an intersemiotic translation modality used to make art museums accessible to visually impaired visitors. Existing audio description guidelines in various countries recommend describing only that which is seen, in other words, to avoid subjective interpretations of the visual message. However, there is evidence that some visually impaired people prefer more subjective audio descriptions. The controversy around this issue has generated reception- and product-oriented studies of audio description which demonstrate that not only is subjectivity present in existing audio descriptions, but also that it may benefit the construction of a more meaningful experience. A methodology which combines corpus and contextual analysis and draws on cognitive linguistics as well as art theories has been followed in this study to examine audio descriptive guides in art museums in four different countries. Results show considerable levels of subjectivity and offer a categorization of this element. Additionally, the level and type of subjectivity appear to be influenced by contextual factors, including the level of abstraction of the artwork and the audio describer’s degree of compliance with existing guidelines.
- audio description,
- visual arts,
- cognitive linguistics
L’audiodescription est une modalité de traduction intersémiotique utilisée pour rendre les musées d’art accessibles aux personnes handicapées visuelles. Les directives existantes dans différents pays recommandent de décrire uniquement ce que l’on voit, en évitant les interprétations subjectives du message visuel. Cependant, il apparaît que certaines personnes en situation de handicap visuel préfèrent les audiodescriptions plus subjectives. La polémique liée à cette question a donné lieu à études de l’audiodescription, centrées à la fois sur la réception et sur le produit. Celles-ci montrent que la subjectivité est non seulement présente dans les audiodescriptions existantes, mais aussi qu’elle pourrait contribuer à l’élaboration d’une expérience plus significative. Une méthodologie associant analyse de corpus et analyse contextuelle, s’appuyant sur la linguistique cognitive ainsi que sur les théories de l’art, a été utilisée ici afin d’étudier les guides audiodescriptifs disponibles dans certains musées d’art de quatre différents pays. Les résultats font état de niveaux considérables de subjectivité et proposent une catégorisation de cet élément. De plus, le niveau et type de subjectivité semblent influencés par des facteurs contextuels, dont le degré d’abstraction de l’oeuvre d’art et la mesure dans laquelle l’audiodescripteur s’est conformé aux directives existantes.
- arts visuels,
- linguistique cognitive
La audiodescripción es una modalidad de traducción intersemiótica utilizada para hacer los museos de arte accesibles para visitantes con una discapacidad visual. Las directrices de audiodescripción existentes en varios países recomiendan describir únicamente aquello que se ve, es decir, evitar las interpretaciones subjetivas del mensaje visual. Sin embargo, existen pruebas de que algunas personas con discapacidad visual prefieren audiodescripciones más subjetivas. La controversia en torno a esta cuestión ha llevado a la realización de estudios de recepción y estudios del producto en audiodescripción. Estos estudios demuestran no solo la presencia de subjetividad en audiodescripciones existentes, sino también que esta podría ayudar a la construcción de una experiencia más significativa. En el estudio que aquí se presenta se ha seguido una metodología que combina el análisis de corpus y del contexto comunicativo y se apoya en la lingüística cognitiva y teorías artísticas para estudiar guías audiodescriptivas en museos de arte en cuatro países diferentes. Los resultados muestran considerables niveles de subjetividad y ofrecen una categorización de este elemento. Asimismo, hay factores contextuales que parecen influir en el tipo y el nivel de subjetividad, como son el nivel de abstracción de la obra de arte y la medida en que el audiodescriptor siguió las directrices existentes.
- artes visuales,
- lingüística cognitiva
Corps de l’article
Most audio description (AD) guidelines recommend a neutral or objective approach (ADC 2008; AENOR 2005; ITC 2000; Morisset and Gonant 2008; Salzhauer Axel, Hooper, et al. 2003; Snyder 2010) to this modality of intersemiotic translation, but more subjective descriptions seem to be beneficial at least for some sections of the visually impaired population (RNIB and VocalEyes 2003: 47-52). Due to this controversy in recent years, subjectivity has become an object of study in reception studies of AD in different contexts. The Open Art project (Szarkowska, Jankowska, et al. 2016) advocates for a universal accessibility approach to AD in art museums in order to create ADs that benefit visitors with and without a visual impairment. The study created and later tested ADs for displayed contemporary art. It concluded that visitors prefer shorter, more interpretative descriptions. For theatre (Udo and Fels 2009; Udo, Acevedo, et al. 2010), results indicate that first person audio descriptions which use more subjective content to portray the scene were positively received by audiences, as they found it enjoyable and entertaining. For film AD, in a study carried out by Mazur and Chmiel (2012), half of the participants stated that subjective audio descriptions are unacceptable. Despite that, almost three quarters of them actually preferred descriptions that contained evaluative adjectives. The results of a recent study (Walczak and Fryer 2017) indicate that more creative ADs provide a more immersive experience of the film by stimulating “presence” – understood as suspension of disbelief and perceptual illusion of non-mediation. In a similar study which measured the impact of neutral versus emotional language on the emotional responses of sighted and visually impaired individuals for film AD (Ramos 2016), results showed that emotional language elicited stronger responses in both groups.
Three product-oriented studies of AD in art museums have shown that subjective language is present in ADs of visual artwork (Lima and Magalhães 2013; Luque Colmenero 2016; Soler Gallego 2013, 2018). The goal of this article is to enrich this research area with a more encompassing study of subjectivity in AD for art museums. This study attempts to answer the following research questions: What are the levels and types of subjectivity in ADs for art museums? What is the connection between this subjective content and the context in which the ADs are created? This combination of product- and context-oriented studies within the descriptive branch of translation studies has been advocated for by Halverson (2013: 35) in her proposal for a cognitive linguistic method that takes into consideration different contextual factors – linguistic, non-linguistic, individual, and cultural – in order to explain translation phenomena.
