AUDIO DESCRIPTION AND DIVERSITY AWARENESS: FLUTUA MUSIC VIDEO

Brazil has the highest murder rate of transgender people in the world. The paper will focus on a debate of how audio description of some audiovisual products should be used to call the attention to the discrimination and violence suffered by homosexual couples. We have chosen a Brazilian music video called Flutua produced and performed by Johnny Hooker with special participation of the Brazilian singer Liniker, a black trans woman. The clip presents an outstanding visual narrative involving contemporary themes such as gays with disability, gender-fluid and homophobia. During the video a gay deaf couple spends a day having fun with friends in an urban scenario composed by known streets and places of Sao Paulo capital. At the end of the day a member of the group suffered a very violent attack. The audio description of the most relevant scenes, the identities and outfit of the singers, and their movements will be presented having in mind the music as protagonist of Flutua.


INTRODUCTION
Some decades ago, a new social paradigm appeared which reconsidered or reinitiated the traditionally established social schemata. Under the concept of "post-structuralism", the ideas proposed by authors such as Derrida, Foucault, Barthes and Deleuze tried to put to the test what had been considered the truth. These authors believed language played an important role in the construction of identities since such construction is related to labels required to refer to what is around and within us. However, though necessary, language can be of little help when we construct sexual orientation and gender identities as "any act of identity affirmation requires a certain level of essentialisation, which is something that has been rejected in recent years by theories" (MARTÍN RUANO, 2018: Preface).
Despite being placed under the same umbrella, queer theories differ from feminist and homosexual theories, in that the queer is never based on a natural identity (man/woman), nor on a definition grounded in (homosexual/heterosexual) practices rather but in the multiplicity of bodies that are discovered when combatting the regimes that build them as "normal" or "abnormal" (PRECIADO, 2005, p.163). In this regard, we can say that language is unable to communicate reality, or at least, the potential reality of the multiplicity of existing identities. However, as paradoxical as it may sound, "identities are constructed through, and not outside, difference" (HALL, 1996, p. 4). How is it possible, first of all, to describe, or, more precisely, "correctly describe" an identity? And secondly, concerning our research interests, how can we translate identities?
Our paper will be focused on the presentation of some results of an audio description collective project produced by Mídia Acessível e Tradução Audiovisual Research Group. (Media Accessibility and Audiovisual Translation). We have chosen the music video Flutua as our study object since it is composed by people who represent the LGBTI community. During the video a gay deaf couple spend a day with friends using Brazilian sign language to communicate in a scenario of well-known streets in the city of São Paulo. Despite being deaf, the group dances together on the streets of São Paulo and inside a night club through to the end of the night when a member of the group suffers a very violent attack. We have chosen to audio describe important information such as the physical characteristics of one of the singers, who is a trans woman (Liniker), and also some gestures of the main characters in the narration, and, finally, certain movements where people are dancing in different locations in the city of São Paulo.
This article will first of all introduce the video clip audio described, Flutua, and will discuss how the audio introduction and audio description were issued. It will then move on to the discussion of transphobia and homophobia crimes and violence in Brazil, highlighting the importance of the message in this audiovisual content. Afterwards, some theories on the relevance on translation in the transference of ideas, stereotypes and representation of non-normative identities are discussed. Finally, the attention is put in some audio introduction and audio description excerpts of the music video, which lead to the conclusions of the present paper.

FLUTUA
Our study object was the Flutua music video, released on December 2017 on singer Johnny Hooker's YouTube channel. The video has an innovative style and has now been viewed more than 5,900,000 times. It tells the story of a gay deaf couple who use sign language and suffer prejudice and violence in the city of São Paulo.
Flutua's first stanza makes it clear that its goal is the search for trans respect, which is a daily battle for the LGBTI community in Brazil.
What will they say about us? Their parents, God, and such things When they hear the rumours of our love Baby, I'm tired of hiding When we are looking at each other, whispering to each other We are two men, nothing more They will not win Baby, nothing will be in vain Before this night ends Let us dance to our song And it floats, and it floats No one can or will tell us how to love.
