Abstract

Media accessibility is an increasingly prominent issue in today’s rapidly developing technological age. The growing awareness of the diverse needs of the population is reflected in legislation and facilities for accessing information, and is starting to also show signs of recognition within the world of entertainment, as demonstrated by legislative acts on accessibility including the Broadcasting Act 1990 and the Disability Discrimination Act 1995. Indeed, accessibility to the arts is becoming a more highly-rated concern with various art forms, including opera, making provisions to improve their access services. Given the multilingual, multisemiotic and increasingly multimedia nature of opera, the translator plays a fundamental role in this process. This article looks at how the strategies, challenges and general role of the translator are affected by issues of accessibility in opera, focusing on methods for making opera accessible to wider audiences including the sensory-impaired.

Keywords: opera translation, audiovisual translation, media accessibility, surtitles, intersensorial translation, audio description ad, audio introduction ai, audio subtitling as, sign language interpreting, touch tours, visibility of the translator, multisemiotic translation

1. Introduction

Opera is breaking away from its stereotypically elitist image to embrace notions of accessibility and diversity, and the role of the translator is rapidly evolving to reflect these developments. Today, the translator's task is to overcome "not only linguistic but also sensorial barriers" (Orero and Matamala 2007:262) bringing the once rarefied world of opera to all audiences. Therefore, various new modalities of audiovisual translation (henceforth AVT) involving interlingual and intralingual translation, defined by Jakobson respectively as "an interpretation of verbal signs by means of some other language" and "an interpretation of verbal signs by means of other signs of the same language" (1959:139), as well as intersensorial transfer (defined later), are being introduced. These innovations have raised complex issues which readdress discussions surrounding Venuti's debatable dichotomy of the visibility and invisibility of the translator, they demand new or adapted translation theories and terminology, and redefine the notion of the translator's task. Firstly, it is important to identify the various modalities of opera translation, considering the role of the translator in terms of strategies, challenges, responsibility and status. Opera translation methods overcoming the linguistic barrier will be analysed briefly initially, followed by an investigation into the semiotic aspects of opera, before moving onto in-depth examination of the translator's task within new opera translation techniques for sensory-impaired audiences. Then, a discussion regarding the opera translator's visibility is required, focusing in particular on the tension between the invisibility and visibility of the translator within certain relevant translation modalities, and also involving topical debate on issues of impartiality. Having explored these issues from a more theoretical point of view, it will be necessary to consider the practice of various types of opera translation from a professional perspective, studying discrepancies between theory and practice, as well as drawing on interviews with expert practitioners and examples from individual performances at the Royal Opera House, Covent Garden (henceforth ROH). Finally, a discussion about the multifaceted nature of opera and opera translation is required, and conclusions will be drawn regarding the overall impact of developments within the world of opera on the general role of the translator.

2. Opera accessibility for all

"Accessibility is a form of translation and translation is a form of accessibility" (Díaz-Cintas et al. 2007:13-14). This statement at the same time affirms and reveals how inextricably linked these two notions are and points towards the fundamental role played by the translator in making opera accessible. The first step towards opera accessibility is overcoming the language barrier, a translation task which can be performed through three different activities: "'straight' libretto translation; subtitle or surtitle translation; and singing translation" (Burton and Holden 2005:1). Secondly, in order to make this medium accessible to sensory-impaired audiences, new modalities of translation are being introduced to the world of opera. These include audio introduction (henceforth AI), audio description (henceforth AD), audio subtitling (henceforth AS) and touch tours for the blind and visually-impaired, as well as sign-interpreted performances and captioning for the deaf and hearing-impaired, all of which will be defined later. If an opera is performed in a language which is foreign to the audience, the sensory-impaired viewers not only require the first step of interlingual translation in order to understand the words of the opera but also an additional transfer process of intersensorial translation.

3. Making opera accessible to international audiences: a historical overview

From the very birth of opera during the Renaissance period in Italy, the importance of making opera accessible to the audience has been recognised and "a strong priority was given to the comprehension of the text" as reflected in the title favola in musica attributed to Monteverdi's Orfeo, first performed in 1607 and now considered to be one of the first operas (Desblache 2008:156). This emphasis on accessibility and comprehension persisted and whilst Italian remained the principal language of opera until the nineteenth century, this mode of entertainment became increasingly popular and widespread across Europe, thus rendering translation paramount. In Germany, Italian opera was translated or adapted from the mid seventeenth century and in England translated operas were performed from the beginning of the eighteenth century, the translation modalities in use at this time being libretto translation and then later singing translation. These methods for overcoming the linguistic barrier continue to be exercised today with the addition of opera subtitles and surtitles which did not appear until the twentieth century but similarly originated from "a reception need" (Mateo 2008:136), that is demands for more accessibility made by "both the public and subsidising organisations" (Desblache 2008:163).

3.1 Libretto translation

The term libretto literally meaning "small book" was coined soon after the emergence of opera, and whilst originally referring to a "printed or manuscript book giving the literary text, both sung and spoken, of an opera (or other musical work)", the word "has also come to mean the text itself" (Macnutt 2007-9). Nowadays there are two types of libretto translation: word-by-word libretto translation (Sario and Oksanen 1996 in Orero and Matamala 2007:262-263) and "straight" libretto translation (Burton and Holden 2005:1). With these translation modalities the translator's role is determined by the function of the translation, and the strategies adopted vary accordingly. The function of word-by-word libretto translation is "to be used as a working document by singers who want to grasp the meaning of the lyrics" (Orero and Matamala 2007:262). They are intended neither for audience viewing nor to be sung. The translator's task is therefore limited to decoding the original libretto with accuracy. An overall strategy of literal or "semantic translation" (Newmark 1981:23; 1991:1,47) is most likely to be adopted as "every word has to be translated" (Burton and Holden 2005:1), although as Burton and Holden point out, there may be occasions where a more communicative approach must be applied, such as in the translation of idioms (2005:1). This form of interlingual opera translation is considered "the most straightforward" and the main demands include familiarity with the "many classical, historical and cultural references, and with certain archaic turns of phrase" (Burton and Holden 2005:1). A very similar role is played by the translator in producing "straight" libretto translations. Again, these interlingual translations are solely to be read, although in this case, they are for audience viewing, as towards the end of the nineteenth century they appeared on programmes and from the twentieth century they have been issued with gramophone record, CD or DVD opera recordings. The same overall semantic or literal translation strategy is applied due to the frequent parallel presentation of the source text and target text, and phonic aspects can be ignored as the source text is often translated as "unrhymed free verse" (Low 2002:100). This type of libretto translation is considered to be literary translation, although Low suggests that they "are perhaps easier to produce than much literary translation, since they are done to complement the musical recording, not to stand alone" (2002:100).

3.2 Singing translation

Singing or "singable" translation (Low 2002:99) involves adapting the source libretto into the target language "in order to be sung and facilitate access to the opera when performed" (Orero and Matamala 2007:263), a trend which was established in the nineteenth century (Desblache 2008:161). In this mode of interlingual opera translation the task of the translator is considered to be "the most difficult" (Burton and Holden 2005:5; Desblache 2004:28) due to the challenge of synchronising music and words. The translator's strategies must take into account the importance of "semiotic cohesion" (Díaz-Cintas and Remael 2007:49-52), that is the interaction between the various semiotic aspects of opera (a concept which will be explored further later), whilst aiming for absolute clarity and naturalness. Amanda Holden, a highly reputed English opera translator for English National Opera (henceforth ENO) and Opera North amongst other companies, declares, "a translation isn't any good if it sounds like a translation" (Burton and Holden 2005:6), thus confirming that the overall approach adopted is most likely to be "communicative translation" where the aim is one of "equivalent effect" (Newmark 1981:23). This notion of equivalence or faithfulness (Desblache 2004:28), as discussed by Lucile Desblache, a renowned French opera translator, involves a delicate balancing act on the part of the translator. Desblache argues that faithfulness is not desirable, especially in the case of comic opera where entertainment value is a high priority, referring to the advice of outstanding librettist and opera translator Eric Crozier "to capture the spirit of the text, but to forget the source words" (Desblache 2004:29-30). A method of free translation is also supported by Low who reiterates the conflict between musical constraints and semantic accuracy, and suggests a flexible translator's role (2002:100; Orero and Matamala 2007:263). Holden also advocates mirroring the spirit of the original, but emphasises the importance of matching the musical phrasing and texture of the source text, stating "you must be faithful to the libretto, and the rise and fall of the phrases of translation must match those of the original words and music" (Burton and Holden 2005:6). Whilst striving for phonetic and stylistic faithfulness, a certain amount of freedom and artistic license is inevitable, thus rendering the translator a skilled juggler. In view of the translator's necessary liberties with the source text, and the almost poetic creativity required to overcome difficulties in phonetic and rhythmic differences between languages[1], it could also be argued that the translator's role is of authorial status as the translation becomes more of an adaptation, often with poetic and rhetorical originality of its own.

3.3 Subtitles and surtitles translation

Opera subtitles, "translated text displayed below the image, as on a cinema or television screen", have been around on film since the early twentieth century and on television since the early 1970s (Burton 2009:1). Surtitles, also known as "overtitles" (Orero and Matamala 2007:264) or "supertitles" (Burton 2009:1), all referring to captions projected above the stage (see picture 1 below) or on seatback screens during live opera or theatre performances, were invented in Canada around 1983, although the first live titles in an opera house, displayed vertically at the side of the stage, were reputedly in Hong Kong in the early 1980s (Burton 2009:1).