2. On subjectivity in art education and museums
An important contextual factor and potential source of norms in museum AD are current theories of Art Education and Museum Studies. Within the latter, museums are conceived of as tools for learning and the democratization of culture. They provide individuals with an accessible and inclusive environment that fosters participation, critical thinking and a diversity of frames of reference through which to experience the world (Asensio, Asenjo, et al. 2014: 99). This concept of the museum can be traced back to Dewey’s philosophy of art and education. According to Ueno (2015: 97), Dewey’s definition of art as experience implies that the individual is shaped by the work of art and at the same time, creates meaning by connecting it with ordinary personal experience of the world. Similarly, for Eisner (2002: 44-5), the fundamental goal of education should be for learners to be exposed to a wide variety of meanings that humans can create and to deepen their knowledge of them while fostering their willingness and capacity to experience the world from those different frames of mind. More specifically, he states that the goal of art education programs should be to help learners develop their capacity to have aesthetic experiences in their daily lives. Eisner considers this type of aesthetic seeing a cognitive activity that needs to be learnt and whose nature he defines as follows:
Being able to see from an aesthetic perspective requires an ability to focus on the formal and expressive qualities of form rather than solely on its utilitarian functions. It requires the ability to slow down perception so that visual qualities can be inspected and savored. It requires one to search for qualitative relationships and to note the quality of experience they engender.Eisner 2002: 26
Hubard (2011: 18) observes that this combination of rational thinking and embodied ways of knowing is still rejected nowadays by many education professionals and art educators. She criticizes the argument advanced in some educational contexts that the importance of art education is based on its use in developing rational thinking skills. Contrary to this limited view of human thought and art education, Hubard defends the need to place the complex multimodality of human learning processes at the center of art education programs. This view of art education is also at the center of “cultural mediation,” an approach to teaching and learning which is defined as “the process of gaining and negotiating knowledge about the arts and social and scientific phenomena through exchange, reaction and creative response” (Mörsch and Chrusciel 2015: 14). When cultural mediation is specifically related to the arts, it is referred to as “art mediation” and is understood
[…] less as knowledge transmission and more as the act of forming relationships of mutual exchange among publics [sic], works, artists and institutions. Under this understanding, the aim of médiation culturelle is to place the various different perspectives involved in relation to one another. These activities focus on the individual perception of artworks by participants. The point is not to fill in lacunae in understanding with specialized knowledge, but to understand the lacunae as a point of departure for dialogue and aesthetic experience […].Mörsch and Chrusciel 2015: 18
The practical application of these principles is found in a curriculum developed by Housen and Yenawine, which is based on Housen’s (1992) study of aesthetic development from a cognitive psychology perspective. In her study, Housen (2000-2001: 4) identifies and describes five stages of aesthetic development: accountive, constructive, classifying, interpretive, and re-creative. The Visual Thinking Strategies curriculum is being used in different art education contexts, including art museums. It consists of educator-facilitated discussions about art and “is designed to match images and questions to the aesthetic developmental needs and naturally occurring capacities of beginner viewers” (Housen 2001-2002: 100). These discussions focus on the accountive and constructive stages and are conceived as group discovery experiences guided by open-ended questions posed by the educator. The three foundational questions are “What is going on here?,” “What do you see that makes you say that?,” and “What more can you find?”
The design, creation and assessment of accessibility resources should be coherent with the principles of museum and art education. ADs should function as tools that help achieve the educational goals of art museums and subjectivity should be studied as one of the factors that could have an impact on achieving these goals. Including subjective descriptions could help educators and museums convey the message that knowledge and skills should be built collectively and democratically, based on visitors’ and experts’ experiences and knowledge. On the other hand, subjective descriptions could hinder visitors’ freedom to have individual, first-hand experiences of art due to the influence of the interpretations provided in the audio descriptive guide. Visually impaired visitors could think that the voice of the sighted expert is the most (and maybe the only) valuable source of interpretation of art. For this reason, an important aspect of subjectivity in visual art AD could be how to flag it by means of linguistic devices so that receivers perceive it as an invitation to experience art and not as an authoritative voice, a carrier of truth.
The importance of carrying out a descriptive study of existing ADs is twofold. Firstly, the definition of subjectivity by both practitioners and users is not clear-cut, as indicated by the studies mentioned in the previous section, but this definition can at least be partly clarified through a systematic analysis of a bigger corpus of real samples of visual art AD. Secondly, a taxonomy of the type of subjective content found in existing ADs should be the starting point for reception studies that aim to assess the adequacy of current practices in this respect.
3. On subjectivity in linguistic and visual communication
The study of visual art AD as the product of an intersemiotic translation process (Jakobson 1959; Gottlieb 2005) should draw on theories of visual art and linguistic theories that allow researchers to describe and explain the connection between the non-verbal source text and the verbal target text. This theoretical approach was implemented by the TRACCE (Translation and Accessibility) research team at the University of Granada to carry out a corpus-based study of film AD. Films were annotated by means of a qualitative data analysis software at the narratology, film language, and grammar levels (Jiménez Hurtado, Seibel, et al. 2010; Jiménez Hurtado and Soler Gallego 2013). However, the study presented in this article differs from the one just mentioned on the grammatical level. I propose that cognitive linguistic theories, namely theories that attempt to describe and explain linguistic communication based on theories of human cognition, be used to study intersemiotic translation phenomena. Therefore, cognitive linguistics and, more precisely, Langacker’s cognitive grammar (1987/2008) and its further development by Croft and Cruse (2004) are applied here to the study of visual art AD. In cognitive grammar, the “construal” is defined as the conceptualization or interpretation of an entity or situation that is conveyed through language. Humans can create alternative construals of the same entity or situation by means of various cognitive operations. Croft and Cruse (2004) use these cognitive operations to name the construal dimensions as Attention, Judgement and Comparison, Perspective and Overall Structure. Each one of these construal dimensions, in turn, includes more specific operations of which two are relevant to the study of subjectivity: Focus of Attention (within the Attention dimension) and Metaphor (within the Judgement and Comparison dimension). This study discusses the Focus of Attention operation.
Focus of Attention is defined in cognitive grammar as the selection of conceptual content for linguistic presentation (Langacker 1987/2008: 57). The audio describer can focus on different components of the artwork. Dondis (1973/2006: 13) identifies three levels of communication in visual data: symbol systems, representational visual material, and the abstract underlying structure. The abstract underlying structure consists of the basic visual elements, the composition, and the medium. These three are the formal components of the message in visual communication. The content or meaning of this message consists of 1) the symbolic signs, 2) the representational or iconic signs, and 3) the concepts, sensations and emotions evoked and triggered by these signs and the formal components on the receiver (Dondis 1973/2006: 15, 105). I argue that, in AD, the subjectivity involved in the verbal representation of these concepts, sensations, and emotions is greater in comparison to the verbal representation of the iconic (there is a tree), symbolic (there is a heart/there is a sentence that reads…), and formal components (there is a green line crossing the surface). Therefore, the study of subjectivity in visual art AD should focus on the representation of this more subjective content, which is here referred to as “opinion.”
The field of “evaluation” (also termed “appraisal” and “stance”) within linguistics is concerned with the study of the linguistic expression of opinion. Bednarek’s (2009: 148) approach connects linguistic and psychological studies of evaluation and proposes a model based on a comparative analysis of these two domains. Bednarek draws on the cognitive dimensions of evaluation advanced by Ellsworth and Scherer (2003) as being common to most appraisal theories that are active in the field of Psychology and relates them to linguistic dimensions of evaluation collected from previous linguistic studies. The result is a “classification of ‘opinion lexis’ that is cognitively motivated” and “takes its input both from authentic discourse data and from existing theories of evaluation, trying to offer a cognitive basis for differentiating evaluative dimensions” (Bednarek 2009: 161). This classification is used in the study presented here to identify and categorize subjectivity in visual art AD. It consists of the following categories:
Expectedness: How expected?
Emotivity: How good/bad? (and how necessary?)
Importance: How important?
Authenticity: How real/true?
Power: How easy to deal with? (and how comprehensible/complex? how possible? how able?)
Reliability: How likely?
Causality: What are the reasons/causes/consequences?
These categories are illustrated with excerpts from the corpus of audio descriptive guides analyzed for this study (see Appendix 2, opinion lexis in bold), except for Authenticity, which had no activation in the corpus. During the corpus analysis, the definition of three of these categories was adapted to our object of study. Thus, Causality refers to opinions about the artist’s technique (1), or the concepts evoked by the iconic and symbolic signs as well as by the formal components of the work (2-3). When the opinion expressed is related to the emotions or sensations felt by the viewer or the subject represented in the work (4-5), or conveys an evaluation of artistic quality (6) or aesthetic value (7), it is classified as Emotivity. Lastly, Reliability refers to expressions of (un)certainty in the categorization of elements of the real world represented in the work (8), except for emotional states (which are categorized under Emotivity).