Since the music video was released, the Brazilian LGBTI movement has used the verse "No one can or will tell us how to love" as a motto of liberation and gender identity affirmation. The Rede Globo owned pay television channel GNT used the song Flutua as their Summer 2017/2018 opening sequence. It was adapted into a one-minute commercial showing GNT presenters swimming in a pool to the song, and at the end of the commercial: "GNT Summer Mode" 1 the slogan is shown.
The seven-minute music video presents a visually rich narrative with contemporary themes such homophobia, transphobia, and the inclusion of people with hearing impairments. One of Flutua's peculiarities is that its main scene is two minutes long and takes place in complete silence. This was one of the factors that made us choose it for our accessibility project. In the remaining audio described scenes, we chose to edit the audio by decreasing the track volume and adding a few seconds of audio description in ten different parts of the song.
As mentioned before, the music video story follows a gay deaf couple who communicate by using Brazilian Sign Language (LIBRAS). They spend the day in São Paulo hanging out with their friends, who are also deaf. They walk around different spots in the city, dancing during the day and at night in various urban environments. Besides these spaces, two scenes take place in the apartment of one of the main characters. The most important scene takes place under an elevated highway in downtown São Paulo, best known as the "Minhocão" [Big Worm].

FLUTUA: A COLLECTIVE PROJECT IN AUDIO DESCRIPTION
Flutua's AD was produced by the Accessible Media and Audiovisual Translation Research Group (MATAV 2 ) team, including students from the Design; Journalism; and Radio, TV and Internet undergraduate programs at Unesp Bauru, together with the assistance of translation professionals.
Several AD professionals, such as Benecke (2004) and Zardo (2017), argue that, as AD is made up of several stages, it must be performed by a team with different roles. According to Zardo (2017), Studies already establish -and the practice of audio description accepts and uses -three essential functions for this process: the screenwriter, the audio describer that writes the descriptive text for a specific product; the consultor, the visually impaired audio describer who analyses and reviews the script; and the narrator, the audio describer responsible for reading the (live or recorded) audio description script. (ZARDO, 2017, p. 14, our translation) We divided the Flutua project into stages and took into account the team roles as described by Zardo (2017). The AD script was completed in a collaborative 1. youtu.be/Q5VIhKjiit8 2. Mídia Acessível e Tradução Audiovisual (MATAV). process with around ten MATAV group members 3 . First, we divided the music video into scenes, considering changes in locations and the narrative time (day or night). The time change is important as the characters experience a transition between feelings of joy, euphoria, sadness, and guilt during the intense day when the narrative takes place.
The scripting team used the time to describe the gender identity of singers Johnny Hooker and Liniker as well as that of the two gay deaf protagonists. For consistency purposes, we have chosen to disclose to visually impaired people that Johnny Hooker is a gay male singer who puts on makeup and dresses in women´s clothes. In our conception, we needed to describe Liniker as a transgender black singer. It was very important to mention these elements because she represents a sizeable black community that has used the audiovisual medium to appropriate a strongly transphobic, homophobic and racist space.
In an interview for Revista Caliban magazine, Johnny Hooker explains how he uses his stage name to create an alter ego for John Donovan, his real name. Johnny Hooker lives within this somewhat nerdy boy called John Donovan. This is an alter ego. When I go up on the stage, it feels like life or death. That is when Johnny Hooker breaks free. Even before entering the stage, in the dressing room, he is already trying to break free. Sometimes it takes him a long time to come out. Sometimes he comes out like a hurricane. He does not have a gender identity. He is neither a man nor a woman. (https://revistacaliban.net/ johnny-hooker-n%C3%A3o-%C3%A9-homem-nem-mulher-%C3%A9-umaentidade-66c8359c188e) Despite stating that his alter ego does not have any gender identity when he goes up on the stage, we decided to objectively describe Johnny Hooker's outfit and makeup as follows: Johnny hooker is a thin white man with straight brown hair. He has blue eyeshadow and light red lipstick.
He has a moustache and is wearing a white shirt with the buttons open, showing his chest in a suggestive manner.
