Scripted or spontaneous? Two approaches to audio describing visual art in museums

ABSTRACT

We report on a comparative analysis of two approaches to live audio-describing (AD) visual art in museums: the first case is a tour with scripted AD (the guide reads written descriptions out loud), and the second case is spontaneous AD (AD is intertwined with the guide’s talk). As previous studies have mostly analyzed pre-recorded AD, our aim was to describe how AD occurs in and as direct interaction between a museum guide and visitors, and how interaction affects the art experience of the (blind and partially sighted) visitors. Data were collected from two authentic settings in which groups of blind, partially sighted, and sighted people visited art museums on guided tours. The data consist of video recordings of the tours and retrospective interviews with visitors. The analysis revealed how the interactive constitution of the tour and the AD format enables or disables the visitors’ participation in experiencing visual art. Most importantly, we show how AD-enriched interaction between the guide and visitors facilitates joint meaning-making about vision and art, in which visually disabled visitors actively participate with multifaceted communicative practices and resources. Our study contributes to the research on (live) AD, demonstrating the role of interaction in the process.

KEYWORDS:

• Audio description
• art
• visual impairment
• experience
• meaning making
• interaction analysis

Transcription symbols

In the transcripts, the text in bold signify AD sequences in speech, the text in cursive font below speech describes non-verbal, bodily actions. The transcription symbols follow the conventions of Conversation Analysis (e.g., Arminen, 2016).

Verbal and non-verbal action:
[…]	=	square brackets indicate overlapping utterances or sounds
(…)	=	omitted part of speech or action
* … *	=	stars indicate the beginning and end of a non-verbal action
-->	=	continuation of a non-verbal action
//	=	A pause longer than 1 second
/	=	A pause less than 1 second
↑kuvassa	=	Raised pitch level at the beginning of a word
.	=	Falling intonation
;	=	Slightly falling intonation
,	=	Continuing intonation
¿	=	Slightly rising intonation
?	=	Rising intonation
kumarassa	=	Stressed word (underlined)
((…))	=	Transcriber’s account of speech or sound
vin-	=	Cut-off word
hhh.	=	Outbreath
NN	=	“no name’; names have been deleted to mask the data

1. Introduction

For quite some time, audio description (AD)—the verbalization of visual contents in visual and audiovisual culture and communication—has been actively researched in translation studies. Scholars and practitioners alike have long been aware of the different forms of producing and delivering AD, which depend on the contexts of use: AD can be made of static or dynamic source material, it can be delivered live or recorded, and it can be produced in a non-scripted or scripted form (see, e.g., Orero 2006, pp. 284–286). Of the genres of AD, museum AD has been practiced and taught less than film and television AD, but more than theater AD (Perego 2019, p. 334 citing the ADLAB PRO reports 2017a and 2017b). While museum AD in various forms and some general principles of good practice exist, ‘a comprehensive overview of the linguistic and textual patterns characterizing this text type is missing, not to mention the fact that we currently lack systematic reception studies illustrating how museum AD is received and processed by end–users’ (Perego, 2019, p. 334).

In our article, we contribute to the research on AD forms and genres by comparing ‘scripted’ and ‘spontaneous’ (non-scripted) AD as these are delivered live in face-to-face interaction with users.¹, ² By (social) ‘interaction’ we mean the concept originating in sociology (e.g., Goffman, 1961) of communicative situations in which at least two people (co-participants) are present to each other and engage in exchange multimodally, that is, using speech and body to communicate and to act together (see vom Lehn & Heath, 2016, p. 361).

Furthermore, our study sheds light on how the two interactive formats are received and experienced by users (cf. Perego 2019). Responding to Hutchinson and Eardley’s (2018) call for AD that enables museum and art experience and does not merely provide factual information about artworks, we analyze visitors’ experiences (see Pekarik et al. 1999), showing which dimensions of experience become realizable in which form—scripted or spontaneous.

We begin by reviewing previous research on the AD of art and museums as well as on the interactive dimension of museum and art experiences in general. In Section 2, we describe the two cases, data, and methodology. Our research is based on video-recordings of two authentic guided tours in Finnish art museums. After the tours, volunteer visitors were interviewed about their experience with the AD. In Section 3, we report on our data analysis by compiling the focal interactive structures, by reconstructing samples of interaction for each case, and by analyzing the interviews. In Section 4, we discuss our results in the framework of museum experience, and point to future directions of research on AD, the accessibility of art, and museums.