This type of product-oriented study of intersemiotic translation aims to describe the connection between the components of the visual, non-verbal source text and the aural, verbal target text. To achieve this goal, the elements of the work being described by means of opinion lexis are classified into the categories of a taxonomy of the components of visual communication, based on the work by Dondis (1973/2006) and Fichner-Rathus (2014). The latter also identifies two levels in visual communication: content and form; and includes technique, materials (or medium), and style within the second one, along with the basic visual elements and composition (Fichner-Rathus 2014: 11-19). The proposed taxonomy is shown in Table 1 (categories used for the analysis in italics).
The formal components are defined as follows, based on the proposals by Dondis (1973/2006), Fichner-Rathus (2014), and Leborg (2006) (see Appendix 3 for examples from the corpus).
Technique: methods used to manipulate and create with the materials.
Materials: physical elements used to create the artwork.
Style: synthesis of medium (material and technique), basic visual elements, composition, and purpose.
Composition: orchestration of the basic visual elements with a specific communicative and expressive purpose, following certain techniques of visual communication. The two basic techniques of visual communication are contrast and harmony, from which derive several sub-techniques expressed as opposite pairs, such as horizontality-verticality, balance-unbalance, and sequential-aleatory.
Dot: a place without area.
Line: a succession of adjacent dots.
Shape: a differentiated two – or three-dimensional area.
Space: the three-dimensional space of the real world and the implicit space created by visual techniques to simulate depth in two-dimensional media.
Dimension: size, scale and proportion of the visual elements.
Movement: the movement of entities in the real world and in a video recording, the implicit movement (it is inferred from the image that its elements moved) and the illusory movement (the image triggers the sensation that its elements move).
Direction: the direction of a movement or a shape.
Texture: perceptible structure made of lines or objects.
Tone: shades of grey, contrast (luminosity) and value pattern (variation and disposition of light and darkness).
Color: visual perception of the different wavelengths.
4. A method to study AD in art museums
In the 1990s, corpus linguistics (CL) experienced a great expansion within various areas of Applied Linguistics and a research agenda was advanced for descriptive translation studies (DTS) that incorporated this innovative methodology. Corpus-based translation (CBT) has never stopped expanding since then and it has become an important and thorough methodology in translation studies (Laviosa 2011: 14). This study applies corpus analysis and contextual analysis to examine subjectivity in audio descriptive guides of art museums. This combined approach was proposed in a recent study on subjectivity in interlingual translation that analyzed a multilingual corpus along with contextual information collected through bibliographical reviews and interviews (House 2011). The following sections outline a proposal for corpus compilation and analysis and for studying the communicative context of this translation modality.
4.1. Corpus compilation and analysis
A recurrent problem within corpus translation studies (CTS) is the relation between balance and comparability on the one hand and representativeness on the other (Laviosa 2011: 19). To resolve this issue, it is vital to offer a detailed description of both the population and the sample that allows researchers to consider the results of the analysis in context. In some cases, the audio descriptive guide was available on the museum’s web site or via an app for mobile devices, but in other cases, I had to request it. There were varied responses from the museums and companies holding the copyright; while some sent the complete audio descriptive guide script, others only sent a percentage of the audio descriptions and a few did not respond. Therefore, not all the museums with an audio descriptive guide could be included in the study. Notwithstanding, I was able to compile a multimodal corpus consisting of the complete or partial audio descriptive guides (audio and/or script) for the permanent collections or specific temporary exhibitions at 14 art museums in France, Spain, the UK, and the USA as well as the images of the works of art (Table 2).
In addition to the ones listed here, the Musée de l’Orangerie in Paris has an audio descriptive guide available for visitors at the museum, but it was not possible to have access to the script or the audio files. Estudios Durero, a company specializing in tactile images, in collaboration with the Spanish Organization of Blind People (ONCE), has developed audio descriptive guides for the Museo de Bellas Artes in Bilbao and the Museo Nacional del Prado, which include directions to explore tactile images of the described works. These audio guides have not been included in this study as they are considered a subgenre of the audio descriptive guide with distinct characteristics. The Spanish branch of the Guggenheim museum has an audio descriptive guide for two sculptures located outside the museum and the building, but it was not included in this study because it focuses on an audio description of two-dimensional art. Art Education for the Blind, a company based in New York that specializes in access to visual art for blind people, has published an encyclopaedia of art history with audio descriptions of works from the Brooklyn Museum and the Whitney Museum of American Art, among others. These have not been included in this study, given that they are educational materials that are not available at the museum, and thus I consider them a subgenre of visual art audio description.
The audio descriptive guide consists of general content, which can include a welcome and a farewell, mobile device features and user instructions, descriptions of the museum and the exhibition space, general mobility directions as well as content related to each work of art included in the guide. The latter includes the identification (title, year, dimensions, materials) and description of the artwork as well as its contextualization and expert interpretation (Soler Gallego 2012, 2016). This type of content dedicated to one specific exhibit is what existing guidelines and other publications refer to as “verbal description” and “audio description” in the museum context. From a linguistic perspective, however, the exhibit’s detailed and vivid description intended for visually impaired visitors has been added to an existing genre, namely the audio tour for museums. This can be defined as the oral, mobile version of exhibition labels (Serrell 2015) or the autonomous version of educational gallery tours (Kai-Kee 2011). For this reason, I use the term “audio description” to refer exclusively to this descriptive content addressed to visually impaired visitors and resulting from an intersemiotic translation process.
In a previous study (Soler Gallego 2018), it was observed that the artistic style could have an impact on the audio description. For this reason, I created subcorpora for three levels of abstraction: Abstract, Semi-abstract, and Representational. These three together make up a corpus consisting of the script or transcription of the section of the audio descriptive guide that deals with a specific piece of art. Since for some museums I could only access some sections of the audio descriptive guide and, to keep a certain balance between museums and styles, I selected 14 works for each subcorpus and two works from each museum featured in the subcorpus. The general statistics of this corpus are shown in Table 3. It is important to clarify that the length of the descriptions varies considerably between museums and companies so their weight in the corpus also varies. The weight of the different countries and museums also differs in the three subcorpora, according to the type of art the museum specializes in. However, it is not the goal of this study, although it could be the focus of future research, to analyze the statistical correlation between these contextual factors and subjectivity.
Using the qualitative data analysis software Atlas.ti, customized codes (Table 4) were created and assigned to text segments. These segments were later retrieved through various analysis functions. To facilitate the corpus analysis, each section of the audio descriptive guide that deals with a specific artwork was saved to a separate text file and then imported to groups of documents in Atlas.ti. The grouping of the individual files allows the creation of subcorpora for different variables such as museum, country and style.