By describing Hooker as a white man who has a moustache and wears blue eyeshadow, red lipstick, and a white shirt with the buttons open showing his chest in a suggestive manner, we allow audio description users to perceive the ambiguity that Johnny Hooker himself defends when he mentions that he "is neither a man nor a woman" on stage.
For the second stage of the project, two professional audio describers reviewed the script proposed by the MATAV team. These two consultants have extensive experience in AD in movies and theatre and made the necessary adjustments to our text.
The last stage involved the narration all the parts of our AD script in a studio. We had a radio journalism professional and a researcher in sound media, who helped to complete our accessible product.

VIOLENCE AGAINST GAY AND TRANSGENDER PEOPLE IN BRAZIL
Currently there are about 20 million Brazilians who self-identify as LGBTI. Scholars and activists say many people still have not declared their gender identity or sexual orientation as they are afraid of prejudice. We consider this data very significant for a country with over 200 million people.
Unfortunately, a large part of the LGBTI community faces a cruel reality in Brazil on a daily basis: murder, violence, and suicide. Brazil has the world's highest LGBTI murder rate. Recent data show that one person is murdered every 27 hours due to their gender identity or sexual orientation. According to the annual report published by the Grupo Gay da Bahia (Gay Group of Bahia) in 2012 there were 338 murders of gays, trans and lesbian people in Brazil, including two Brazilian trans women killed in Italy. In 2012 gay people lead the ranking of victims of murder 188 (56%), followed by 128 transgender people (37%), 19 lesbians (5%), and 2 bisexual people (1%).
As an example of the violence against transgender people we can analyse the 2016 Trans Murder Monitoring (TMM) report which showed that 868 trans and gender-diverse people were murdered between January 2008 and June 2016 in Brazil. This makes Brazil top the ranking and shows how cruel is the prejudice and violence against gender-diverse people is. It is essential to highlight that the transgender immigrant population with the highest victims of murder rate in the same period (2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016) in Europe was also Brazilian. Figure 1 presents the graphs issued by the 2016 TMM ranking on killings: All the data presented show the concern of a number of sectors of Brazilian society that attempt to protect gender diverse people. Several of these organizations and associations defend trans respect, in addition to a significant number of members and research groups in the Brazilian academic community, which seek to find ways to discuss the defense of minority rights through numerous projects.
Recently the Brazilian Supreme Federal Court started considering homophobic and transphobic conduct as racist crimes (Article 20, Law 7.716/1989), now referred to as "criminalisation of homophobia." The next challenge is for the National Congress to make a specific law covering every legal aspect to protect LGBTI people, thus punishing those who commit homophobic and transphobic acts.

AUDIO DESCRIPTION AND THE LANGUAGE OF MUSIC VIDEOS
Flutua has been chosen as our object because we wanted to show how audio description (AD) allows visually impaired people to understand the importance of gay and transgender people being represented in Brazilian audiovisual products, namely in the music video genre. We defend the position that visually impaired people would not perceive the leading role of these people unless they have access to audio description tracks that mention gender identity details.
According to Bernd Benecke (2004), audio description can be defined as: […] the technique used for making theatre, movies and TV programmes accessible to blind and visually impaired people: an additional narration describes the action, body language, facial expressions, scenery and costumes. The description fits in between the dialogue and does not interfere with important sound and music effects. (BENECKE, 2004, p.78) Although in our conception of audio description technique is only one aspect of the whole process, we have considered in our practice audio description as an audiovisual translation resource that translates movements, costumes, and facial expressions into words in movies, music videos, and other audiovisual genres.
As we will show in the next section, our toughest challenge was to audio describe critical items in a music video with several narrative elements: the characters' and performers' body language; the facial expressions of the deaf characters, who communicate mostly by using gestures; the scenes of São Paulo city; and the singers' outfits.