Picture 1: Tosca, ROH, July 2009 (kindly provided by Judi Palmer)

The advent of both opera subtitles and surtitles, the most recent forms of translation which render opera accessible to a wider general audience, was inspired by audience demands, but was made possible due to technological progress. Dewolf describes titles as "a technical device to make opera more user-friendly" and suggests that this progress is "revolutionising the translator's profession" (2001:180,179). The task of opera subtitlers and surtitlers does indeed demand a previously unnecessary degree of technical competence, as demonstrated by Burton's description of the various software, digital and slide projection systems used (Burton 2009:2-3), thus rendering the translator a technician as well as a communicator (see picture 2 below).

Picture 2: Judi Palmer in the surtitles box at the ROH (kindly provided by Judi Palmer)

In this respect, and in general terms, the role of the translator in producing subtitles and surtitles is much the same, although given that surtitles are for live performances and subtitles are usually for recorded performances[2], there are some distinctly differing requirements, most notably regarding pacing and synchronisation. For live opera surtitles "the pacing should be kept slow and simple, to avoid distracting the audience" (Burton 2009:10), whereas for opera subtitles "the pacing can be faster" (Burton 2009:10). The surtitler's task demands greater versatility, as a "musically literate" operator (Low 2002:104) is required to follow changes in tempo. This flexibility can also be seen in terms of the surtitler's potential collaboration with the director and other members of the production team, as the task of the surtitler involves more teamwork and adaptability in accommodating peculiarities and last-minute changes within a given production. As Low states, "opera surtitles are devised during a 'production in progress", whereas a subtitler usually works from a completed film" (2002:102). This notion of the translator's role within a team is discussed further below. Similarities of the translator's role in writing subtitles and surtitles include aims towards simplicity (Burton and Holden 2005:3), readability (Orero and Matamala 2007:265) and conciseness, aspects which are explored in more detail in sections 6.2 and 9.1. Although surtitles are more likely to contain omissions and condensation, both surtitles and subtitles impose constraints of time and space with a usual "maximum of 2 lines per title and 32 characters per line" (Low 2002:103). In both modalities the translator also seeks to avoid interference with the "listening and viewing of the opera" (Desblache 2004:28), thus recognising the importance of semiotic cohesion (discussed later). This approach makes it likely that the translator will adopt an overall strategy which does not favour literal translation of the source text because an aspect of design, whether details of props or staging or any other visual or audio feature, might be "at odds with a literal translation" (Burton and Holden 2005:4). Therefore, whilst aiming for discreet titles, the translator may on occasions be required to assume the role of adaptor due to the need for a production-specific translation. The significance of the other semiotic modes in opera (see figure 1) may aid an unobtrusive style, as sometimes "the musical and emotional meaning could come across without the verbal content" (Low 2002:107), or at least with minimal written translation, such as in repetitive arias and choruses. However, in other more narrative passages of recitative, for example, the translator may have to convey verbal communication to a maximum, especially in comic opera where this is particularly important in terms of humorous function. It could therefore be argued that there is an element of flexibility in the task of both the surtitler and subtitler given the diverse functions of different genres of opera, including amongst others opera seria and opera buffa (Brown 2007-9), and of varying narrative elements of opera including aria, chorus and recitative. Subtitling and surtitling also share their "functional nature" to "facilitate comprehension" of the text (Mateo 2008:137) and therefore Low recommends a "skopos-based" strategy (2002:109; also cf. Reiss and Vermeer 1984:119), thus focusing on purpose and audience needs. Originally surtitles were produced for audiences unfamiliar with the source language but the reception of these interlingual surtitles, whilst initially causing controversy amongst traditionalists, gradually changed the expectations of receptors to consider understanding the text to be part of the artistic experience. These altered requirements combined with difficulty understanding words sung in high registers led to the introduction of intralingual surtitling and a new type of translation role for surtitlers who became "communication facilitator[s]" (Mateo 2008:151). Furthermore, this focus on the audience, now showing "a desire to understand the verbal text at the same time as they receive the music and seem to have realised that full comprehension of the opera can only be achieved through the simultaneous interpretation of all semiotic signs in it" (Mateo 2008:137), together with an aspiration to make opera accessible to all, has led to a new task for the opera translator involving intersensorial transfer.

4. Making opera accessible to sensory-impaired audiences: opera semiotics

Intersensorial translation (De Koster and Mühleis 2007:189) can be defined as the transfer process between one sensory communication channel and another, for example, translating visual aspects into audio format or vice versa. This type of transfer transcends the traditional concept of translation, redefined by Gottlieb as "any process, or product thereof, in which a combination of sensory signs carrying communicative intention is replaced by another combination reflecting, or inspired by, the original entity" (Gottlieb 2005:3). It involves interpretation of the semiotic signs of opera, illustrated below in figure 1, which is based on categorisations by Delabastita (1989:199; also cf. Díaz-Cintas and Remael 2007:46) and, as reported by Cabeza, opera scholars such as Pahlen, Arregui and Vela (Pahlen 1963; Arregui and Vela 2007 in Cabeza forthcoming:1), with the addition of tactile, olfactory and gustatory aspects. It must be noted that this scheme is merely representative of the numerous factors to be considered.

Figure 1

The semiotic signs of the audio and visual systems can be interpreted by hearing and seeing audiences without aid because two senses which can function simultaneously yet independently are used. However, for the sensory-impaired, the translator has to convey the aspects to which the visually-impaired or hearing-impaired do not have access, performing processes of intersensorial transfer or as Gottlieb labels it "intersemiotic translation", defined as translation in which "the one or more channels of communication used in the translated text differ(s) from the channel(s) used in the original text" (Gottlieb 2005:3). Geryzmisch-Arbogast extends the definition of this type of translation characterised by multiplicity by defining "multidimensional translation", as she denotes it, as "a form of translation which transfers - with a specific purpose - a speaker or hearer's concern expressed in a sign system 1, formulated in a medium 1, via the same medium or a medium 2 or a combination of media into another sign or semiotic system 2" (2005:5). In view of these various definitions, it would seem that there are terminological issues which need to be addressed. However, for the purpose of this paper with its forthcoming focus on the translator's role within techniques specifically aimed at sensory-impaired audiences, the expression "intersensorial translation" has been integrated as defined at the beginning of this section. Furthermore, the term "multisemiotic" has been chosen here instead of Gottlieb's "polysemiotic" (2005:2) in order to maintain terminological consistency, as various other words with the Latin prefix "multi-" are employed. This term "multisemiotic" which can be defined, like "polysemiotic", as communicating through more than one semiotic channel (Gottlieb 2005:2) can be applied not only to surtitles, subtitles and singing translation, but also to all types of opera translation for the sensory-impaired, as well as opera itself. Indeed, one of the defining characteristics of opera is its multisemiotic nature. As Buyssens states:

The richest collection of semical facts seems indeed to be that produced by the performance of an opera. The artists communicate with the audience in a variety of ways: through words, music, mime, dance, the costumes of the actors; through the orchestra, the setting and the lighting on-stage and in the auditorium; through the architecture of the theatre (Buyssens 1943:56, translated by Marvin Carlson, in Puigdomènech et al. 2008:382).

It must be noted that if analysed from "the end user's perspective" (Gottlieb 2005:5), it might be considered that to a profoundly deaf or totally blind person opera is not a multisemiotic genre because these patrons only perceive it via either the audio or visual channel. However, it can be argued that opera may appeal to additional channels of communication including tactile, olfactory and gustatory sign systems, as illustrated in figure 1 above, thus rendering opera multisemiotic for deaf and blind audiences. Furthermore, translation techniques and the translator also play a salient role in rendering opera multisemiotic by helping the sensory-impaired to engage with these supplementary sign systems, whether directly, for example through tactile experience such as touch tours (discussed later), or by using verbal language such as allusions and rhetorical devices including metaphors and similes which have references to the senses. These additional semiotic elements of opera, especially tactile features, are exemplified in two recent productions of Mozart's opera, The Magic Flute, and also of Mozart's opera buffa, Don Giovanni. The colourful, dynamic adaptation of The Magic Flute which was performed at the Duke of York's Theatre in London in April 2008, implicitly engaged various senses, particularly communicating through the tactile system with the orchestra of marimbas and various African-style rhythmic dances both causing substantial vibrations. The ROH production of Don Giovanni in September 2008 also included features with a tactile dimension, most notably in the final scene when Don Giovanni descends into hell, as shown below in picture 3, in which large hot flames shot up from the stage. In addition to the visual and auditory aspects, this dramatic spectacle not only has the potential to appeal to the sense of touch for example through the non-verbal signs of increase in temperature, but also to the sense of smell via the non-verbal signs of odours produced by the flames.