In order to identify the audio descriptive content, the texts were first coded for rhetorical moves. For this purpose, I used the categories of rhetorical moves for audio descriptive guides for art museums identified in a previous study (Soler Gallego 2016), which were transformed into codes in Atlas.ti. Once the coding process was complete, the “Code Document Table” function within the “Analysis” option was used to determine the word count for the “Description of the Artwork” move, which amounts to 4,365 tokens for the Abstract subcorpus, 4,849 tokens for the Semi-abstract subcorpus and 4,764 tokens for the Representational subcorpus. At this point, it is important to make a distinction between the “Description of the Artwork” and the “Interpretation of the Work” moves. In the museum context, the term interpretation has several meanings. When used in the context of curatorial work at the museum, it refers to “how scholars tease out the meaning of an object or artwork as the result of research” (Wells, Butler, et al. 2016: 38). In addition, it is used more broadly to refer to “education, visitor experiences, and informal teaching and learning in museums” (Wells, Butler, et al. 2016: 38). In this study, the “Interpretation of the Work” move is related to the first of these definitions and is thus applied to textual segments that 1) offer an interpretation of the meaning of the artwork based on reference materials or expert knowledge of art history and 2) inform receivers – more or less explicitly – of the expert source of this interpretation (see Appendix 4 for examples from the corpus, interpretive content in bold). This study is not concerned with this type of interpretive content since it is not unique to audio descriptive guides for visually impaired visitors. This study is concerned with interpretations of the meaning of the work that refer to specific visual components and that are not attributed to expert sources. Thus, it can be assumed that their primary function is to help visually impaired visitors create a mental image of the work. These interpretations were coded as “Description of the Artwork” to be analyzed for subjectivity.
As mentioned before, this study focuses on the Focus of Attention operation within the Attention construal dimension and, more specifically, on the “opinion” component of visual communication. To study this component, the codes in the Opinion group were used to analyze the ADs, based on lexicographic definitions in different monolingual dictionaries and the contextualized analysis of the text segment (Figure 1). Next, this quotation was embedded into a larger one coded with the component of visual communication being described. If two or more non-consecutive text segments referred to the same component, each segment was treated as a separate quotation and was assigned an Opinion code, and all of them were then enclosed within a single quotation indicating the component being described. The goal was to be able to later retrieve the relative word count (percentage of the AD) for each Opinion code and the frequency of the co-occurrence of these categories as well as the Visual Communication categories.
Once the coding process was complete, the “Code Document Table” function within the “Analysis” option was used to determine the relative word count for the different Opinion categories, that is to say the percentage of the “Description of the Artwork” move that was categorized under each of them. This operation was completed for the whole corpus and for the three subcorpora. Next, the “Quotations Enclosing Quotations” filter in the “Quotations Manager” option was used to quantify the co-occurrences of the Opinion categories with the categories of the components of the Visual Communication for the whole corpus and for each subcorpus. The Visual Communication codes group was specified as the enclosing quotation, while the different codes within the Opinion group were specified as the enclosed quotations, one at a time. In addition, a filter was added to retrieve data for each level of abstraction (Figure 2). This filter retrieved a list of enclosing quotations along with their code and file for the specified category and subcorpus. The list was exported to a spreadsheet, where the occurrences of each category were quantified.
Through this analysis, it was possible to determine the number of subjective statements in the analyzed corpus and to classify them into different categories. In addition, the connections between these categories and the components of visual communication were quantified for the whole corpus and for each subcorpus.
4.2. Contextual analysis
In order to describe the context in which visual art ADs are created and used, it is necessary to collect information from two sources: documents and individuals. In my studies of museum AD, the main documentary resources used are the museum’s website and AD guidelines published by different organizations. In addition, I interview the accessibility coordinators and educators employed by the museums as well as the freelance describers and companies involved in the creation of the audio descriptive guides. Whenever a face-to-face or online interview was not feasible, I administered a questionnaire. In this study, I was able to compile information through interviews and/or questionnaires from all the aforementioned museums except for two.
From the full set of questions included in both the questionnaire and the interview script for museums, freelance describers, and companies that I used in my research on museum AD, only those questions that were relevant to the study of subjectivity are included below, since they are the ones that were analyzed for this study. The full set of questions was adapted to each participant and to the order in which individuals involved in the same project were interviewed. This means that the questions that were already answered by one individual and did not enquire about the interviewee’s opinion were not put to the other individuals in the project. Both the questionnaire and the interview script began with questions that aimed to collect general data about the creation of the audio descriptive guide. These were followed by questions that focused on specific features of the ADs that were identified as relevant in existing guidelines, previous research in AD, and corpus analyses of visual art AD.
Who created the audio descriptive guide for visually impaired visitors for the museum? What is the professional profile of the audio describer? Was s/he trained in accessibility and audio description? From whom did s/he receive this training?
What were the main stages in the development of the audio descriptive guide for visually impaired visitors?
Did the audio describer follow any set of guidelines to create the audio descriptions? Please specify.
If the audio description was created by an external company or freelance audio describer, how was the collaboration between the museum and the company or the freelance audio describer? Did the museum give any directions to the company or audio describer to create the audio descriptive guide? Please specify. Did the museum review the content created by the company or audio describer?
Was the audio descriptive guide assessed by consultants with a visual impairment during the creation process?
Has the audio guide for visually impaired visitors been evaluated by users? What were the results?
Is including both contextual and interpretive information about the work and the audio description of the work in the same audio track, instead of including these two types of information in separate audio tracks, beneficial for visually impaired visitors?
Should the language used to audio describe works of art for visually impaired visitors be as objective as possible?
A set of codes was created in Atlas.ti to classify the answers to the questions in the audio recordings of the interviews and a “Comment” with a transcription of the answer was added to each quotation (Figure 3). The quotations for each code were retrieved by means of the “Quotations Manager” option, along with their code, file name and comment. Questionnaires were administered through an online service which included a function to export questions and answers to a spreadsheet which was combined with the one generated by Atlas.ti. The combined discussion of results from the corpus-based and contextual analyses allowed us to formulate explanatory hypotheses for the linguistic features observed in the ADs.
5. Results and discussion
The method described in the previous sections was followed so as to analyze both the corpus of visual art AD for art museums and the context in which these ADs were produced. The following sections include a summary of these results and a conjoint discussion of the findings.
5.1. Corpus analysis
The results of the analysis described in the previous section show that the Opinion categories are activated in the corpus and they amount to 15.03% of the word count in the “Description of the Artwork” move. The Opinion categories amount to 14.6%, 19.9% and 10.5% in the Abstract, Semi-abstract and Representational subcorpora respectively (Figure 4). Among the Opinion categories, Causality (18.5%) and Emotivity (18.2%) show the highest activations, followed by Reliability (7.43%). The Authenticity category is not activated in the analyzed corpus and Importance (0.46%), Power (0.37%) and Expectedness (0.04%) show very low rates as compared to the other categories. Causality shows the highest activation rate in the Semi-abstract and Representational subcorpora, while Emotivity occupies the first place in the Abstract subcorpus (Figure 5). Overall, these results indicate that describers include subjective content in the ADs for art museums, regardless of the artistic style of the source text. In addition, these data indicate that the type of subjective content varies according to the level of abstraction of the work.