1.1. Translatory and linguistic aspects of museum and art AD

Museums are central spaces for aesthetic experiences, knowledge acquisition, and cultural education (vom Lehn & Heath, 2016, p. 363). Translation features as a central transfer phenomenon in the museum context at least in two ways: first, museums perform cultural translation through their selective and narrative practices of representing culture and, second, museums use methods of inter-, intra- and intermodal translation and interpreting to cater for multilingual and -sensory communication (see Neather 2018, p. 361). The hitherto modest amount of research on translation in museums (ibid., p. 367–368) is now expanding, in particular in the area of translation-driven accessibility (see e.g., Rizzo 2021; the references below).

Among translation-driven accessibility, the audio description of visual art is a significant research object. To date, two lines of research have dominated: proposing methods for AD (De Coster & Mühleis, 2007; Neves, 2012; Soler Gallego, Luque Colmenero & Rodríguez Posadas, 2016), and characterizing the discourse, or typical features, of AD (Soler, 2015; Perego, 2019; Hirvonen & Wiklund, 2021). These characterizations are based on scripted, pre-recorded AD (either audio descriptive audio guides or recorded museum AD). However, the AD research seems to be shifting from the perspective of accessibility of artworks to a more holistic view of accessibility of experience (see Eardley et al., 2017, p. 199). Following this new perspective, visitors should be able to recall whole experiences of art instead of remembering descriptions of single artworks (Eardley et al., 2017, p. 201). Thus, visits with AD should provide for social, cognitive, and emotional elements of experience (Hutchinson and Eardley, 2018, p. 51, following Pekarik, Doering and Karns, 1999). This shift of focus calls for the study of museum visits and live AD as social interaction and collective experience.

1.2. Interactive and experiential aspects of museum visits

Previous research on museum visits has analyzed several aspects of visitor behavior (see vom Lehn & Heath, 2016) and, more recently, interaction between the guide and visitors in guided tours (Best, 2012; Mondada, 2017). Visitors’ behaviors and their interactions with each other and with the artworks have aroused interest in, i. a., structures of walking and bodily orientation in the museum space (vom Lehn & Heath, 2016, p. 363) or how visitors look at artworks (Smith & Smith, 2001; di Giovanni, 2020). Another interest is in learning: how people turn to social interaction, objects, and digital devices in museums to enhance their learning (vom Lehn & Heath, 2016, pp. 364–365). The third area focuses on the organization of social interaction during visits (ibid., p. 366). Among the found regularities of social organization during visits is that visitors tend to respect the individual ‘use space’ (Goffman, 1971) so that they rarely cross the sight line of another visitor looking at an object (vom Lehn & Heath, 2016, p. 367), and people respond to fellow visitors’ bodily orientation as clues for where others’ attention is directed (ibid., p. 369). Furthermore, the nuanced analysis of the interaction between the guide and visitors has shifted the traditional perception of museum guidance from a monologic performance to a joint accomplishment, as well as it has changed the view of museum audience from a passive group of listeners to active participants with multimodal resources to interact with the guide (Best, 2012).

Social interaction becomes all the more relevant in the context of blind and partially sighted visitors who must typically rely on external aid (e.g., another person, a guide dog, or an audio guide) in order to move in unfamiliar places and to assimilate visual information. In consequence, blind and partially sighted people experience the museum with and via other people and agents. Museums usually offer two types of guidance and AD for visitors with visual impairment: prerecorded audio-guides, including audio-descriptions (Soler, 2015, p. 17) or guided (AD) tours, which in many museums are non-scripted and provided live (Perego, 2019, p. 335). These may be supplemented with multisensory experience, for example via sound effects and touchable objects, which is another way of providing access to museums (Kreplak & Mondémé, 2014). In this article, we describe two different approaches to realizing a guided tour with live AD.