Picture 3: Simon Keenlyside as Don Giovanni (photo by Catherine Ashmore, courtesy of the ROH)

Both of these examples also demonstrate the important role of the translator because although the sensory-impaired can engage with these features to a certain extent without aid, some explanation and contextualisation via verbal description may be needed. This is in order to prevent these aspects from causing undue alarm and to allow them to most effectively and fully contribute to making the opera real for the sensory-impaired, as well as heightening their enjoyment of the complete opera experience. On some occasions, as in the case of Don Giovanni, it may be that an effect is employed on stage with the purpose of provoking shock. Therefore the translator's role involves striking a balance between allowing the audience to experience this whilst also providing contextualisation where necessary in order that the sensory-impaired can receive the same impact as fellow seeing and hearing spectators. This discussion regarding additional channels of communication raises the issue of the translator's role in communicating not only the elements of opera but also the features of the opera experience. It is an important distinction to consider, although the boundaries between the two are very subjective and flexible. For example, whereas gustatory sensations produced by consuming food or drink during the performance could be labelled more readily as features of the opera experience as opposed to the opera itself, the classification of surtitles, or indeed the tactile aspects in the above examples, as elements of the opera or opera experience is debatable. Therefore, in this paper the semiotics of opera encompasses all aspects including those which might be considered to be semiotics of the opera experience, as the role of the translator is explored here in terms of making the opera experience as a whole accessible to all audiences. The difficulty arises in prioritising these aspects for which there is "no general agreement" as to which is the most important (Puigdomènech et al. 2008:382), whilst also avoiding interference with the audience's interpretation of the various other semiotic signs to which they have access. These complexities are present in translation which overcomes linguistic barriers, but are heightened in translation which overcomes sensorial barriers, as aspects which are not available to sensory-impaired audiences must be compensated for in the translation strategies and final product. The tension between music and text, for example, a controversial subject which raises varying opinions amongst translators regarding the relative importance of these aspects, is a particular challenge within translation methods for the visually-impaired, especially if the opera is sung in a language unfamiliar to the audience. According to Low, "above all, opera is a genre where music dominates" (2002:103), a statement which reflects popular opinion, although academics and composers have challenged this (Matamala and Orero 2007:205) to reiterate the importance of text and words as "an intrinsic element not only in the expression of the meaning of what is sung, but of the dramatic action" (Dewolf 2001:182), a change in attitude which perhaps reflects an increasingly text-oriented society. This debate continues to rage and is explored further in section 9.2. Another challenge facing the translator is recognising the importance of semiotic cohesion and then achieving unity between the various semiotic aspects. The notion of semiotic cohesion, previously applied in the context of film as referring to the "interaction between words and images" and between "speech and gesture" (Díaz-Cintas and Remael 2007:49-52), is particularly important in opera translation as there are a multitude of varying aspects which all come together to form a whole. Just as Díaz-Cintas and Remael suggest that if subtitles "are to function effectively they must interact with and rely on all the film's different channels" (2007:45), Mateo states, quoting from Virkkunen "'[s]urtitling opera is about seeing and hearing, reading and writing' and it can only be approached from a multimodal view of opera texts" (Virkkunen 2004:96 in Mateo 2008:136). This perspective which applies to all opera translation modalities emphasises the need for compensation in translation for the sensory-impaired, and highlights the complexity of the translator's task in producing a translation which will accommodate varying degrees of visual or hearing impairments and will appeal to the appropriate alternative senses.

5. Translation techniques for the blind and visually-impaired

Opera translation for the blind and visually-impaired (henceforth the umbrella term visually-impaired will be used to denote varying degrees of visual impairment including total blindness) entails methods of communicating the many and varied visual aspects of opera through alternative senses such as hearing and touch. It is important to note here that in view of the explication of intersensorial translation and Gottlieb's innovative redefinition of translation provided in the previous section, the agent in producing AIs and the other three methods for rendering opera more accessible to the visually-impaired may be referred to as the translator. In touch tours, which will be defined later, it is the visually-impaired patrons (henceforth VIPs) who are essentially the translators or interpreters themselves as no other agent is involved in the process, a notion which will be elaborated upon later. For the other three translation modalities of AI, AD and AS, the role of the translator in facilitating the access of VIPs to the opera experience as a complete whole is crucial. The translator performs a kind of ekphrasis, verbalising the visual aspects of the opera for the VIPs, perhaps drawing on "allusions which have reference to the senses other than sight - especially touch and smell" (York 2007a:13) (see figure 1), with a change in mode from the visual to the written (printed scripts) and then to the audio (spoken scripts). The challenge for the translator is achieving a smooth transition from the visual to the audio mode, a manoeuvre which involves amalgamating the manifold visual elements in opera, in order to produce a "synthetic and vivid description" (Matamala 2005:9). These visual features are displayed below in figure 2, in an arrangement which builds upon Matamala's tripartite categorisation (Matamala 2005:10).

Figure 2

It must be noted that this figure provides a preliminary but by no means exhaustive list of the components which may or may not all be involved in a given opera production. Surtitling and subtitling, for example, is not available in all opera houses or all performances. Regarding libretto and leaflet, sighted audiences may or may not have access to the opera libretto and a programme containing details of the plot synopsis, production, cast list and so on. In addition to these details, information about the auditorium may be supplementary and provided in order to improve the VIPs' access to the overall experience of attending the opera, or may in fact be a feature of any given production (Matamala 2005:10). The role of the translator will entail considerable flexibility as the prioritising of the features in figure 2 will depend upon individual productions and may involve last-minute alterations. The translator will also have to take into account the importance of explaining sounds not included in the libretto which VIPs may need help interpreting as although they can hear, they cannot see the action on stage. Due to the visual nature and therefore inaccessibility to VIPs of comprehension aids such as surtitles and programmes, the translation methods for VIPs will need to contain vital information regarding plot and characterisation, especially if the opera is performed in a language unfamiliar to the audience. Furthermore, certain aspects of scenography may prove more significant in some productions, whether due to their support in communicating plot and characterisation, or as a result of personal preferences of the director. In addition to these considerations, the translator's choices will also be affected by the differing time constraints for each of the diverse methods explored below (in sections 5.1-5.4) which may be used individually or in conjunction with each other. It is also important to note that at present the following opera translation modalities for the visually-impaired and also the hearing-impaired are all usually intralingual, although as interlingual subtitling for the deaf and hard of hearing is being introduced on some DVDs (Díaz-Cintas and Anderman 2009:13) interlingual opera translation facilities may gradually become more widespread.

5.1 Audio introduction (AI)

AI is "an audio text which describe[s] the opera production prior to the representation" (Orero and Matamala 2007:269) or as York defines it "an extended introduction [...] which includes a clear synopsis of the plot, as well as a lot of colourful detail about the visual elements" (2007b:3). In fact, it was Gregory York who, in 1993, set up the first enterprise entitled Talking Notes to provide AIs at musical events, and these services were subsequently commissioned by numerous concert halls and opera companies including the Royal Festival Hall, Glyndebourne Touring Opera and Opera North. This project was, like surtitles, inspired by audience demand and continues to be enjoyed by patrons of ENO and the ROH where "a recorded Talking Notes service [is] freely available at every performance" (York 2007b:7) or on request in advance in audio cassette or CD format. AIs are also offered by the British company Vocal Eyes and by the Teatre del Liceu (henceforth Liceu) in Barcelona, where the recording is available online in mp3 format (Orero and Matamala 2007:269). The task of the audio presenter or translator entails various skills including technical, linguistic and narrating expertise. His or her role could be considered a performance, whether the introduction is delivered live or not, and in both cases the greatest challenge is dealing with time constraints, as AIs usually last only fifteen minutes before the start of the opera with further five minute commentaries before successive acts. The strategies employed must therefore involve substantial condensation or "compression" (York 2007a:6) combined with précis proficiency, as the translator has to communicate the essential elements of the plot in a "nutshell synopsis" (York 2007a:7), as well as providing concise and vivid descriptions of as many visual features as possible, without causing information overload. The task of the translator also requires teamwork because although the AI will most likely be presented by one individual, the translation process from stage to page to recording will probably necessitate a team of researchers, technicians, writers and presenters, because it involves studying costumes and staging by attending model shows and rehearsals and liaising with production members, as well as writing the script. Finally, it is important to note the role of the translator in maintaining a sense of realism whilst interacting with the audience. With this aim in mind the presenter may choose to avoid technical theatrical references and to refer to the audience using the pronouns "you" and "us", for example, strategies which highlight the performance and narrating aspect of this exercise.

5.2 Audio description (AD)

AD is "the descriptive technique of inserting audio explanations and descriptions of the settings, characters and action taking place in a variety of audiovisual media, where information about these visual elements is not offered in the regular audio presentation" (Orero 2005:7; López Vera 2006:1). This method was first introduced on the radio in Spain in the 1940s (Orero 2007a:111-112), then in the theatre in the US, Canada and UK around the mid eighties (Orero 2005:8), and has more recently been offered by several opera houses and companies including the Liceu in Barcelona, Opera North, Scottish Opera, Welsh National Opera and the Grand Opera House, Belfast, and by the UK firm Vocal Eyes. The role of the audio describer varies according to the audiovisual medium in which the AD is employed, and in opera the task is particularly challenging given the prominence of music alongside a wealth of visual elements. Furthermore, opera AD, like AD for other performing arts, is delivered live. In view of these and other specific requirements of opera AD, the role of the opera audio describer must be considered in its own right. Matamala describes opera AD as "live planned audio description" (Matamala 2007:124 in Cabeza forthcoming:1) which as Cabeza explains "means that although delivered live, it has been prepared beforehand and the audio describer has to be alert and react to unforeseen events which may occur during the performance" (forthcoming:1), thus highlighting the demanding, spontaneous and improvisational aspect of the opera audio describer's task. The advantage, however, of AD for opera over other types of live performance such as theatre, is that "the music score helps with the timing of the narration" (Orero and Matamala 2007:272) although the audio describer has to be sensitive to changes in tempo, as with surtitles. It is the live delivery of AD and its intermittent commentary aspect throughout the performance which distinguishes it from pre-recorded AI which is essentially pre-performance, although very brief introductions are given during the intervals before each act. Nevertheless, the audio describer's role has many similarities with the task of AI, including technical, linguistic, narrating and performance skills, as well as creativity and inventiveness in providing evocative verbal depictions. Both must apply strategies of condensation, although the audio describer has more time to give greater detail, and both strive to engage with the audience, even if this interactive quality is more prominent amongst audio describers due to their live performance. This issue of frequent interaction with the audience leads to the question regarding interference with the VIPs' reception of the music, a subject which is central to discussions about opera AD strategies. There are currently two established approaches to opera AD, and one new method which was first implemented in a trial AD by researchers from Universitat Autònoma de Barcelona (henceforth UAB) for a performance of the opera Andreu Chénier at the Liceu in 2007, has now been accepted at this opera house. Matamala and Orero propose AI as a type of AD, therefore comparing the Anglo-Saxon approach to AD, which includes the UK method of AI and the USA technique of involving up to three describers narrating elements of the plot, costumes and stage production independently (2007:206), to the Catalan approach, which tends towards more comprehensive AD during the performance (Matamala and Orero 2007:205-207). It is important to acknowledge where these differing techniques originated, but these methods are now no longer limited to performances in individual countries. For example, in the UK pre-performance AI is no longer the only approach and some opera companies such as Opera North and Scottish Opera offer AD during the performance. Therefore, in this paper, AI is considered as a separate modality (pre-performance) and AD approaches will be explored in terms of the degree to which they overlap with the music. One opera AD approach which was, for example, practised by the Spanish Association for the Blind, Organización Nacional de Ciegos Españoles (henceforth ONCE), at the Liceu from 2004 until 2007 involved trying to "avoid overlapping on the lyrics" (Orero and Matamala 2007:270). Another approach, as was used for example by the Catalan Association for the Blind and Visually-Impaired, Associació Catalana de Cecs i Diminuïts Visuals (henceforth ACCDV) during the same period at the Liceu aimed "to deliver a complete AD of everything even if it meant treading on an aria or relevant music" (Orero and Matamala 2007:270) and "provides information not only on what is happening visually but also on the feelings of the characters and on the plot" (Matamala 2005:10). The new approach currently employed at the Liceu and which received a "mostly positive reaction" in the aforementioned trial (Cabeza forthcoming:6), integrates the various existing AD practices as well as AI methods. It offers "a concise audio introduction (lasting eight to ten minutes) of the whole opera before curtain up, including a coherent account of the plot, overall descriptions of costumes, characterisation and scenography, with the suggestion that this audio introduction could be posted on the Liceu's website for consultation prior to the performance" (Cabeza forthcoming:3). It also includes a summary before each act, and a comprehensive description during the performance in which "the elements to be described would be: costumes, hairstyles, stage and props, sounds not included in the libretto, plot and crucial moments of drama" (Cabeza forthcoming:3), but which does not overlap with the singers and therefore "never impede[s] on lyrics" (Orero and Matamala 2007:270). These varying strategies illustrate that the role of the opera audio describer is evolving and emphasise the complexity of this multifaceted task.