The next step in the analysis was to determine the co-occurrence of all the Opinion categories as a group and the Visual Communication categories (Figure 6). The result is a visual representation of the connection between the components of the visual, non-verbal source text (the work of art exhibited at the museum) and the aural, verbal target text (the audio descriptions included in the audio descriptive guide). The Opinion categories mainly co-occur with the Icon code and, to a lesser extent, with Composition and two other visual elements, namely Shape and Line. The analysis then proceeded to study co-occurrence for the three subcorpora. Within the Abstract subcorpus, the Opinion categories co-occur more frequently with Composition and, to a lesser extent, with Shape, Line, Icon, and Color. In the Semi-abstract and Representational subcorpora, however, the Opinion category mainly co-occurs with the Icon category. For semi-abstract works, the Opinion category also co-occurs, although less frequently, with Shape, Composition, Color and Line. Overall, these results suggest that subjective content is more frequent in the intersemiotic translation of certain visual components, namely the iconic signs (elements of reality represented in the work) and the composition (orchestration of the visual elements and visual techniques used by the artist). The specific visual component of the work of art that the subjective content refers to varies according to the level of abstraction of the work.
The final step was to analyze the co-occurrence of each Opinion category and the Visual Communication categories (Figure 7). In the Representational subcorpus, both Emotivity and Causality are mainly related to the Icon category. In the Semi-abstract subcorpus, the Icon category is mainly connected to Emotivity and Reliability. The other formal components (Shape, Color, Composition, Line, Space, and Dot) show a frequency that is similar to that of the Icon category, while in the Representational subcorpus the Icon category occupies a clearly predominant position. In the Abstract subcorpus, Emotivity and Causality are once again the most frequent Opinion categories. However, unlike in the Representational subcorpus, the activated Visual Communication categories are not iconic signs, but the formal components of Composition in the first place, followed by Shape, Line, Color, Space, Materials, and Texture. Overall, these results indicate that Emotivity is strongly activated in the three subcorpora, but the visual components to which it is connected vary depending on the level of abstraction. While in descriptions of representational art, the Emotivity category almost exclusively refers to icons, for semi-abstract art, the formal components are equally activated and in the corpus of abstract art, the formal components are the only visual components activated for this type of subjective content.
The previous quantitative analysis was complemented with a qualitative analysis of the quotations allowing us to better understand and illustrate the inclusion of subjective content in the ADs. Starting with abstract works, the Opinion category refers to concepts evoked by the composition (see Appendix 5, Example 1) as well as to feelings and sensations triggered by this formal component (2-3). For semi-abstract works, the Opinion category is related to the identification – with various degrees of certainty – of the referents in the real world of iconic signs (4) and different formal components (5-6), concepts evoked by iconic signs (7), and emotions and sensations felt by the figures represented in the work (8) or triggered by the formal components in the viewers (9-10). Finally, in representational works, the Opinion category refers to human facial expressions (11), the aesthetic value of the iconic signs (12), the judgement of the artist’s technical skills (13), and the significance of the iconic signs in the scene represented in the work (14) or its socio-cultural context (15).
5.2. Contextual analysis
There were varying degrees of participation from museums, companies and freelance describers related to the audio descriptive guides included in the corpus. For the GP, MNCARS, NYG, and NYMoMA museums, both museum staff and the describer participated in the study. Describers for the MNCARS and the NYG are members of the department in charge of education and accessibility at the museum. In the case of the MFA, TM and MNP, only the museum’s representative participated in the study, while for the MDO, CIM, and RAA, only the describer participated in the study. The audio descriptive guide for the MCT was created by a company specializing in accessible multimedia mobile devices and they, in turn, hired an audio description company. Both companies participated in the study. Unfortunately, no data could be collected by means of an interview or questionnaire for the NGA, CP or MJR. Overall, contextual data were collected in this way for eleven out of the fourteen museums represented in the corpus.
This section offers a discussion of the results from the corpus analysis and the results of the contextual analysis that are more relevant to better understanding current practices regarding subjectivity in ADs for art museums. First, there is a lack of coherence between the statements of the interviewed participants regarding the inclusion of subjective content in the ADs and the textual samples. For five of the fourteen museums, participants declared that the language used in ADs should be as objective as possible and for one more museum it was determined that objectivist guidelines had been followed to create the ADs. However, the results of the corpus analysis show that subjective language was used in the ADs created for all the museums studied. This dissonance could be due to participants having different conceptions of what subjective language is. Guidelines are not sufficiently clear and precise in their definition and illustration of subjective content and this lack of conceptual definition also affects some studies in this field. Studies, such as the one presented in this paper, can help clarify this concept for practitioners, trainers, and researchers alike.
Two elements of the context that could be especially relevant to the matter of subjectivity are authorship and AD guidelines. Audio descriptive guides are created either by the museum or by external translators. These external professionals are specialists in editing, creative writing and audiovisual communication, while the museum staff in charge of creating ADs are experts in art history and/or art education. Most of them have been trained in accessibility and audio description by more experienced professionals. Among the published guidelines, two of them only address film AD (Morisset and Gonant 2008 in France; ITC in the UK), some offer general recommendations on film, dance, theatre and museum AD (ADC 2008 and Snyder 2010 in the USA; AENOR 2005 in Spain), while others specifically address the AD of displayed artwork (Salzhauer Axel, Hooper, et al. 2003 in the USA; RNIB and VocalEyes 2003 in the UK). Most of them recommend objectivity or neutrality in order to help visually impaired receivers build their own mental image and interpretation of what is being described without being influenced by the translator’s subjective impressions. The only exception is the publication by the RNIB and VocalEyes, which includes comments by visually impaired participants supporting more subjective descriptions. Those focusing on works of art state that ADs should be detailed and focus on subject matter, composition, technique, materials, style, and color.
This defense of subjective ADs is also found in the interviews with the describers for the CIM and the RAA as well as with the accessibility coordinator at the TM. The CIM and TM are among the seven museums in the corpus with a percentage of Opinion content at 20% or higher, but interestingly, in the RAA, this percentage drops to 11.8%. This could be related to the representational style (realistic portrait) of the audio described works, which contrast with the more abstract ones at the other two museums. The describers for these three British museums are external professionals, have a background in linguistics, literature, editing and creative writing, and have been trained in AD. The describer for the RAA followed guidelines (which could not be obtained for this study) created by Antenna International, the company hired for the project, and the same may have been the case for the TM, but this information could not be confirmed. Meanwhile, the describer for the CIM stated that there were no guidelines. A somewhat similar context is found at the MDO and GP in France, with 18.3% and 21.3% Opinion content respectively. The describer for these two museums is also an external professional, but he has a background in history and sociology as well as professional experience in accessibility. He has a visual impairment and partners with a sighted describer. He did not follow any set of guidelines and considers it important to find a balance between offering subjective descriptions and, at the same time, giving receivers the freedom to create their own subjective interpretation of the visual components. No information could be collected regarding the authorship of the ADs for the CP, whose ADs contain 20.6% Opinion content.