2. Materials and methods: two approaches to AD in art museums

In a microethnographic and conversation-analytical study of the multimodal constitution of AD (the MUTABLE project³), the first author of this paper was allowed to follow and video-record two guided tours (‘cases’) at Finnish art museums with live AD.⁴ The data were gathered by a convenience sample in 2017 as the project was recruiting research settings from the author’s network of audio-describers and users. The recorded tours were organized for a group of blind and partially sighted art enthusiasts and their sighted companions. The museum, the exhibition, and the guide were different in each case but many of the visitors were the same in the two tours (approximately 14 people each time). The cases were named SAAD1 and SAAD2 (Simultaneous Art Audio Description).

SAAD1 involves a guided tour of an exhibition in a private art museum in Finland. The tour lasted an hour and included the AD of three artworks. The artworks were paintings by an artist of the Finnish classical era, dating from the beginning of the twentieth century. The tour was guided by an inhouse museum educator who had had training in AD (a three-day course organized by a museum institute), and this tour was the first time she gave AD. The guide performed the AD by reading a text/script out loud to a headset microphone, which she carried in her hand during the tour. The visitors sat on foldable chairs in front of the artworks, and they heard the AD via a headset on one ear.

SAAD2 includes a guided tour to an exhibition in a large, publicly owned art museum, also in Finland. The tour lasted 1 h 15 min and included the AD of eight artworks. The artworks are paintings of the same era as in SAAD1 but by a different artist. The tour was led by a freelance museum guide whose background and experience in AD is not known to us.⁵ The guide spoke her descriptions freely, intertwined with her guidance talk. She did not use technical equipment for voice transmission, or any paper or other memory aids. Visitors were mostly standing during the tour, although some sat on benches where these were available.

Consent to collect data from the visits was given by the guides, the museums, and the visitor groups prior to or at the beginning of the tour. The first author of this article was present on the tours, recording them with a video camera and conducting the interviews afterwards, but remained in the background to avoid influencing the situation. Immediately after each tour, on the museum premises, volunteer participants (both visually disabled and sighted) were interviewed.

Our study applies microethnography (LeBaron, 2005; Streeck & Mehus, 2005; see also Risku et al. 2022) to empirically analyze occurrences of ‘big’ social and cultural issues (here: experience) in ‘small’ communicative behaviors and action (here: interaction among the tour participants). We use video recordings as primary data and study the visible and audible behaviors of social actors embedded in a social and material environment. Video enables repeated observation and verification of the analysis. To reconstruct the interaction in detail, we used multimodal conversation analysis (e.g., Best 2012; Hirvonen & Tiittula, 2018) to reveal the practices which interactants use in constructing their actions and in orienting to each other and the artworks. Transcribing the multimodal data (speech, bodily behavior, etc.) is constitutive of the analysis because it enables the reconstruction of social interaction (see Mondada, 2018).

As in microethnography, we supplemented the interaction analysis with retrospective participant interviews to gain an understanding of the experiencing subjects’ perspectives; that is, how the participants themselves describe their experience (Perttula, 2005, pp. 136–143). In SAAD1, this was a group interview, and in SAAD2, they were carried out individually. The interviews first asked questions about participants’ backgrounds (age range, grade of blindness/visual disability, background in fine arts, and education/profession). Then the participants were asked to share their thoughts on the AD they had just experienced and, if applicable, to compare this experience with previous experience with AD. The interview in SAAD1 had four participants, three of them blind and one sighted. The interview in SAAD2 had four participants, of whom two were blind, one partially sighted and one sighted. One visitor participated in both interviews. All interviewees were adults (age range 40–80) and none of the blind persons reported themselves as congenitally blind.

3. Results

In this section, we first summarize how the two tours unfolded interactively and reconstruct some typical sequences of interaction in scripted and spontaneous AD (3.1.). Then, we provide an overview of the interactional and discursive characteristics of the two formats (3.2.). Finally, in Section 3.3, we use the interview data to shed light on the visitors’ experiences of the AD formats.

3.1. Interactive organization of scripted and spontaneous AD

Table 1 presents the general unfolding of activities in the tours. Both cases entailed phases which were either focused on ‘guide-talk’ (background information and narrative accounts of the art and artists) or on AD (the description of artworks and the environment), as might be assumed to happen in guided tours in which AD is provided. However, a significant difference occurred with the delivery of AD (phases 2–4 in Table 1): whereas in SAAD1 the AD and the guide-talk were separate phases, in SAAD2 these activities intertwined so that AD was part of guide-talk.