5.3 Audio surtitles (AS)

AS are surtitles which are read aloud, a method which at present is not widely used in the field of opera, but is considered a potential alternative to AD and perhaps a more viable option given that it is more economical and time-efficient. The audio describer or voice-talent and audio subtitler share the aim of helping VIPs to follow the plot of the opera as it progresses, whilst avoiding treading on the music too much, although this is inevitable to a considerable extent. They therefore assume the role of a narrator or commentator guiding the VIPs through the opera. There are many other similarities between these two tasks, but audio subtitles differ from AD in a fundamental way because whilst they provide VIPs with access to all the information on the subtitles or surtitles, they do not accommodate description of any other visual element of scenography or dramatic action. It might therefore be argued that this modality is better suited to concert version opera where there is minimal dramaturgical representation, and it was in this scenario that the ACCDV tested the reception of audio subtitles for Donizetti's opera Roberto Devereux, performed in Italian at the Liceu. This pioneering initiative in the field of opera was inspired by the Dutch television audio subtitling project "Spoken Subtitles" (Orero 2007b:141-142) and feedback from both was positive, even "very encouraging" in the opera context (Orero and Matamala 2007:273). In a questionnaire following the opera performance, one hundred per cent of the VIPs responded that they enjoyed "both the music and the commentary" (Orero 2007b:145). Interestingly, different presentation techniques were employed in these two ventures. Speech-synthesis software and equipment was used in "Spoken Subtitles", whereas in the Catalan opera experiment the audio subtitles were broadcast by an experienced voice-talent from a separate room (Orero 2007b:141-142). Orero reports the success of the approach adopted for the opera audio subtitles using the example of the voice-talent's specific strategy of making "clear cuts coinciding with syntactic units, which makes understanding much easier" (2007b:143), thus highlighting the importance of the performance aspect of the audio subtitler's role combined with technical competence. Orero and Matamala also partially attribute the favourable outcome of this project to the "proximity of Catalan and Italian" resulting in an "overall effect of voice-over and a feeling of reassurance in the listener" (2007:273). Therefore, they suggest that "other types of opera should be subtitled" and assessed, and that in general further research is required (2007:273). Although audio methods may be becoming more common than Braille amongst the blind community, another potential alternative to AS might be Braille surtitles, which could be delivered electronically to the Braille keyboards of individual audience members from the standard surtitles. However, this method, which could avoid interference with the VIPs reception of the music by employing the tactile channel of communication rather than the audio channel, also requires research.

5.4 Touch tours

Touch tours are visits to the opera stage and backstage in which VIPs have the opportunity to touch items of the set, stage design models, costumes and props, in order to gain information about these visual elements of the production prior to the performance. These tours are often accompanied by verbal description, including contextual information about the displayed items, and whilst mostly aimed at VIPs, may also be of benefit to people with developmental or cognitive disabilities as well as sighted audiences. These services are currently offered at some opera houses such as Scottish Opera one hour and fifteen minutes prior to all audio described performances as well as at Opera North, Welsh National Opera and the Grand Opera House, Belfast, and in the past have been tried out at the ROH. As has already been suggested, the VIPs are the translators themselves as they translate the information gathered by means of touch, such as texture and weight, in order to create their own mental image without any external agent. However, the tours are most likely to be guided, with the items for touching pre-determined and the time given for the VIPs to process these tactile experiences limited, and therefore their role as translators is also affected by external factors and constraints. This guided element of the tours, often seen as a way to "supplement the lack of detail that can be provided during a live event" (Udo and Fels forthcoming:2) whether via AI, AD or AS, also renders the participants' task multisemiotic, as they respond to tactile, audio and olfactory stimuli (see figure 1). Thus, the interactive aspect of their role is also highlighted, and as they feel the set and listen to information regarding the significance of props, they have the opportunity to experience dimensions and the geographical layout of the stage, which allows them to contextualise "a character's actions and onstage events" (Udo and Fels forthcoming:11) and understand the director's vision. Little if no research has been carried out into this area of touch tours for opera, although art education for the blind and visually-impaired is a dynamic interdisciplinary field of research, which involves cognitive, educational and social psychology, neuroscience and philosophy studies, and in which issues regarding touch tours have been explored ("Online Accessibility Training", 2009). This provides a solid foundation from which to build upon, with translation studies focusing on opera touch tour facilities. Udo and Fels' research into touch tours for theatre-goers also presents a useful model, advocating reference to well-documented theory and practice of museum-based touch tours and universal design theory (forthcoming:2), which could be applied to development of opera touch tours. Furthermore, the "overwhelmingly positive" results of reception studies, which reveal the enhanced "fulfilling and engaging" (Udo and Fels forthcoming:13) entertainment experience of theatre-goers due to touch tours, serve to prove the value of this modality within the field of dramatic representation including opera.

6. Translation techniques for the deaf and hearing-impaired

There are currently two methods of opera translation for making opera accessible to the deaf and hearing-impaired (henceforth the umbrella term hearing-impaired will be used to denote varying degrees of visual impairment including total deafness[3]): sign interpretation and captioning or adapted surtitles. The role of the translator of both of these modalities is to communicate the information and emotion of the acoustic elements of opera, as displayed below in figure 3, via alternative means, especially visual, although the translator may appeal to other senses through rhetorical devices, as mentioned before.

Figure 3

Again this figure is a non-exhaustive scheme and requires some elaboration. Music covers a vast range of elements including melody, harmony, rhythm, dynamics, textures, pitch and intonation, and instrumental musical aspects may comprise a pit orchestra as well as onstage or offstage instrumental groups. Language variation also encompasses multifarious factors, composed of diatypic variation which includes matters of register and archaisms, for example, and dialectal variation which refers to issues such as dialects and accents (Halliday 1978:110,35). Unfortunately the aforementioned task of the translator is often faced with resistance from "people who consider signed language and surtitling obtrusive and annoying practices that hinder the full reception of the visual component" (Orero and Matamala 2007:274). Sign interpreted opera performances are offered by some opera companies such as the ROH, ENO and Opera North but only for a limited number of performances, and other facilities for hearing-impaired opera audiences are neither widespread nor frequently available. This may also be due to the common misconception that the hearing-impaired cannot enjoy music and appreciate opera, a narrow-minded view which stems from not only an inability to recognise the multilayered and multisemiotic creative complexity of opera, in which a "mix of artistic disciplines" (Williamson 2008:31) play an important role, but also from a lack of understanding of hearing impediments. There is a large range of varying hearing abilities including "hard of hearing, profoundly deaf, deafened, cochlear implant user, deaf British Sign Language user and deaf SSE[4]user" (Shaw 2003:38) which must be considered, because "only a small percentage of hearing-impaired individuals do not hear at all" (Darrow 1985:33). Furthermore, research has proven not only that "hearing-impairment does not vitiate music responsiveness" (Darrow 1985:33), but also that listening to music is one of the musical activities most enjoyed by deaf individuals (Darrow 1993:105). Therefore, "a theatre that is aiming to be fully accessible to all deaf people will offer all three assistive devices: sound enhancement, sign language interpretation and captioning for their performances" ("Stagetext: Current Access Provision", 2003). Music is a multisemiotic experience and therefore can be appreciated on many levels by the hearing and the hearing-impaired (see figure 1). Evelyn Glennie, an internationally renowned percussionist, highlights the tactile element of hearing, stating "hearing is basically a specialized form of touch. Sound is simply vibrating air which the ear picks up and converts to electrical signals which are then interpreted by the brain. The sense of hearing is not the only sense that can do this, touch can do this too". She also emphasises the significance of sight in this process, confirming "we can also see items move and vibrate. If I see a drum head or cymbal vibrate or even see the leaves of a tree moving in the wind, then subconsciously my brain creates a corresponding sound" (Glennie). Furthermore, the entertainment value of opera for the hearing-impaired is reiterated by Orero and Matamala who state that "the deaf and hard-of-hearing community includes people who are partially able to hear, people who hear only certain frequencies and people who cannot hear anything at all but who can feel the vibrations of the orchestra, making opera an enjoyable experience for all of them" (2007:274). In short, these statements highlight the salient social role played by the translator whilst raising issues regarding the translator's visibility, both of which aspects will be discussed later.