Intriguingly, the MCT is the museum with the highest percentage of Opinion content (29.9%). The describer followed the Spanish AENOR guidelines, which recommend objectivity, and expressed support for this method. However, it is the only museum in the study that follows a universal accessibility approach. The aim is for visitors with and without a visual impairment to have a similar experience at the museum. The company in charge of the project created the text for adult visitors along with the museum’s team of art educators. This text was sent to a company specializing in AD which added a limited amount of audio descriptive information to the text, following the requirements established by the first company. This means that any subjective content used for the general audio guide remained in the audio descriptive guide, which could explain this high percentage. The percentage of the Opinion category is lower for the other Spanish museums: MNCARS (13.2%), MJRT (10.8%), and MNP (5.7%). The describers of the last two were external professionals with experience in accessibility and audio description and followed the Spanish guidelines. In contrast, the three guides for the MNCARS were created by art historians and educators working for the museum and they followed both the Spanish guidelines and those of Art Beyond Sight. One describer for the MNCARS defends the complementarity of objective and subjective content, while the other describer and the MNP representative are of the opinion that subjectivity should be avoided in general. Besides, the describers for the MJRT and the MNP worked for the two organisations that drafted the two sets of Spanish guidelines. Overall, it seems that the use of objectivist guidelines could be related to lower Opinion percentages for these museums.
Among American museums, the percentage of the Opinion category varies considerably. The describers and museum staff interviewed for the NYG, NY MoMA and MFA stated that Art Beyond Sight’s guidelines had been followed and/or that ADs should follow the objectivist maxim. Despite this apparent agreement, the Opinion content for these museums amounts to 23.3%, 20% and 5.2% respectively. The NGA is somewhat in the middle at 10%, but no contextual data could be collected for this museum. The describer for the NY MoMA is an external professional with a background in creative writing, while those for the NYG and the MFA are educators and accessibility experts at the museum. These data suggest that despite supporting the objectivist approach, in practice some museums create ADs with higher levels of subjective content compared to other museums in the corpus. One factor that could be related to these numbers is that both describers for the NY MoMA and the NYG were trained by Art Beyond Sight. It is possible that formal training aids in comprehending the role of, as well as determining the appropriate levels of, subjectivity in visual art AD.
Another contextual factor that could have an impact on subjectivity is the AD creation process. It was determined that in five out of the fourteen museums (MNCARS, CIM, RAA, MoMA, and MFA), one visually impaired consultant or a group of them, reviewed the script before it was made available to visitors. No consultants were involved in the creation of the ADs for the MCT, GP, MDO and MNP, although the describers for the last three have a visual impairment and a wide range of experience in accessibility. Given the limited number of consultants involved and the great variety of individual profiles among visually impaired people, it is possible that a small number of individual preferences regarding subjectivity in the ADs are being taken into consideration, while other opinions remain unrepresented. Besides, the only two museums that carried out evaluations once the guide was available for visitors are the MCT and MNCARS. Results of the former could not be accessed and those of the latter are not relevant to this study, since most of the respondents were visitors without a visual impairment. Some museums mentioned their interest and plans to carry out formal evaluations of the audio descriptive guides and some may have taken place since then. Overall, the lack of formative and summative evaluations of audio description resources makes it difficult to assess existing practices in general and the effect of subjectivity on visitors’ experience and understanding.
Audio description in art museums is a modality of intersemiotic and multimodal translation. As such, it is fundamental that descriptive studies of this translation phenomenon analyze the connection between the visual, non-verbal source text and the verbal target text. Qualitative Data Analysis software, like the one used for this study, is a useful tool to achieve this goal. It allows ADs to be annotated at various levels related both to the visual components of the artwork and the linguistic elements of the AD and for these layers of information to be analyzed conjointly. In this study, the focus of attention on the Opinion category within the visual components is considered an indicator of subjectivity in this intersemiotic translation modality. Results have shown that subjectivity is found in the analyzed corpus and both the level of abstraction of the work and the context of communication seem to have an impact on the subjectivity level. The descriptive results and explanatory hypothesis presented in this article are a useful source of data to design reception studies, which are necessary to determine the effects of subjective descriptions on visually impaired people’s experience of the visual arts and the museum event.
Appendix 1. References for audio description guidelines
ADC (Audio Description Coalition) (2008): Standards for audio description and code of professional conduct for describers. Visited 29 September 2019, <https://audiodescriptionsolutions.com/the-standards/download-the-standards/>.
AENOR (2005): Norma UNE: 153020. Audiodescripción para personas con discapacidad visual. Requisitos para la audiodescripción y elaboración de audioguías. Visited 1 October 2019, <https://www.aenor.com/normas-y-libros/buscador-de-normas/UNE?c=N0032787>.
ITC (Independent Television Committee) (2000): ITC guidance on standards for audio description. Visited 3 September 2019, <http://msradio.huji.ac.il/narration.doc>.
Morisset, Laure and Gonant, Frédéric (2008): L’audiodescription: principes et orientations. Visited 1 December 2019, <https://www.csa.fr/Media/Files/Espace-Juridique/Chartes/Charte-de-l-audiodescription>.
RNIB and VocalEyes (2003): Talking images guide. Museums, galleries and heritage site: Improving access for blind and partially sighted people. Visited 2 December, 2019, <http://audiodescription.co.uk/uploads/general/Talking_Images_Guide_-_PDF_File_5.pdf>.
Salzhauer Axel, Elizabeth, Hooper, Virginia, Kardoulias, Teresa, et al. (2003): AEB’s guidelines for verbal description. In: Elizabeth Salzhauer Axel and Nina Sobol Levent, eds. Art Beyond Sight: A Resource Guide to Art, Creativity, and Visual Impairment. New York: AFB Press, 229-237.
Snyder, Joel, ed. (2010): Guidelines for audio description standards. Visited 3 December, 2019, <http://www.acb.org/adp/ad.html>.
Appendix 2. Examples for each category within the classification of subjectivity in visual art AD
Many colors look as if they were thinly applied to the canvas…
His arms are crossed in a casual fashion and he glances up to his right, the viewer’s left, as if looking directly at the upper left corner of the canvas.
The two figures represent the Moon and Earth of the painting’s title.
This sense of rhythm is heightened by evenly spaced patches of yellow and blue – and smaller blotches of red, purple and other colors.
Her face is shown in full profile, her mouth expressionless…
Watteau’s use of red chalk is equally eloquent.
His body is slender and boyish…
Three versions of a black boy of perhaps ten or twelve years of age are spread out across the page.
On note la présence incongrue d’un chien de chasse, racé, qui se tient au premier plan.
We notice the incongruous presence of a pedigree dog standing in the foreground.
Of particular significance is a dark red narrow line that begins just below the figure’s waist, runs up along the entire contour of her back and over her bony left shoulder, which juts conspicuously upwards.
El Guernica es una obra de gran complejidad, no sólo en términos de representación sino también en los aspectos interpretativos.
Guernica is a work of great complexity, not only in terms of representation but also in its interpretive aspects.
Appendix 3. Examples illustrating the formal components of visual communication
Technique: It is a detailed etching showing a narrow road or track into Wivenhoe across fields of tall grass.
Material: The final zip is tan, and almost looks like raw canvas. But closer examination shows that it too is paint.
Style: This is an abstract expressionist painting.
Composition: In front of her, another woman sits or squats, elbow on one raised knee, which juts toward the center of the painting.