After searching for Brazilian studies on music video language, we found a study by Caldas (2013), whose master's research is called Music Video 2.0: Interactivity and Interaction Schemes in the Digital Age, where he affirms: [ ] thanks to music videos, funding and technological resources are being allocated to experimental productions, which in the past were paid by the artists themselves. Machado also states that, due to pop culture, to which music videos are structurally associated, maybe "certain transgressive attitudes in the creative audiovisual environment can reach the mass audience" (CALDAS, 2013, p. 25, our translation).
Besides the ever-present difficulties inherent in the audio description of audiovisual products, the music video genre is naturally centred around a song, which must be reedited with great caution. As with subtitling, audio description will always be an insertion transforming an audiovisual product into a new work with oral records (audio description) that were not foreseen in its pre-production.
Pablo Romero-Fresco (2013) argues that accessibility features should no longer be an afterthought in audiovisual and film production. That is, producers, filmmakers and audio describers or subtitlers need to think together to create an audiovisual accessibility project, which includes scenes with dialogues that are designed so that the audio description track is inserted as naturally as possible, rather than only in ten to 15 seconds of a scene. Flutua, though, is an example of an audiovisual production that has not been designed with accessibility, which is the main trend in real audio description practice. It would be much more inclusive if pre producers also included visual impaired people as their target public.
To make the video accessible to visually impaired people from the very beginning, we have included an audio introduction (AI) so users can understand the summary of Flutua's narrative. Historically, audio introductions emerged at the same time as audio descriptions. According to Fryer and Romero-Fresco (2014), audio descriptions and audio introductions started being used in plays at the same time. Later, movies and other audiovisual genres also took advantage of audio introductions. Fryer and Romero-Fresco (2014) state: An AI for film may be defined as a continuous piece of prose providing factual and visual information about the source material that must engage the listener´s attention, whet their appetite, and, most importantly, enable them to appreciate the film's inherent cinematic qualities. (FRYER; ROMERO-FRESCO, 2014, p. 12)

TRANSFERRING DIVERSE IDENTITIES
In the current globalised world, identity construction processes are necessarily characterised by the intersection of cultures, and, in terms of gender, sexuality and visibility, discourses are transferred into other cultures by means of translation. In the present work, the translation of sexual orientation and gender identities takes a step further by creating audio description for an audiovisual content starred by nonnormative identities. Audio description, which is considered a form of mediation which is "cross-modal, involving essentially a 'translation' of visual images into verbal text" (BRAUN, 2008, p. 2), mediates between the visual content and the consumer that cannot extract information from what they see. Audio description will inevitably serve as filter.
When transferring non-normative sexual and gender identities, the importance lies in how we see and transmit such information. The simple task of describing can adopt a whole new political essence since the mere action of speaking is never neutral (MARTÍNEZ PLEGUEZUELOS, 2018). Therefore, describing diverse human identities seems an impossible achievement since, in essence, no categories should be found, and there is no tangible skeleton that helps us to situate what is perceived. The categorisation by which such identities are ordered is a mere human convention that is politically loaded (CAMERON, 1995, p. 122). It is worth signaling that diversity also accounts for other aspects of the human experience, such as ethnic group or racial features, that might have implications in the way a person is read and understood in different contexts around the world.
The role of translation as promoter of visibility discourses is not always liberating. Contrarily, it can also make a message go back to the normative and traditional molds that divide and organise society. It can easily forget the accents, nuances and particularities of the experiences of some individuals, and these can also be read differently in different social contexts. The loss of nuances is particularly important in the field of audio description since time restrictions in the practice require summarised and understandable contents that should generally aim to provide the maximum information in the minimum time.
The reduction of identities to a recognisable group of tags or labels such as "Muslim", "gay" or "woman" frequently results in the multidimensional complexity of each human being hidden. The use of such reduction when describing a person can make such human experience transparent (CRONIN, 2009, p. 218). The intention of portraying complex and diverse identities in audio description really poses a challenge, and this has been emphasised in the 21 st century thanks to the development of post-modernist ideas and their call for the re-construction of selfidentities.