6.1 Sign language interpreted performances

A sign language interpreted opera performance involves a sign language interpreter translating the opera for the hearing-impaired "usually standing to one side of the stage interpreting the language used by the performers" ("British Sign Language Interpreted Performances", 2009). These services are provided at selected performances by individual sign interpreters such as Wendy Ebsworth at the ROH, and across the UK by organisations including the charities Music and the Deaf and SPIT (Signed Performances in Theatre), in order to enable "people who use sign language, whether as a first or second language, to enjoy live performance in the language that is most appropriate to them" (Shaw 2003:11). There is little if no academic research into this translation modality for opera and therefore the following observations are drawn from attendance at sign interpreted performances at the ROH. The translator's presence on the stage highlights his or her involvement in the representation and yet subtle spotlighting and dark clothing ensure avoiding interference with the action on stage. In this way there is a tension between visibility and invisibility or unobtrusiveness, and the sign interpreter, whilst playing an integral part in the performance, remains discreet, rather like the orchestra. Nevertheless, the task of the translator entails a strong performance aspect because body language, including movement and facial expression, is employed alongside signing in order to portray the details of the plot, characterisation and emotions of the characters. Indeed, as a teacher writing to the ROH following a sign interpreted schools matinee opera performance states, "this has enabled our pupils to access the dialogue in their own language and to tune into the emotional experience expressed by the performers" (personal communication). Furthermore, musical phrasing and rhythm are sometimes indicated through hand gestures and dance-like bodily movements. There is also an evident interactive element to this role, not only with other visual features on stage, but also with the audience, as the sign interpreter faces the audience as if engaging them in a direct conversation. It is also interesting to note a comment made about the sign interpretation of an opera performance given by Westgate College for Deaf People as part of the ROH educational initiative, Write an Opera, that "signing adds to the drama of the performance" (Williamson 2008:34). In this case, the students performing the opera were themselves hearing-impaired and so signing was incorporated into the dramatic action of the opera, rather than as an additional service at the side of the stage, but this observation also applies to standard sign interpreted opera performances. Indeed, whilst providing an invaluable service for the hearing-impaired who have limited reading ability or those for whom sign language is their first or preferred language[5], sign interpretation also enhances the beauty and richness of the opera experience, thus underlining the translator's artistic role.

6.2 Captioning and adapted surtitles for the hearing-impaired

Captioning is a new assistive device for hearing-impaired audiences, which is used in live theatre and opera performances and "converts the spoken word into text, which is displayed on a caption unit that is viewed by the audience" (Shaw 2003:14). Sometimes the captions are presented to individual audience members via seat-back screens, hand-held screens and even special glasses (Shaw 2003:14), all examples of closed captioning. Otherwise, open captions are offered on "a screen on, above, below or beside the stage" (Shaw 2003:14) for everyone to see. Captioning "builds on the experience of television subtitles" (Shaw 2003:11), but this term should not be confused with the American use of the word "caption" which can refer to standard subtitles[6], because in the case of live theatre or opera performances, captioning always provides additional information on audio elements in particular for the benefit of hearing-impaired audiences, one of the main features which distinguishes captioning from surtitles. Indeed, it is the task of the captioner not only to relay the full text of the production, but also, where relevant, to include the acoustic features discussed in section 6 and figure 3, as well as to display character names and indicate asides. Therefore, this exercise entails intersensorial and intermodal transfer from acoustic features to verbal description and then to visual written script, thus confirming the role of the captioner as a translator. However, it is important to note that given the conventions within captioning of presenting the spoken text verbatim, this method is only used for intralingual translation. Perhaps it is due to this restriction that, whilst captioning is becoming more widely available in theatre, in the world of opera captioning is less widespread. There is some evidence that captioning is starting to be accepted as a valid opera translation method. For example, Stagetext, a leading UK company delivering and promoting captioning, founded in May 2000, trained staff from English Touring Opera as captioners and the company has its own captioning unit, and Wales Millenium Centre offered a captioned performance of Gershwin's opera Porgy and Bess on 24 October 2009. However, in order for this technique to be applicable to interlingual translation, which is often the case in opera performances, for example at ROH where all operas are performed in their original language, the convention of presenting the full text verbatim will require adaptation. A method which combines surtitling and captioning to create foreign opera captioning might be an option, especially in view of the similarities between these two tasks, although this proposition requires further research. There are some fundamental differences between the role of the surtitler and that of the captioner who has additional considerations when formatting the text to include supplementary information about acoustic features, as well as when outputting the text due to the differences in software equipment. For example, with captioning software the captioner can skip lines and clear the screen (Shaw 2003:31), thus allowing the captioner more spontaneity. Furthermore, the caption unit is moveable, so can be positioned in the most appropriate place for each individual production, also bearing in mind the preferences of some hearing-impaired audiences to be able to lip-read the actors as well as read the captions. This flexibility with software and hardware also allows the captioner to deliver "real-time messages" (Shaw 2003:27) such as emergency announcements, thus giving the captioner the safety responsibility of keeping the hearing-impaired audience informed. However, the tasks of the surtitler and captioner both involve a combination of technical and linguistic skill. Technical concerns have already been mentioned and a marked linguistic difficulty for both is overcoming the "levelling effect of the mode-shift and in particular the way in which features of speech which are in any way non-standard tend to be eliminated" in the written mode (Hatim and Mason 1997:79). A written style which reflects spoken language and specific tones of voice such as sarcasm might be indicated through appropriate punctuation, and phonological language variation can be signalled through labelling, for example "Scottish", or phonetic representation, for example, "Did yow call sir?" (Shaw 2003:16). They also share the challenge of dealing with space constraints, providing all the necessary information whilst avoiding distraction from the action on stage. Surtitle screens usually only have space for 2 lines of text whereas caption units allow for 3 lines, but with the additional information which must be provided, captioning space is still restricted. Both also aim for optimum readability, and the issue of the comparative reading speeds of hearing audiences and the hearing-impaired must be considered here. Strategies adopted by captioners for improving readability tend to be more varied, for example in changing the colour and size of text, but the role of both opera captioners and surtitlers entails a musical sensitivity regarding the timing and synchronisation of text delivery. Indeed, the recognition of the importance of semiotic cohesion in general is fundamental in both tasks, and therefore surtitlers and the captioners play an indirectly interactive role with the action on the stage as well as with the audience (see figures 1-3). Finally, both tasks involve a combination of teamwork and individual effort, as captioners and surtitlers attend rehearsals and discuss last-minute changes with the stage manager, for example, but text delivery is a solitary undertaking and in fact it is thought that the "best results are obtained when a single captioner prepares and outputs the text" (Shaw 2003:24). The task of adapting surtitles to render them fully accessible to hearing-impaired opera audiences, whether through combining them with captioning techniques or otherwise, will most likely face resistance from traditionalists, just as surtitles caused ample controversy when they were first introduced. Financial and technical considerations may also prove to be obstacles, as discussed below. However, as awareness of disability and accessibility issues increases, opera translation strategies will have to evolve to ensure the inclusion of all audiences.

7. Social responsibility of the translator

Social inclusion is not a concept which is traditionally associated with opera. However, the exclusive status of this unique artistic form has been challenged from the end of the nineteenth century, as testified by riots at Covent Garden in 1809 "in protest of price increase" (Desblache 2008:161), and opera company policies have been changed to attract wider audiences, as demonstrated, for example, by the aforementioned ROH initiatives involving open-air opera and multilingual cinema broadcasts. The role of the translator in this continuing evolution towards opera accessibility for all is fundamental, as shown by the previous discussions on the various new modalities of opera translation, and indeed this task entails significant social responsibility. This responsibility is heightened not only because of innovative technological developments which have increased the potential for inclusion, initially in the form of surtitles overcoming linguistic barriers and also making a social impact by promoting multilingualism, and latterly through facilities such as AD and captioning which contribute to the fight against the social marginalisation of the sensory-impaired, but also due to legislation. UK laws such as the Broadcasting Act 1990, the Disability Discrimination Act adopted in 1995 and amended in 2005, and the Communications Act 2003 enforce equality of services for disabled people and other members of the public. Furthermore, European directives and actions underline the importance of universal media accessibility. These include the "Television without Frontiers" directive adopted in 1989 by the European Union Council and amended in 1997 by the European Parliament, the Council Directive 97/36/EC (the European legal version of the US Telecommunication Act 1996), and projects by the European Committee for Electrotechnical Standardisation (CENELEC), "the institution in charge of media accessibility in Europe" (López Vera 2006:3). The recognition of accessibility as "a human right" means that it must be considered, perhaps most advantageously at the inception stage of design (Díaz-Cintas et al. 2007:14), a notion purported by universal design theory which is promoted by Udo and Fels as having the potential to result in "a more inclusive experience for all audience members" (Udo and Fels forthcoming:6). This concept in turn increases the social responsibility of the translator to aim to supply a service which not only improves access but gives full access to the complete opera experience. This might be achieved by offering various methods in conjunction with one another, for example providing touch tours to complement AI and AD, or may be accomplished by one single technique as is asserted by Stagetext regarding open captioning, stating that this system ensures that the hearing-impaired "have, as far as possible, the same experience as a hearing audience" (Shaw 2003:4). Through visible techniques such as open captioning and sign interpreting, the opera translator also makes an important social contribution towards increasing awareness of disability, as is stated in the Good Practice Guide for Open Captioning, "open captioning raises people's awareness of hearing loss, makes the production more accessible to everyone, and fosters inclusiveness" (Shaw 2003:16). The increased recognition of the social significance of accessibility within political, legal and other spheres also has financial implications. Although at present sources of funding for the development of facilities for improving access to opera are scarce, the need for economic support for the arts is becoming more widely acknowledged. This growing awareness, along with the heightened perception of opera houses regarding the potential for economic benefits in view of possibilities for increasing audience figures, presents an improved financial outlook. The need for financial support is particularly significant in developing translation modalities for the sensory-impaired due to the expensive audiovisual, technical equipment often required such as seat-back screens, and therefore economic considerations are fundamental. It might be suggested that opera companies seek ways to "widen their audience-base, for the sake of selling tickets and gaining sponsors" (Low 2002:98) therefore implying a more commercial than social role for the translator. Nevertheless, whether for social reasons or financial gain, or a combination of both, the various aforementioned opera translation modalities, especially those for the sensory-impaired, confront accessibility issues at the forefront of current political and social concerns, and thus emphasise the translator's social responsibility. Furthermore, these opera accessibility strategies have the capacity to improve social integration of not only the sensory-impaired but also "people with language impairments" such as dyslexia, as well as those with developmental or cognitive disabilities (Orero 2007b:141). This widened audience-base, thus presenting a more economically viable venture, may also help to attract a more diverse spectrum of sponsors keen to publically support a topical social cause. Indeed, the social and financial aspects of opera accessibility are closely interlinked, and another related factor which may contribute to positive financial possibilities is the improvement of the status of the translator.