Dot: Little dabs and dashes of paint seem to dance over the surface of the canvas – white, red, blue, deep purple, pink and green.
Line: This expanse of color is interrupted by five thin vertical stripes. Each is an inch or two wide, but no two are exactly alike.
Shape: In the center is an open cylinder representing the sound hole; it is about three inches wide and protrudes out about five inches.
Space: There is only the slight illusion of depth created by their circle of joined hands.
Dimension: That is due partly to the subdued lighting – as requested by the artist Mark Rothko – and partly to the fact that the paintings are huge – and, since they are without glass, sound-absorbent.
Movement: Little dabs and dashes of paint seem to dance over the surface of the canvas – white, red, blue, deep purple, pink and green.
Direction: The next pole, in the back corner of the platform to your right, is about 8 inches wide and shaped like a surfboard, tapering at both ends. It faces in, toward the center pole.
Texture: Knowing that the work is made of steel, you might imagine it being smooth, shiny and reflective. In fact it is the opposite. It is dark grey and the surface is dull and appears rusty and weathered.
Tone: The shadows are somewhat heavier across the left side of his torso and along his left shoulder and upper arm.
Color: The upper right corner is an especially bright and sunny yellow, as if it was an extension of the landscape at sunset.
Appendix 4. The “Interpretation of the Work” move
Interpretation of the work
Reference materials: Explicit
But he [Rothko] didn’t want his work interpreted intellectually. “I am not an abstract painter. I am not interested in the relationship between form and colour.”
But Watteau’s studies are respectful of his subject. Exhibition co-curator Pierre Rosenberg: “So here we see Watteau’s mastery of his technique, but also his quality of not looking down on his models.”
Reference materials: Implicit
Kandinsky wanted his paintings to capture a mood, very much in the same way that music does.
Mais on pense aussi à la calligraphie chinoise tracée au pinceau… Pour cette série, Brice Marden s’inspire des écritures de Han Shan, poète chinois du IXème siècle…
But it also makes us think of Chinese calligraphy made with a brush… For this series, Brice Marden drew his inspiration from the writings of Han Shan, a 9th century Chinese poet…
Knowledge of art history
…On pense aux fameux drippings de Jackson Pollock, où l’artiste laisse goutter (drippings en anglais) la peinture sur une surface plane.
… It makes us think of Jackson Pollock’s famous drippings, in which the artist dripped painting on a flat surface.
Appendix 5. Co-occurrence of the Opinion and Visual Communication categories
The overall look resembles a vigorously brushed watercolor… (Kandinsky, Improvisation 31, National Gallery of Art, Washington)
But mainly it’s because of the quietly throbbing energy that seems to exude from their deep maroons, wines, reds and blacks… (Rothko, Seagram Murals, Tate Modern)
A purple swath, highlighted with bright blue and orange strokes, draws the eye along the left side of the canvas in a thin column. (Lewis, Untitled, Museum of Fine Arts, Boston)
Otro niño, presa del pánico, se aferra por detrás a su madre, la cual parece estar embarazada, a juzgar por el abombamiento de su vientre. (Picasso, Masacre en Corea, MNCARS)
Another child, gripped by panic, clings to his mother from behind, who seems to be pregnant, judging by her swollen stomach.
The colors are subdued, perhaps it is dusk. (Gauguin, The Moon and the Earth, MoMA, New York)
…a blotched greeny-grey patterns the background, suggesting both crumbling paintwork and ghostly reflections… (Thornton, Roman Pots, Colchester and Ipswich Museums)
Esta lámpara… podría ser una alusión a las bombas incendiarias lanzadas por los aviones alemanes. (Picasso, Guernica, MNCARS)
This lamp… may be an allusion to the incendiary bombs that were launched by German airplanes.
She is situated toward the right side of the canvas with what seems to be the weight of the world on her shoulders. (Picasso, Woman ironing, Guggenheim, New York)
The weak wash of colour gives them an almost ghostly appearance… (Thornton, Roman Pots, Colchester and Ipswich Museums)
Le tableau, de petit format, 80 cm de hauteur et 64 cm de largeur, surprend par les contrastes violents de couleurs… (Delaunay, Jeune Finlandaise, Centre Pompidou)
The small scale painting (31” by 25”) is surprising due to the violently contrasting colors…
Le premier, attentif, tient le seau d’eau bénite et le goupillon tandis que l’autre, dissipé, s’intéresse à un autre spectacle. (Courbet, Un enterrement à Ornans, Musée d’Orsay)
The first child, intent on his task, holds the container of holy water and an aspergillum; the second, distracted, is more interested in other sights.
Sus facciones suaves y delicadas están marcadas por una nariz fina y una pequeña boca. (Romero de Torres, Ángeles y Fuensanta, Museo Julio Romero de Torres)
Her soft and delicate features are highlighted by her narrow nose and diminutive mouth.
L’expression de la surprise est magistralement rendue, avec l’expression des visages et les gestes interrompus. (Velázquez, La Forge du Vulcain, Grand Palais)
This expression of surprise is masterfully rendered, with facial expressions and gestures frozen in time.
… en la parte inferior se amontonan varios cadáveres de fusilados. Nadie se ha molestado en retirarlos antes de continuar con la macabra condena. (Goya, El tres de mayo de 1808, Museo Nacional del Prado)
… bullet-riddled bodies lie in a heap below. Nobody has bothered to remove them before continuing with the grisly punishment.
La pénibilité de la glane qui oblige à travailler le dos cassé, ressort nettement… (Millet, Les Glaneuses, Musée d’Orsay)
The onerousness of the act of gleaning which obliges the worker to work bent over double is evident…
I would like to express our most sincere gratitude to the museums, companies, and freelance professionals that have generously participated in this study and to Maria Olalla Luque Colmenero and Nuria Cabezas Gay, my dear colleagues at the TRACCE research team at the University of Granada, who compiled some of the corpus materials. This study was funded by the Faculty Development Award, granted by the College of Liberal Arts of Colorado State University.
Mörsch, Carmen and Chrusciel, Anna, eds. (for Pro Helvetia): Time for cultural mediation. Zurich: Institute for Art Education, Zurich University of the Arts. Consulted on December 1. 2019, <https://prohelvetia.ch/app/uploads/2017/09/tfcm_0_complete_publication.pdf>.
Visual Thinking Strategies (Last update: 2. March 2019): Research and Theory. Visual Thinking Strategies. Consulted on December 29, 2019, <https://vtshome.org/research>.
ATLAS.ti Scientific Software Development GmbH (4 December 2017): Atlas.ti. Version 8.1. Consulted on September 4, 2019, < https://atlasti.com/>.
- Asensio, Mikel, Asenjo, Elena, Castro, Yone, et al. (2014): Evaluación implicativa: hacia una visión generativa y participativa en la gestión de audiencias. In: Iñaki Arrieta, ed. La sociedad ante los museos. Públicos, usuarios y comunidades locales. Bilbao: Universidad del País Vasco, 97-137.
- Bednarek, Monika (2009): Dimensions of evaluation. Cognitive and linguistic perspectives. Pragmatics and Cognition. 7(1):146-175.