Although the discussion of such topic within audio description has been little explored, the manner in which language shapes bodies has been widely investigated. Judith Butler (1997) argued that it is not the language that sustains the body and it brings it into being, but it is by being interpellated within the terms of languages that a certain being becomes possible (p. 5). As a matter of fact, some years earlier the renowned author Simone de Beauvoir wrote "you are not a woman but become a woman". That is, the interpellation of woman makes her existence real. Martel (2013) makes the analogy with the homosexual, who is not born a homosexual but becomes one in the sense of a socially constructed product (BUTLER, 1999(BUTLER, /2007. The complexity lies in the fact that the affirmation of any act of identity, the interpellation, as stated by Butler, requires a certain level of essentialisation, which is something that contrasts with theories rejecting binarism and advocating the forging of diverse identities. The challenges for the translator/audio describer are thus presented when representing minority sexual and gender identities nowadays (MARTÍNEZ PLEGUEZUELOS, 2018, p. 2). The simplification that may be found in audio description due to these particularities could hinder the expression of the multidimensionality of identities by restricting the contents being transferred to the user.
Having in mind the intrinsic challenges posed by the characteristics and use of audio description, as a mediation between the visual and the verbal channels, some examples of the audio description and audio introduction of the video clip Flutua will be presented. These excerpts are proposed to point out some aspects which the audio describer(s) must take into account when transferring a message to visually impaired audiences.

DISCUSSION OF EXAMPLES
As mentioned before, we have produced the audio introduction, which will be presented first, and audio description of ten scenes, of which we will present below six of them.
For the creation of the audio introduction we did not find any precedent example of an AI for a music video, we took into consideration most of the elements mentioned in Fryer & Romero-Fresco (2014) in the AI for Flutua. Our AI is about 46 seconds long.
Audio introduction for the Flutua music video: The music video tells the story of a gay deaf couple who uses sign language. They spend a pleasant day with other deaf friends. the scenes take place in the street and at a club at night. One of the men is tall, thin, white, and has a moustache. The other is tall, well-built, white, and has a full beard. At the end of the night, an unfortunate event takes place.
Johnny Hooker is a thin white man with straight brown hair. He has blue eyeshadow and light red lipstick. He has a moustache and is wearing a white shirt with the buttons open, showing his chest in a suggestive manner. Liniker is a black transgender woman with an afro hairstyle. She has black lipstick and is wearing a sexy, deep v-neck gold blouse.
The audio introduction presents the characters in the music video, both the two singers and the young men who star in the story. The word "gay" has been used to talk about men since in Brazilian context gay refers only to man who is affectively attracted and/or sexually attracted by people of the same sex/gender. Then we use the word men to refer to the two of them and point at the beard and moustache to first, characterise them and then to highlight their cis-male identity.
On the contrary, the two singers are described with more queer characteristics, Matav team decided that the presentation of those characteristics would be relevant to reinforce their identities. We have also checked in many reliable sources (Official webpages and interviews) their preferred nouns in Portuguese. Johnny Hooker prefers the pronoun he and Liniker announced in 2018 that the pronoun she should be adopted when referring to her. We thought it was important to provide information about the ethnic characteristics of Liniker as a queer black transgender woman because, as we mentioned before, she represents a black community that has used the audiovisual medium to appropriate a strongly transphobic, homophobic and racist space.
After the AI, we created audio descriptions of ten scenes we judged the most important for the audience to understand the music video script. The audio description of these six scenes is presented in the following paragraphs. Each audio description excerpt is accompanied by an illustrative image of the scene being described.
Scene 1: The man walks into the apartment and speaks in sign language with his boyfriend, who is lying down and whose face is hurt. He is smoking.
Scene 2: Daylight. Several happy young people are on a walkway. They speak in sign language. They are moving and dancing to the music. The couple are dancing as if they were floating. Image 3. Scene 3.