8. Status of the translator

Despite the salient social impact of new professional translation practices within an increasingly multimedia society, the audiovisual translator is still perceived to hold inferior status to literary or specialist translators. It is generally considered that as a whole, opera translation falls under the category of AVT, although some types of libretto translation are thought to be "literary translation" (Low 2002:100). Therefore, the opera subtitler, surtitler, audio presenter, audio describer, audio surtitler, sign language interpreter and captioner are, as audiovisual translators, each seen as "un acteur passif, invisible, mal reconnu" (Gambier 2009:152). Indeed, these translators hold little prestige in academic settings "due to the consideration of AVT as a lesser manifestation of literary translation" (Chaume 2002:4), hence the relatively small amount of written scholarship about these translation modalities. This inferior status may also be due to the lack of attention paid to opera translation within its own professional setting in terms of written standard practices and guidelines, for example, although an increasing amount of literature is being written by expert practitioners in the field, as demonstrated by articles by Jonathan Burton, Gregory York and the company Stagetext amongst others (Burton 2001, 2009; York 2007a; Shaw 2003). There is also an increase in academic scholarship due to growing interest in media accessibility which according to Orero "points towards a level of professionalization in the job which will go in tandem with research" (Orero 2005:10). Indeed, the professional status of translators of media accessibility techniques for the sensory-impaired, such as audio describers, is improving, because whilst AD was initially performed by volunteers with no formal training, there are now professional training opportunities at private and public institutions as well as postgraduate courses, for example at Roehampton University and Surrey University in the UK, and at UAB and Universidad de Granada in Spain (Orero 2005:10). There is still a long way to go to achieve full professional recognition of translators of these translation methods for the sensory-impaired, as for example, even nowadays "the profile of audio describer as a profession does not exist in most countries - with the lack of social recognition and legal protection" (Orero 2005:15-16). However, progress is being made and Orero affirms, "audio describers and subtitlers (SDHH)[7] will soon enjoy social recognition, which they have not at the moment - taking into consideration copyright, working conditions and open market prices" (2005:10). As well as promoting greater social recognition, public and academic discussion about the role of the audiovisual translator and opera translator may also enhance financial prospects for the development of media and opera accessibility, by making the fundamental role of the translator in improving access to information and opera more widely known. There are also advancements regarding the authority of the opera translator, although this varies according to the translation type. The libretto translator of a singing translation, such as Eric Crozier for example, is more likely to enjoy authorial distinction than surtitlers and translators for the sensory-impaired whose authority is subject to the director. When surtitling was first introduced, surtitlers suffered considerable opposition from directors, most notably from David Pountney, former Director of Production at ENO, who "famously compared surtitles to 'a prophylactic between the opera and the audience'" (Burton 2009:5). More recently, however, some directors and composers specifically request surtitles, for example "Harrison Birtwistle for his opera Gawain" (Burton 2009:6), thus increasing the translator's influence. As new translation modalities for the sensory-impaired are established within opera, there is a similar resistance in recognising their value and indeed the authority of the translator. However, steps are being taken to improve their status, as demonstrated for example by the following quote from "A Good Practice Guide to Open Captioning": "the production team needs to see the captioner and technician as equal members of the production team" (Shaw 2003:30). Greater internal recognition of the translator as an integral part of the production and increasing visibility from the audience's point of view, whether due to physical presence on the stage or written acknowledgement in programme notes for example, may help to raise the status of this profession, although the question of the visibility of the translator within this field of opera remains a complex, polemical issue.

9. The illusory invisibility of the opera translator

The concept of the translator's "invisibility" conceived by Venuti (1995:1,4-5,17) raises, amongst others, issues of fluency, partiality and recognition, all of which form the subject of topical debates within the sphere of opera translation. In the following discussion, the visibility of the translator is considered in relation to these subjects, particularly in terms of the translator's aims regarding the obtrusiveness of his or her intervention on the original text or representation, as well as in terms of physical visibility. As in all types of translation, the opera translator's role involves making necessary choices and prioritising as regards objectives to be achieved within the context and confines of the given translation modality, therefore rendering the task inevitably visible. Indeed, due to the wealth of semiotic aspects of opera enumerated earlier (see figures 1-3) all opera translation modalities demand substantial decision-making and intervention on the part of the translator. However, this need is not always recognised, as already discussed above with regard to authorship and a lack of recognition of the translator as a valued member of the production team, leading to invisibility in terms of status. By way of example, the translator's decision processes are examined below in the case of surtitling, AD and AI, contemplating in particular the translator's choices with regard to embracing visibility or pursuing the aim to maintain the illusion of invisibility. This discussion highlights the tension between Venuti's dichotomy of the translator's visibility and invisibility (Venuti 1995:1,4-5), an issue which is currently particularly prominent given the introduction of new methods of translation for the sensory-impaired, or indeed the adaptation of existing techniques, which cause the balance of priorities to be reassessed.

9.1 The inevitable and potentially increasing visibility of the translator in surtitles

This friction between the translator's invisibility and visibility, and the change of emphasis regarding this division are particularly well exemplified within the practice of surtitling. Like subtitles, although invisibility is the aim, by their very nature as co-existent with the source language at the bottom of the screen (Hargan 2006:68) or at the top of the proscenium, interlingual surtitles are evidently a translation. Furthermore, due to the necessity to manage the restrictions imposed on the surtitler by the shift in mode from speech to writing, by constraints of space and pace causing reduction of the source text, and by requirements to match the visual image (Hatim and Mason 1997:78), methods which render the translator more visible may be required. Nevertheless, just as Díaz-Cintas and Remael suggest that subtitling should be "preferably 'invisible' work" (2007:192), Burton states about surtitles that "the titler's aim should be transparency or even invisibility" (2009:7). This approach adheres to "the basic principle of surtitling [...] to avoid distracting the attention from what is happening on the stage" (Dewolf 2001:181) and is guided by an emphasis on fluency and readability, bearing in mind the limited amount of time the audience has to read the surtitles. Indeed it follows Venuti's notion that "under the regime of fluent translating, the translator works to make his or her work 'invisible', producing the illusory effect of transparency that simultaneously masks its status as an illusion: the translated text seems 'natural' i.e., not translated" (1995:5). However, the surtitler remains inevitably "visible", as implied by Venuti's reference to illusion and supported by Balfour's statement about subtitles, which can also be applied to interlingual surtitles, that they "are the marks of difference, the written words that visibly render the voice of another language" (Egoyan and Balfour 2004:532 in Hargan 2006:68). If surtitles were to be adapted for the hearing-impaired, this inevitable visibility of the translator would have to be embraced to a certain extent, as the additional information required (see figure 3) would demand techniques such as the use of brackets, essentially a translator's note, as well as more subjective input regarding the description of music, for example. Indeed, in captioning, such methods are employed, thus heightening the translator's visibility in linguistic terms and also with regard to authorship. This visibility is also boosted in terms of public profile and recognition by increased efforts of companies such as Stagetext who have initiated schemes "to raise awareness of captioning" (Shaw 2003:37). Nevertheless, as regards translation methods, the tension between the translator's visibility and invisibility remains, because whilst individual strategies which increase the translator's visibility might be used to provide additional information, the overall approach may tend towards being discreet or even invisible. As Burton states, "the surtitler must remember that the audience has come to see the opera not the surtitles; the titles should be discreet and not distracting" (2009:7), although paradoxically in order to achieve this aim of unobtrusiveness, more "visible" translation techniques, as mentioned above, may need to be used to produce a more concise and accessible option which avoids distraction from the other various semiotic aspects to which the audience has access (see figure 1).