Table 1. Comparison of the organization in the two visits.

	SAAD1	SAAD2
1	The guide leads the visitors into the exhibition and waits for them to organize themselves in front of the artwork.
2	The guide performs the AD.	The guide presents background information and AD. A discussion between the guide and the visitors, with shared conceptualizations and specifying questions, is intertwined with the guide’s presentation of background and AD.
3	The guide asks the visitors a question.
4	A discussion starts in which the visitors describe their conceptions of the artwork and ask specifying questions, and the guide gives background information about the artwork.
5	The guide invites the visitors to move to another location.
6	The group moves to another artwork.

In SAAD1, each object viewing began with AD. SAAD1 followed the typical, guideline-based order (Soler 2015) in which technical details of the artwork precede a detailed account of the visible ‘facts’ (De Coster & Mühleis 2007) and, finally, (personal) impressions were voiced. The guide, who had positioned herself to the left of the painting from the visitors’ perspective, performed the AD by reading the descriptions from the script on paper in her hand. The visitors were seated on foldable chairs facing the painting and the guide. While performing AD, the guide’s eyes were fixed on the paper. She stood mostly in one place, occasionally turning her torso slightly more either toward the painting or the visitors. These shifts in posture signal a shared space between her, the object and the observers/visitors (Schmitt & Deppermann 2010, pp. 201–211); however, her visual attention remained with the AD script. The most distinctive bodily action is the gesturing with her free hand: she moved it along the paper while reading to set pace, emphasis, and rhythm to her reading. The visitors sat silently for most of the time but occasionally gave vocal and bodily feedback (e.g., by interjecting mm, by nodding, and by shifting body posture or face/gaze toward the artwork or the guide). Transcript 1 illustrates the scripted style of AD delivered in SAAD1 in which descriptions are read out loud (G1 = the guide in SAAD1; for transcription symbols, see the appendix).

Transcript 1: The guide performing AD by reading out loud (SAAD1_S1280004 at 00:03:57-00:04:19)

01 G1: (…) teoksen päällä on lasi. //

     the work has glass on it

02    ↑kuvassa oleva nainen on / kaksikymmentä viiva

     the woman in the picture is   twenty to

03    kolmekymmentävuotias; //

     thirty years old

04    hän istuu kuvan oikeassa reunassa aidalla / kädet helmassa,

     she is sitting on the right side of the picture on a fence hands on her lap

     *moves slowly towards the painting* 

05  hieman kumarassa asennossa, /

   in a slightly hunched posture

   *waves left hand pointily at the paper-->

06  niin että suorana olevat jalat / lepäävät maassa /

   so that her legs which are straight  rest on the ground

   -->      *

07  kuvan vasemmassa alareunassa näkyvän lankun päällä /

   on a plank that is visible at the left lower edge of the picture

08  rinnakkain.

   side by side

In this excerpt, the guide describes the appearance of the woman figure in the painting with features like her age (line 3), position (lines 4, 7–8), action (line 4) and posture (lines 4–8). This sequence illustrates the syntactic complexity and semantic richness of the scripted AD: the description of the figure’s posture and position expands over several complementary constructions (4–8), of which one is a subordinate clause (6–8). The speech consists of longer and shorter utterances delineated with short pauses. Longer pauses (1, 3) separate distinct thematic entities, and intonation (a sentence-initial high pitch, Hirvonen & Wiklund, 2021) marks transitions into new foci (Holšánová, 2008, pp. 9, 40); here, from a general introduction of the motif by a summarizing focus (the woman figure) to an elaboration of the motif by substantive and localizing foci describing position and posture of the figure.

In SAAD2, while the spatial organization of the group is similar to SAAD1 (the guide standing next to the painting, on the visitors’ left), the difference lies in the presentation of AD. SAAD2 did not follow such a strict order as SAAD1 but the guide performed the AD in a spontaneous style and in interaction with visitors; in addition, the AD intertwined with guide-talk. The visitors were spread more widely in front of the artwork, standing and sitting. One visitor had her companion draw on her back while the guide was audio-describing. The discourses of guide-talk and AD tended to blur: descriptive passages that can be categorized as AD were often given in-between guide-talk, or guide-talk was added to the passages of AD, as Transcript 2 illustrates (G2 = the guide in SAAD2).