- Croft, William and Cruse, D. Alan (2004): Cognitive Linguistics. Cambridge: Cambridge University Press.
- Dondis, Donis (1973/2006): La sintaxis de la imagen. Introducción al alfabeto visual. (Translated from English by Justo González Beramendi) Barcelona: Gustavo Gili.
- Eisner, Elliot W. (2002): The Arts and the Creation of Mind. New Haven: Yale University Press.
- Ellsworth, Phoebe C. and Scherer, Klaus R. (2003): Appraisal processes in emotion. In: Richard J. Davidson, Klaus R. Scherer, and H. Hill Goldsmith, eds. Handbook of Affective Sciences. Oxford: Oxford University Press, 572-595.
- Fichner-Rathus, Lois (2014): Foundations of Art and Design. Boston: Wadsworth.
- Gottlieb, Henrik (2005): Multidimensional translation: Semantics turned semiotics. In: Heidrun Gerzymisch-Arbogast and Sandra Nauert, eds. MuTra – Multidimensional Translation Conference Proceedings. (MuTra: Challenges of Multidimensional Translation, Saarbrücken, May 2-6, 2005). Visited 5 May 2020, https://www.euroconferences.info/proceedings/2005_Proceedings/2005_Gottlieb_Henrik.pdf.
- Halverson, Sandra (2013): Implications of cognitive linguistics for translation studies. In: Ana Rojo and Iraide Ibarretxe Antuñano, eds. Cognitive Linguistics and Translation: Advances in Some Theoretical Models and Applications. Berlin: Walter de Gruyter, 33-73.
- House, Juliane (2011): Using translation and parallel text corpora to investigate the influence of Global English on textual norms in other languages. In: Alet Kruger, Kim Wallmach, and Jeremy Munday, eds. Corpus-based Translation Studies. Research and Applications. London: Continuum, 187-208.
- Housen, Abigail (1992): Validating a measure of aesthetic development for museums and schools. ILVS Review: A Journal of Visitor Behavior. 2(2):213-237.
- Housen, Abigail (2000-2001): Voices of viewers: Iterative research, theory and practice. Arts and Learning Research Journal. 17(1):1-12.
- Housen, Abigail (2001-2002): Aesthetic thought, critical thinking and transfer. Arts and Learning Research Journal. 18(1):99-131.
- Hubard, Olga (2011): Rethinking critical thinking and its role in art museum education. The Journal of Aesthetic Education. 45(3):15-21.
- Jakobson, Roman (1959): On linguistic aspects of translation. In: Reuben A. Brower, ed. On Translation. Harvard: Harvard University Press, 67-77.
- Jiménez Hurtado, Catalina and Soler Gallego, Silvia (2013): Multimodality, translation and accessibility: A corpus-based study of audio description. Perspectives. 21(4):577-594.
- Jiménez Hurtado, Catalina, Seibel, Claudia and Rodríguez Domínguez, Ana, eds. (2010): Un corpus de cine. Teoría y práctica de la audiodescripción. Granada: Ediciones Tragacanto.
- Kai-Kee, Elliott (2011): A brief history of teaching in the art museum. In: Rika Burnham and Elliott Kai-Kee, eds. Teaching in the Art Museum. Interpretation as Experience. Los Angeles: J. Paul Getty Museum, 19-58.
- Langacker, Ronald W. (1987/2008): Cognitive Grammar: A Basic Introduction. Oxford: Oxford University Press.
- Laviosa, Sara (2011): Corpus-based translation studies: Where does it come from? Where is it going? In: Alet Kruger, Kim Wallmach, and Jeremy Munday, eds. Corpus-based Translation Studies. Research and Applications. London: Continuum, 13-32.
- Leborg, Christian (2006): Visual Grammar. New York: Princeton Architectural Press.
- Lima, Pedro H. and Magalhães, Célia M. (2013): A neutralidade em audiodescrições de pinturas: Resultados preliminares de um descrição via teoria da avaliatividade [Neutrality in audio descriptions of paintings: Preliminary results from a description based on Appraisal Theory]. In: Vera L. Santiago Araújo and Marisa Ferreira Aderaldo, eds. Os novos rumos da pesquisa em audiodescrição no Brasil [New directions in research on audio description in Brazil]. Curitiba: Editora CRV, 73-87.
- Luque Colmenero, Maria O. (2016): The embodiment of the metaphor: An analysis of the metaphors used to convey the human body in audio descriptive guides of museums for people with visual functional diversity. E-AESLA: Revista digital del Instituto Cervantes. 2:326-334.
- Mazur, Iwona and Chmiel, Agnieszka (2012): Audio description made to measure: Reflections on interpretation in AD based on the Pear Tree Project data. In: Aline Remael, Pilar Orero, and Mary Carroll, eds. Media for All: Audiovisual and Media Accessibility at the Crossroads. Amsterdam: Rodopi, 173-188.
- Ramos Caro, Marina (2016): Testing audio narration: The emotional impact of language in audio description. Perspectives. 24:1-29.
- Serrell, Beverly (2015): Exhibit Labels. An Interpretive Approach. New York: Rowman & Littlefield.
- Soler Gallego, Silvia (2012): Traducción y accesibilidad en el museo del siglo XXI. Granada: Ediciones Tragacanto.
- Soler Gallego, Silvia (2013): La traducción accesible en el espacio multimodal museográfico. Doctoral dissertation, unpublished. Cordoba: Universidad de Córdoba.
- Soler Gallego, Silvia (2016): A corpus-based genre analysis of art museum audio descriptive guides. In: Francisco Alonso Almeida, Laura Cruz García, and Víctor González Ruiz, eds. Corpus-based Studies on Language Varieties. Linguistic Insights. Bern: Peter Lang, 145-166.
- Soler Gallego, Silvia (2018): Audio descriptive guides in art museums: A corpus-based semantic analysis. Translation and Interpreting Studies. 13(2):230-249.
- Szarkowska, Agnieszka, Jankowska, Anna, Krejtz, Krzysztof, et al. (2016): Open Art: Designing accessible content in a multimedia guide app for visitors with and without sensory impairments. In: Ana Matamala and Pilar Orero, eds. Researching Audio Description. London: Palgrave, 301-320.
- Udo, John P. and Fels, Deborah I. (2009): Suit the action to the word, the word to the action: An unconventional approach to describing Shakespeare’s Hamlet. Journal of Visual Impairment and Blindness. 103(3):178-183.
- Udo, John P., Acevedo, Bertha, and Fels, Deborah I. (2010): Horatio audio-describes Shakespeare’s Hamlet: Blind and low-vision theatre-goers evaluate an unconventional audio description strategy. British Journal of Visual Impairment. 28(2):139-156.
- Ueno, Masamichi (2015): Democratic Education and the Public Sphere: Towards John Dewey’s Theory of Aesthetic Experience. London/New York: Routledge.
- Walczak, Agnieszka and Fryer, Louise (2017): Creative description: The impact of audio description style on presence in visually impaired audiences. British Journal of Visual Impairment. 35(1):6-7.
- Wells, Marcella, Butler, Barbara, and Koke, Judith (2016): Interpretive Planning for Museums. Integrating Visitor Perspectives in Decision Making. London/New York: Routledge.