Scene 3: Night time at a club. The bearded man is dancing with a blond man while his boyfriend watches from a distance and writes "Flutua" (Floating) on a mirror. Then they passionately kiss. Scene 4: The couple are under an elevated highway. In the background there is a concrete column with graffiti on it. They approach each other. They kiss in a restrained manner. They look to the side. They walk away, then walk back, and kiss. The bearded man gently pulls the man with a moustache by holding his striped jacket. The bearded man walks away and lights a cigarette. The one with a moustache chews gum. He looks at his boyfriend and turns around. He walks away from the bearded man. He keeps strolling. In the blurry background, two men appear with an iron bar and violently hit the bearded one. He is kicked several times as he is on the floor. The man with a moustache keeps walking without realising what is going on thinking his boyfriend is behind him. The beating continues.
Scene 5: At the club, the man with a moustache leans toward the bar. He goes to the dance floor. He dances. The bearded one arrives and walks to the mirror wall. He places his fingers on the graffiti "Float." He walks towards the man with a moustache. They kiss.
Image 6. Scene 6 Scene 6: Retrospective scenes. Friends are having fun during the day. The man with a moustache spray-paints the word "Float" on a wall at night. The bearded man is beaten up. He is in bed. At the club, the couple are dancing holding each other. The bearded man holds his boyfriend around his waist and lifts him off the ground. Liniker and Johnny Hooker kiss. The couple are hugging on the dance floor.
The story described by the song offers less problematic in terms of gender identity and expression since the two protagonists show what can be labelled as normative male expression. However, the different scenes are relevant in terms of content, since they show homosexual romantic and erotic scenes, which later on lead to the violent attack to one of the men.
Some issues, though, are worth mentioning. The type of sign language they speak or the characteristics of this sign language have not been stated in the AD, however, in other occasions this could have made a point in terms of activism for people with disabilities, and particularly for the deaf community.
Afterwards, a group of people is composed by men and women, in English the pronoun "they" refers to all kinds of gender, in Portuguese there are two different pronouns (eles and elas) but the neutrality is produced by the omission of the subject. The sentence, then, results in a gender-neutral description of the participants in the scene, avoiding the division of women and men, and the exclusion of gender nonbinary identities.
The AD created for this video clip has tried to use a language understandable easily by all kinds of audiences. However, there are certain points that have been brought to the attention of the reader that can arise some debate, such as gender expression, the use of female and male pronouns or the reference to the expression of a non-binary character. Other strategies could have been used to address these topics in the video, and it is the responsibility of the audio describer to present all these items in a certain way. By taking a descriptive approach, the article had the intention to comment on these issues, which will certainly be more and more frequent in future audiovisual production worldwide.

CONCLUSION
Flutua was one of the projects produced by Mídia Acessível e Tradução Audiovisual Research Group (MATAV) whose focus is including blind and deaf people in a more democratic society.
In our audio description of the music video, we have proposed to work collaboratively so we could make accessible a product that was not only a song, but something produced with a cinematic narrative that presents a contemporary story of love and resistance to homophobia and transphobia. We wished to think of audio description as a tool to strengthen gender and sexual identity through which we could allow visually impaired people to understand the reality presented in Flutua 4 . In their quest for freedom to express their gender and sexual identity, many LGBTI people have been victims of violence and murder.
The audio description of this kind of audiovisual content provides access to new topics to people who are blind or who have sight loss. The relative recent inclusion of non-normative and every day more diverse characters in audiovisual production opens the door to new challenges. This is particularly relevant when tackling queer theories, that propose the permeability and the diffusion of the limits between the categories established traditionally for gender and sexual orientation. New terms and concepts are arising, and others are disappearing. However, the advances and new proposals in the field of gender and sexuality must find compromise with the reality in which people who are blind or are visually impaired live.
Nonetheless, the portrayal of diverse identities in audio description is a necessary task that should also aim to make such identities exist. How diversity is expressed visually provides a lot of information about identity, and this is a key element when constructing characters in the minds of blind and visually impaired users. However, this task could bring some difficulties to professionals since technical restrictions may hinder the detailed transfer of information. In this case, plain language and affordable concepts and terms have been used to take this music video close to the blind and visually impaired audience. This topic has not been duly tackled in academia perhaps because of the recent presence of this kind of characters in mainstream audiovisual production or the relative recent incursion of the branch of Gender Studies in other fields, more research is required on the topic.