Transcript 2: The guide merges AD with guide-talk (SAAD2_S136003 at 00:05:05–00:08:50).

[00:05:05]

01 G2 eli tuota: NN:n ensimmäinen vaimo¿ /

    okay so [the artist]’s first wife

    *gaze wanders around the audience-->

02   NN¿ / ähm / oli kotoisin (…)

    [the wife] uhm was from

   *turns to look at the painting*

03   ((for appr. 2 min 30 sec, the guide talks about the wife’s

04   background and about the couple’s relationship, glancing back

05   and forth between the painting and the audience))

 [00:07:40]

50 G2 öö tää ensimmäinen, /

    uhm this first one

    *looks at the painting*

51   tämä on NN nimeltään ja tää on näistä vanhin tästä

   this is called [name of the painting] and this is the oldest of these

   *glances between the title frame and the painting-->

52   niin sanotusta NN-sarjasta¿

    of this so-called ‘[name] series’

    -->

53   tää on NN kokoelmasta öljy kankaalle. /

    this is from the [name] collection oil on canvas

   -->

54   ja tää on vuodelta tuhat yheksän sataa yheksän,

    and this is from nineteen oh nine

   -->   **turns to look at the audience-->

55   eli siltä vuodelta ku he on avioitunu ja ja lähteny pariisiin¿

    that is from the year when they got married and and left to Paris

    -->      *

56  ((the guide turns toward the painting and pauses,

57  then turns to the audience and talks more about the artist’s

58  background and his relationship with his wife, gaze wanders in the

59  room))

[00:08:35]

65 G2 ähm tää maalaus on vähä suurempi ku- kuin tuota se edellinen

    uhm this painting is slightly larger tha- than well that previous one

    *glancing aroung the painting and the exhibition label-->

66  mutta ei välttämättä kovinkaan paljoa ehkä, //

   but not necessarily very much perhaps

   -->

67  seiskyt //

   seventy

   -->

68  oiskohan kuu- viiskyt kertaa seiskyt tai kuuskyt kertaa seiskyt;

   could be six- fifty times seventy or sixty times seventy

   -->      *

69  näis ei oo mittoja meillä valitettavasti näissä merki-

   these don’t have measurements here unfortunately these mark-

   *glancing at the audience-->   *

70  näissä taulu- taulujutuissa esillä, //

   visible in these label- label things

   *turns to look at the painting**looks at the painting-->

71  ja täs on nyt sitten NN //

   and so this one has [the wife]

   -->  *glances at the exhibition label*

72  tai NN mallina /

   or [nickname of the wife] as model

   *glances at the audience*

73  katsoo suoraan ahm maalauskohteeseensa ja ja tota

   looks directly at uhm to her object of painting and and well

   *looks at the painting-->   *

74  tämä NN niin ku nää muutkin (…)

   this [name] like these other ones

   *turns to look at the audience-->

75  ((the guide continues explaining a fun fact about the painting))

Transcript 2 shows two passages of AD (lines 50–54 and 65–69) intertwining with guide-talk. The guidance begins with background facts of the artwork (lines 1–7), which altogether lasts for almost three minutes. Then, the guide shifts the discourse to AD both verbally and bodily by topicalizing the painting with a deictical expression and its name (lines 50–51) and by giving further basic information of it (lines 52–54). However, the AD is not fully accomplished as the guide shifts to narrating about the artist’s background (lines 55–59). Overall, this type of intertwining discourse was the basic strategy for the guide in SAAD2, as illustrated by the continuation of Transcript 2 (line 65 onwards). In SAAD1, the two discourses were clearly separated although some merging happened as the guide was prompted by the visitors to talk more about a detail in the art.

The spontaneous style of SAAD2 is further demonstrated by Transcript 3. This passage occurred while the group was viewing the first artwork of the tour.