9.2 The tension between the visibility and invisibility of the translator in AD and AI

Achieving this balance between visibility, accessibility and unobtrusiveness is particularly complex in AD, and as Holland states "in order to be non-intrusive a description has to make decisions and not pretend it is not there" (2009:184). In AD the translator has the challenge of providing sufficient information about the visual elements to which the visually-impaired have limited or no access (see figure 2), including the plot details, especially if the opera is sung in a language unfamiliar to the audience, whilst also trying to avoid treading on the music to which the visually-impaired do ordinarily have access. An audio describer will inevitably interrupt the music at times and the extent of this varies according to different AD strategies, although Puigdomènech, Matamala and Orero state that "it has to be agreed that music should be respected above commentaries" (2008:383). In view of the possible implicit message and emotions frequently portrayed by the music, this awareness of the importance of the music is particularly crucial if it is considered that "providing an audience with the subtext or the unspoken message of a production is often as important as conveying facts, if not more so" (Díaz-Cintas and Anderman 2009:13). Bearing in mind the communicative aspect of the music, it could be argued that AI provides the VIPs greater access to the subtext of any given opera than AD, as it is pre-performance, therefore allowing the audience to "listen to every note of the music without interruption" (York 2007a:3). In this respect the audio presenter of AI remains somewhat hidden or "invisible", because the audience is free to interpret the implicit meanings of the opera conveyed via the music without aid or intrusion, and therefore parallel implications reiterated through visual means may not need to be explained by the translator. For example in Wagner's opera Tristan and Isolde it might be considered that the famous Tristan chord which Wagner leaves repeatedly unresolved, although open to differing interpretations, could be sufficient in conveying the thematic sense of longing in this opera as well as the building tension, such that the characters' facial expressions and gestures which reiterate these emotions may not always need to be described. Another example of the music conveying the emotional meaning, arguably without requiring verbal description, was observed in the performance of Britten's opera A Midsummer Night's Dream at the Theatre Royal in Newcastle on 5 June 2008 during the scene in Act IV in which Bottom and his friends put on a play about Pyramus and Thisbe. It might be considered that, in an AI which allows full uninterrupted access to the music, the effect achieved by glissandi in the brass instruments during this passage could be sufficient in portraying the comic self-importance and clumsiness of the characters without verbal description of comic visual aspects. The audio describer, on the other hand, may have to provide more subjective details about implicit elements of visual features, in order to compensate for the loss of implied meanings conveyed by the music which may be interrupted by the AD (see for example how the semiotics of opera work in figure 1 and especially the visual aspects in figure 2). However, on occasions individual visual elements alone may communicate subtexts or symbolism which the translator may want to explain to the VIPs and whereas the audio describer will have more opportunity to include this information, in AI these details may have to be omitted due to greater time constraints. These different advantages and disadvantages of AD and AI may have contributed to the motivations for the aforementioned UAB project at the Liceu to adapt an opera AD strategy which combines AD and AI, in order to find a method which balances visibility, impartiality and unobtrusiveness, whilst communicating the sufficient information required for the VIPs to enjoy the full opera experience. The audio describer certainly has more opportunities than a presenter of AI to describe visual elements, and therefore may have the potential to be more impartial and therefore "visible". However, both the audio describer and the audio presenter of AI will most likely have to express some personal interpretations of visual elements, and it might be suggested that this impartiality and subsequent visibility are inevitable.

9.3 Impartiality in AD and AI

The tension in AD between the translator's subjectivity and objectivity, which plays a part in Venuti's visibility and invisibility dichotomy (Venuti 1995:32-34), is discussed by Holland within the context of theatre (2009), although many of his observations are also applicable to opera. Although from a background of training which encouraged objectivity (2009:173), Holland advocates an element of subjectivity, highlighting the limitations of impartiality, as he says "there is no direct equivalent between a moment on stage and the words chosen to describe it. The exhortation to be 'impartial' doesn't recognise this fact" (2009:184). There are, however, varying stances on this topic of impartiality and differing AD and AI traditions for different aspects. For example, York highlights the necessity at certain times of "telling it as it is", referring specifically to the importance of naming objects accurately, "however inappropriate it might seem", using the illustration of the description of a bucket in Verdi's Simon Boccanegra at ENO, an expression which was considered by the dramaturge to be "undignified" but was ultimately named simply as a bucket in the AI (York 2007a:10). On other occasions, however, York recommends subjectivity, stating for example, "you have to characterise the manner in which the character appears", such as "stride determinedly" (2007a:12-13), also suggesting the inclusion of specific details of a costume to "illuminate the nature of the character", such as "an ostentatiously-trimmed hat" to imply "the vanity of its wearer" (2007a:9), thus employing evocative adverbs which reveal impartiality. One of the most complex issues causing debate over this subject concerns the expression of emotion, and in the US, for example, "a total objective depiction of the emotion" is promoted, as is the case for attractiveness (Puigdomènech et al. 2008:387). On the contrary, Holland suggests that AD "can never be transparent. By its very nature it will change the experience someone has of the art" (2009:183), thus supporting a notion of inevitable impartiality and visibility due to the choices made by the describer which influence what the audience focuses on. With this approach in mind he also encourages collaboration with members of the production team (2009:173-174), a strategy which is also adopted by York who sometimes includes a "recorded interview with the designer" in the opera AI, although this is more common in ballet AI where an interview with a choreographer "can really bring the introduction alive" (York 2007a:7,10). Bringing vitality and reality to a production, is generally considered a crucial requirement of AD and AI, as, for example, Holland states "description should aim to get to the heart of a work of art and to recreate an experience of that work by bringing it to life. It should not be content with telling someone the physical details of something they cannot see" (Holland 2009:184). One of the principal methods in maintaining the illusion of realism is to avoid theatricality or technical terminology, for example by refraining from making references to "props", the "teaser", or the stage (York 2007a:10; Puigdomènech et al. 2008:387), as well as taking care over choices of vocabulary to prevent use of verbs such as "exit" and prepositions such as "on" and "off" which "can suggest theatricality" (York 2007a:12-13). These techniques contribute to the creation of a real event for the audience, thus somewhat diminishing the visibility of the translator, although paradoxically this approach also inevitably entails an element of impartiality because the translator has to make choices in order to strike a "balance between the literal truth and imaginative truth of something" (Holland 2009:175), thus highlighting the translator's presence. The tension between the translator's invisibility and visibility is therefore reiterated within these modalities of AD and AI. Furthermore, a question is raised regarding the effect on the audience's reception of any given opera production, in view of changing practices in these and other opera translation modalities which embrace the translator's inevitable visibility to a greater extent. This area of audience reception within the field of opera has so far been accorded little if no academic research. However, it would seem that the introduction of new translation methods to improve accessibility to opera, including surtitles and modalities for the sensory-impaired, is gradually altering audience attitudes towards the acceptability of the translator's visibility. Indeed, this change is illustrated by the changes in opinions of surtitles discussed in section 8 and the increase in audience figures since the introduction of surtitles. This is also demonstrated by the increasing number of performances with AI offered. For example, initially "just one or two introduced performances" per production were provided, whereas now "all ENO performances at the Coliseum and all Royal Opera and Royal Ballet performances at the Royal Opera House are introduced by Talking Notes" (York 2007a:3). In general, the translator aims to maintain the illusion of invisibility, but in some opera translation methods, the visibility of the translator from the audience's point of view, is more evident. This is the case, for example, in sign interpreted opera performances due to the translator's physical presence on the stage and ample recognition of this facility, as demonstrated by a full page dedicated to British Sign Language in the cast list leaflet of the ROH schools matinee performance of Don Giovanni, on 25 September 2008. This kind of visibility is not yet tolerated by all audiences, but as policies change to embrace the notion of accessibility which seeks to provide as equal as possible an entertainment experience of opera for all, translation practices and the role of the opera translator will continue to evolve. This progression towards inclusiveness may also be furthered by an increase in visibility of the translator in terms of status, authorship and professional recognition by promoting awareness of the demand for accessibility strategies and of the value of such facilities. In order to achieve this heightened professional recognition of the necessity of the translator's intervention and increased authorial visibility, expert practitioners must be consulted, establishing links between academics and practitioners, theory and practice.

10. Discrepancies between theory and practice

At present there are still distinctive discrepancies between the ideals of translation theory and the realities of translation practice, and therefore having explored the role of the opera translator from a theoretical standpoint, it is important to consider the perspectives of expert practitioners in the field. These divergences concern terminological issues, as well as practical applications, which may be determined by factors such as commercial considerations of marketability and professional constraints including time pressure. For example, practitioners appear to tend towards protecting their own market and promoting their product as individual and different from others, as in the case of captioning where the distinctions of this method compared to surtitling are highlighted by Stagetext who state, "captioning is different from surtitling and subtitling" (Shaw 2003:31). Indeed, there seems to be a general reluctance to amalgamate methods even if they have similar applications, such as with surtitles and captioning, as well as a reticence towards experimenting with either adapted or innovative opera-specific translation techniques. Furthermore, although Stagetext promotes open captioning for all audiences (Shaw 2003:14,16), in general there seems to be an inclination within opera towards keeping services for the sensory-impaired separate from the services for hearing and seeing audiences. For example, Judi Palmer, surtitles coordinator at the ROH, suggests that adapting surtitles on the main screen above the proscenium for hearing-impaired audiences might cause offence due to the excess of information for hearing audiences. She does consider surtitling for the deaf and hard-of-hearing on seat-back screens a feasible possibility "as an option that could be selected by patrons should they wish to use it", as this would not interfere with the hearing audience's reception of the opera and also because there is additional space on these screens which have room for three lines of text as opposed to the two lines on the main surtitles screen, but she also highlights potential problems regarding technical and financial considerations (personal communication). This tendency amongst practitioners towards establishing opera translation methods as distinct and separate entities is contrary to the movement in the academic world towards theories which propose all-embracing notions of translation and which encompass an increasingly broad spectrum of overlapping translation types, as illustrated by Gottlieb's semiotically-based taxonomy of translation with 30 types of translation which challenge traditional categorisations (2005:11). Despite this "gap between theoreticians and practitioners" (Gottlieb 2005:16), favourable partnerships have been forged, such as the aforementioned collaboration between researchers at the UAB and professional practitioners at the Liceu, resulting in the successful adoption at the Liceu of an innovative method merging AD and AI techniques. Bridging this gap is the task of both academics and practitioners, and requires further cooperation between these two parties to share insights and discuss ideas with the purpose of finding the most appropriate and accessible opera translation methods which take into account practical and theoretical factors, including social and financial concerns, as well as requirements of the translator and the audience.