Transcript 3: The guide renders AD as spontaneous speech (SAAD2_S1360002 at 00:18:34–00:18:56)

34 G2: (…) mutta hyvin hyvin keskellä;

    but very very much in the middle

    *glances at visitor and assistant on her right*

35   ähm / auringon molemmilla puolilla, /

    uhm on both sides of the sun

   *turns to artwork*

   *raises left hand and moves it vertically by the painting*

36   on tämmösiä, /

    are these kinds of

    *touches her hair with left hand*

37   ähm, /

    uhm

   *makes circular movements with her hands in front of chest-->

38   on puita¿ / ää, / tämmösiä / ää /

    are trees  uhm these kinds of uhm 

39   kapeita, / lehtipuita; /

    narrow   broadleaf trees

    *sways upper body and left hand towards the wall*

40   jotka on maalattu tämmösillä vaaleen violetin ja

    that are painted with these kinds of light violet and

    *moves left hand vertically alongside the painting*

                  *turns towards audience*

41  liilan sävyillä / vin- tämmösiä:  /  ähm /

   lilac shades win- these kinds of  uhm

   *touches her chest with left hand*

                *turns towards artwork*

                       *fast vertical

   movements with left hand along left side of painting-->

42  [villin hhh. villein vedoin hih]

   with wild   with wild strokes

   and towards audience----------*

43   [((visitors chuckle    ))]

In this excerpt, the guide shifts from describing the ‘sun’ element in the painting to its surroundings. The guide’s talk is filled with markers of spontaneous speech. She uses shortened demonstrative proadjectives (tämmösiä ‘these kinds of,’ 36, 48, 41; tämmösillä ‘with these kinds of,’ 40) as well as multiple hesitating or contemplating interjections (ähm 35, 37, 41; ää 38). The Finnish demonstrative proadjectives serve not only to ‘fill’ speech, but they refer deictically to an object close to the speaker and verbalize the speaker’s thinking process and signal to co-participants that the process is ongoing (Larjavaara, 1990, pp. 96, 99; Larjavaara, 2007, pp. 327, 337).

Multimodal interaction is rich. The guide switches her orientation between the audience (34, 41, 42–43) and the artwork (35–40, 41) with her body and gaze. When describing the artwork, she looks at the artwork and uses her hands to depict features of the painting, such as producing vertical movements while describing (on both sides [of the sun], 35) and circular movements while looking for a specification (trees, 37). As other research has found, hand gestures and gazing away from co-participants are typical of searching for words (Goodwin & Goodwin, 1986, pp. 56–57; Hayashi, 2003, pp. 112–113, 133). This excerpt ends in a joint laughter initiated by chuckling from the audience (43), which the guide responds to by chuckling (42). Joint laughter is found to be a collectivity-building device in interaction (Glenn, 2003).

We spotted also other descriptive sequences within the guide-talk. Descriptions can concern not only the objects in focus in the museum but also the environment in which the visitors move and in which the objects are installed. Describing as activity can take place during other ongoing discourse, not only in the form of AD (as a specialized descriptive discourse), but also during the guide-talk, intertwined with narratives about the artist’s personality, history, and so on (Transcript 4; VV = a visitor).

Transcript 4: The guide describing art during guide-talk (SAAD1_S1280007 at 00:53:00–01:27:00)

01 G1: (…) että tämä hahmo / tämä rangaistusta kärsivä mies /

       that this figure  this man who is suffering a punishment

     *pointing at the painting with open hand*

     *looking at the painting-->   *

02    hän on syntinsä jo sovittanut et hän on hyvin tyyni ja rauhallinen;

     he has already atoned for his sins so he’s very calm and peaceful

                       *pointing at herself   *

     *looking at the audience-->

03     //

     -->

04      mutta tässä se pahis on olluki se herjaaja / se joka ivaa ihmistä

     but here the bad guy has been the slanderer that which derides a human being

     -->

05      joka on katunut ja syntinsä sovittanut.

     who has rued and atoned for his sins

     -->

06 VV:   mhm,

07      //

     -->*

08      *nodding from the audience*

09 G1: tän miehen kasvoja oli vähän vaikea kuvailla mutta hänellä on tosiaan

     it was a bit difficult to describe this man’s face but he has indeed

     *looks at the painting-->

10      parta ja ruskea tukka, / hänen silmänsä on luotu alas ja hänellä on

     a beard and brown hair  his eyes are laid down and he has

     -->

11      aika rauhallinen ilme / kapeat posket, / tästä teoksesta on myös

     a rather peaceful expression on his face thin cheeks this work has also raised

     -->      *

12      arveltu että NN olis yrittänyt kuvailla NN

     speculations that [the artist] would have attempted to describe [a Finnish national author]

     *looks at the audience-->

13      (…)

     -->

As noted above, in both AD cases, interaction between the guide and the visitors was an integral part of the tour. This interaction allows the guided tours to become spaces for joint meaning-making. As AD is provided in face-to-face interaction, it allows the describer (guide) and the users (visitors) to engage in a discussion about art and the perceptions related to it; as aspects of the artworks are being verbalized, they become available to the entire group and, if deemed relevant, the perceptions can be re-constructed or modified jointly. In the data studied, we observed that both the guide and the visitors initiated public meaning-making and invited others to join in by sharing their views or opinions about the artwork or by raising issues for discussion. Given the analytical perspective on AD here, we focus on how the description of visuals can be publicly complemented. In what follows, we first present data excerpts that illustrate the interaction (Transcripts 5–6; VV = a visitor) and then give a comprehensive overview of meaning-making practices in the data (Table 2).

Table 2. Practices for joint meaning-making (translations from the Finnish data by MH).

	practices	data example		type of meaning
#1	request & proposal  + confirmation	- about those hands, I don’t now did you say in the AD that they’re like on the belly, like protecting like this …  (man_1) - yes they rest on the lap below the belly (guide_1)	SAAD1/S1280004	visuals facts
#2	request  + elaboration	- what kind of frame does it have (woman_3) - it’s got a huge frame (…) (guide_2)	SAAD2/S1360002	visual facts
#3	request  + interpretation  + confirmations	- what kind of image did you get from this painting (guide_1) - daddy’s on his last leg …  (man_1) - mm / yep (various visitors) - yes (guide_1) - yes maybe they … to the municipal doctor (woman_4) - yeah (guide_1) - it must be [the artist] there with a thumb in his mouth (man_1) - a little bit distressing feeling (man_2) - yeah (woman_4)	SAAD1/S1280006	mental image interpretation visual facts emotional effect
#4	statement  + confirmation	- it must be a pretty beautiful work (man_1) - it is (woman_4)	SAAD1/S1280004	aesthetic effect
#5	statement  + confirmation	- that’s exactly [the artist] style (man_2) - yeah (guide_2)	SAAD2/S1360002	genre
#6	Proposal  + confirmation  + elaboration	- (…) well maybe in part but that this is rather also a kind of like … a study of a human being (…) (guide_2) - of a woman, study of a woman (man_3) - uhm like a portrait … yes an illustration of a woman indeed, and perhaps even of a this kind of sensual and erotic woman (…) (guide_2)	SAAD2/S1360003	genre visual facts
#7	+ proposal  + confirmation	- (…) and well the gaze is somehow like … begging, I don’t really know what or how the gaze is but it’s a pretty sorrowful gaze that comes across and like uhm …  (guide_2) - would it be like apologetic (woman_4) - could be that too (guide_2) - or is it ceding (man_1) - a bit like that, ceding, too, that’s right (guide_2) - that everything’s alright, that she is/has …  (woman_4) - you know what happened, [the artist] took off with (…) (guide_2)	SAAD2/S1360004	visual facts interpretation

Transcript 5: Shift to interaction with visitors (SAAD1_S1280004 at 00:08:02-00:08:24)

20 G1:  nyt mä laitan tän:, / mun, /

     so now I turn this     my

21     mikin pois päältä? / [niin] että me voidaan

     mic off         so   that we can

22 VV:             [mhm?]

23 G1:  keskustella siitä; /

     discuss it

24    mä ↑toivoisin että jos / joku¿ /

     I would hope that if      something

25    tässä teoksessa / [on sellanen yksityiskohta]

     in this work      is that kind of a detail

26 VV:            [((mumbles something)) ]

  G1:                    *glances at VV-->

27 G1:  josta te haluaisitte tietää vähä enemmän ni, /

     about which you would like to know a bit more so

     -->    *

28    keskustellaan minkälainen mielikuva teille tuli

     let’s discuss what kind of a mental image you got

29    tai mä voin kertoa enemmän jostain.