10.1 Findings from interviews with practitioners

Bearing in mind the task of bridging the gap between theory and practice through increased discussion, interviews were conducted with the expert practitioners Judi Palmer (i) and Gregory York (ii) at the ROH on 22 June 2009, and a questionnaire was sent to Cristóbal Cabeza (iii), audio describer at the Liceu, and a summary of their responses are discussed below. Firstly, the interviewees were asked for their views on the main function of surtitling, AI and AD respectively (1), and secondly how they see their role within these processes (2). The following responses highlight the multifaceted task of the opera translator and the emphasis on accessibility: (1) (i)"To make opera more accessible to a hearing audience." (ii)"To make productions as real as possible for somebody who cannot see. To give enough information so that they know what it is about and what it all looks like." (iii)"Making opera accessible for blind and visually impaired people" (2) (i)"Jonathan (Burton, surtitler at the ROH) is a (there are a number of others that we use) translator. I am the technician. I see my role as a communicator." (ii)"Producer, writer, studio manager, technician, translator and interpreter." (iii)"Translator, narrator, author and communicator." Regarding particular difficulties of their tasks, Palmer and York emphasised the challenge of space constraints, whereas Cabeza mentioned terminology problems (when describing costumes, scenography, props etc.) as well as issues of synchronisation due to variances during individual live performances. York also referred to the difficulty of accommodating last-minute changes, as well as the restriction within pre-recorded AI of neither being able to accommodate spontaneous stage business during the performance nor maintain any elements of surprise. For example, in the ROH final dress rehearsal of Verdi's opera Un ballo in maschera on 23 June 2009, York noticed that in Act II Riccardo's costume had been changed so that he was now wearing the "dark green uniform tunic" from earlier (see picture 4 below), rather than "his fisherman's disguise" as reported in the AI, but at this point it was too late to alter the AI script.

Picture 4: Angela Marambio as Amelia and Ramón Vargas as Riccardo (photo by Catherine Ashmore, courtesy of the ROH)

Also, in the AI for Un ballo in maschera, the surprise element is removed from the otherwise spectacular revelation in Act III of a second lower ballroom which is reflected in a large mirror facing the audience. This scene is shown below in picture 5, followed by the excerpt from the AI describing it, which the audience could listen to during the interval before Act III.

Picture 5: Ballroom scene (photo by Catherine Ashmore, courtesy of the ROH)

Now comes an absolute coup de théâtre. The vast mirror angles backwards to reflect not only the front of the stage but also a large lower salon some fifteen feet below, and suddenly the monochrome stage pictures we've witnessed are changed to brilliant colour because the carpet of this lower room is a rich claret red. Guests enter the salon through a pair of double doors in one wall or by two staircases, one each side of a balustrade. As people descend them, they appear to be going up the scene as it is reflected (York 2009).

York commented that this removal of elements of surprise in AI can also detract from humour, because whilst VIPs can anticipate the humour, the joke will lose its sting. Regarding humour, York affirms that comedy can be conveyed to some extent in AI through humorous description, and Cabeza states:

I don't normally change my AD in operas where there is some humour. Users get humour (if there is any) through the opera itself and the description of what the characters do and what they say. Also sometimes through the way they sing. I've never done a humorous description, since I always try to be a narrator who does not take part in the opera, that is, who does not interpret what they should get from the opera, but help them understand what the opera itself is trying to convey.

Another issue which was discussed is register, and whereas both York and Palmer state that the formality of the language depends on the production or the period and tone of the opera, Cabeza declares that he always uses formal language whilst taking into account that it is written language to be read aloud and that he never uses slang. York, on the other hand, suggests that it is sometimes appropriate to use slang in some operas such as Mozart's Don Giovanni. The variability of choice of register according to the type of opera and indeed character is also confirmed by Palmer, who states that although in general a simple, concise writing style which avoids idiomatic phrases is adopted, "the style of writing, use of colloquialisms and contractions can also be determined by the character: whereas Leporello might say 'things are hotting up', Don Ottavio would use more formal language and we would try to reflect that in the titles", referring here to the surtitles by Kenneth Chalmers for the ROH performance of Don Giovanni on 25 September 2008. The differing approaches adopted by these practitioners who all report that they have had to develop their own guidelines to some extent also points towards the enterprising and individual role of the opera translator. Finally, the topic of the visibility of the translator was addressed, and responses reiterated the aforementioned tension between visibility and invisibility. Palmer confirmed the conventional aim for invisibility in surtitles, and Cabeza declared a dichotomy between his role of invisibility for the general public and visibility for AD users, also affirming the importance of visibility for the social recognition of AD. Although York preferred not to think in terms of visibility or invisibility, talking rather about "personalising and making oneself real", his responses also reaffirmed a struggle between invisibility and visibility because whilst he inclined towards declaring invisibility at first, given "a target audience to whom everything is invisible", he then implied visibility by stating "I think it is very important that the presenter is a real person, a companion for the patron" as demonstrated by the traditional opening to each AI, "Hallo. Greg York here". He also asserts that he considers his role as subjective, as he says "I interpret what I see on stage", and this subjectivity is exemplified in the AI for the ROH performance of Un ballo in maschera at ROH in June and July 2009. For example, the first scene in Act II is described as "a truly dreadful place" and Amelia is described as "a handsome woman". York also suggests that from the perspective of the audience, the AI might be viewed "as part of the performance, as a sort of prologue" thus highlighting not only the versatile nature of opera but also the multifaceted role of the opera translator as creator, narrator and performer.

11. Concluding remarks

Opera, opera translation and the role of the opera translator are all epitomised by a notion of multiplicity. The multiple types of opera and narrative elements of opera, along with its multilingual, multimodal, multisemiotic, multidisciplinary and increasingly multimedia nature, render this art form unique in its creative complexity. Opera translation is consequently equally multifaceted as the strategies, challenges and general role of the translator are determined by a multitude of variables. The translator's task is not only affected by the aforementioned innate multiplicity of opera and the opera experience with its varying audio, visual, tactile, olfactory and gustatory elements, but also by the individuality of each opera production and indeed performance, open to multiple interpretations by both different directors and audiences, thus highlighting the flexibility and potentially interactive role of the translator who may assume the role of performer. Furthermore, technological developments demand increasing technical expertise from translators and enhance the translator's visibility. Indeed, these advancements reflected in developments within the field of AVT, seen as an "extremely volatile translation form" (Remael et al. 2008:76) of which opera is considered a marginal genre, contribute to the constant evolution and multifaceted nature of the role of the opera translator whether librettist, surtitler, subtitler, audio describer, audio presenter, audio subtitler, sign interpreter or captioner. Also, changing attitudes towards accessibility, coupled with the characteristic diversity of opera which lends it to increased accessibility for all, are bringing about the introduction of even more target-oriented translation strategies within this entertainment medium, which by its very nature demands focus on audience reception. Consequently, the role of the translator increasingly entails promoting and celebrating the intrinsic universality of opera, as well as forging links with researchers in order to find the most effective and appropriate translation methods for diverse audiences. The constraints and subsequent challenges imposed by the audiovisual medium heightened within opera translation modalities for the sensory-impaired, also call upon the translator to go beyond the rules, thus aspiring to heights of creativity and flexibility, whilst also aiming for consistency and synchrony between the audiovisual and textual elements of this multisemiotic mode. Therefore, having examined the theoretical considerations as well as the commercial and practical implications of the various new modalities of opera translation, it can be seen that developments within this field have had a significant impact on the role of the translator which is evidently a fundamental although often undervalued component in the path towards accessibility for all to the complete opera experience. The main effect appears to be the notorious complexity of the translation task due to the concurrent variability and multiplicity of the elements of opera and its audiences, as well as the constraints imposed, most especially by the audiovisual mode. Indeed, paradoxically it is its multisemiotic nature which lends opera to inclusion rather than exclusion, but also renders its translation strategies towards accessibility for all particularly complex. The translator's task becomes an elaborate juggling act between visibility and invisibility as well as diversity and marketability. Ultimately, therefore, the translator is inspired to embrace a creative, dynamic and above all multifaceted role, with the purpose of stimulating the audience's imagination and metaphorical thinking to open the doors to the rich, beautiful and diverse world of opera.

Acknowledgments I would like to sincerely thank the following people for their invaluable contribution towards my research: Dr Federico Federici, University of Durham; Dr Pilar Orero and Cristóbal Cabeza, Universitat Autònoma de Barcelona; Greg York, Talking Notes; members of staff at the Royal Opera House, Covent Garden including among others Jonathan Burton, Tessa Forsey, Stanley Hamilton, Greg Jauncey, Paul Reeve, Clare Thurman, Eleonora Claps and in particular Judi Palmer; Judy Dixey, VocalEyes; Lynn Jackson, Stagetext; and all those who kindly provided articles and photos, including Catherine Ashmore.

About the author(s)

Sarah studied for a BA in French, German and Italian and an MA in Translation Studies at Durham University where she will be starting PhD research in October 2010. As a linguist and keen musician with experience of disability, her research interest is in opera accessibility and translation for sensory-impaired audiences. This is the topic of her MA specialism, a forthcoming paper to be presented at the Turin ESSE conference 2010, and her prospective PhD studies. Sarah is currently working full-time for a translation company near London, as well as pursuing individual freelance translation assignments and research projects.

Email: [please login or register to view author's email address]

•  Previous article 
•  Next article