     or I can tell more about something

The guide announces that she will turn off her microphone by verbalizing—audio describing!—her own concurrent bodily action (so now I turn this mic off, 20–22). Someone from the audience recognizes this (feedback interjection in line 22). The guide proposes to discuss it (‘it’ referring to either the painting or the AD) and renews her proposal (I would hope that … which you would like to know … let’s discuss … I can talk more, 24–29). Her speech displays careful planning as she takes longer pauses between utterances (23, 24, 27) and reformulates (what kind of mental image you got or I can tell you more …  28–29). The guide becomes bodily attentive to the visitors at this point and responds to the feedback by a visitor by glancing at this person (26).

A notable change in the guide’s speech is that she shifts from the written format to the spoken language format. In spoken language, pauses do not always follow syntactic structures but occur within sentences, and some words come in shortened forms (such as the first-person singular pronouns mun ‘my,’ mä ‘I,’ instead of the formally correct forms minun, minä (‘my,’ ‘I,’) in spoken Finnish) (Tiittula & Voutilainen, 2016, p. 49; VISK § 595).

In SAAD2, the guide regularly makes eye contact with visitors, and the visitors’ verbal and bodily feedback affect the guide’s performance, as demonstrated by the following data excerpt (Transcript 6).

Transcript 6: The guide and visitors in interaction (SAAD2_S1360002 at 00:19:11-00:19:57)

41 G2: (…) hyvin herkkiä ja erilaisia vihreän sävyjä, /

       very delicate and different shades of green

42    mun (sukulainen) rakastaa kevättä just sen takia että

     my (relative) loves spring exactly because of that

               *turns to audience and touches chest

43    niitä vihreän sävyjä [on niin valtavan] paljon /

     there is such an enormous number of those shades of green

     with left hand*

          *gaze moves in the audience*

44 VV:             [mmhm¿       ]

                          *nods*

45 G2:  et on kaikki keltasen ja vihreän sävyt ja [ne on] kaikki

     that there are all the yellow and the green shades    and they’re all

     *waves with hands*

46 VV:                         [hmm, ]

47 G2:    siellä yhtä aikaa /

     there at the same time

                *turns back to the artwork*

48    ja vaikka niitä vihreän sävyjä on vaan muutamia ja

     and even though there are only a few of those green shades and

49    siellä [täällä] /

     here and there

    *movements with left hand-->

50 VV:     [mm  ]

51 G2: reunustamassa näitä violettia oksistoja ja /

    lining these violet branches and

    -->               *turns towards audience*

52   ja puunrunkoja niin ni mä nään siellä niitä

    *gazes at VV-->   *

     and the tree trunks I can still see these

53   vihreitä; /

     green colours there

    *turns back to the artwork*

 VV:     *nods*

54 G2:  ähm / kymmenittäin erilaisia / ne ei oo sekottunu

     uhm  dozens of different kinds of   they are not blended

 G2:                     *small movements with

55    toisiinsa / vaan ne on / ne on aika tämmösiä rajuja ja

together but they are they are quite these kinds of rough and

    left hand**vertical movements with left hand*

56   välillä /

   sometimes

    *turns towards audience, left hand high*

57    paksuja semmosia [laastimaisiakin] vetoja /

     thick and that kind of even plaster kind of brush strokes

58 VV:            [mm,  ]

 G2:                        *gazes at VV, turns back to

 VV:                        *nods slowly repeatedly

59 G2: mitä ne on /

     that they are

     artwork*

 VV: ---------------*

60   ja nää puut tekee tähän (…)

   and these trees make here

A blind visitor, who is having the description/the painting drawn on her back, responds verbally and bodily (nodding and with the feedback interjection mmhm, line 44) to the guide’s talk and body orientation (43). While describing colors and shades (45–57), the guide looks back and forth between the artwork and the visitors (52). This visitor becomes active (58) and her verbal and bodily action establish interaction with the guide: the visitor utters minimal feedback (mm), to which the guide responds by looking at her, and complements her verbal action with a lengthy nodding (58–59). Then, the guide turns back to face the artwork and changes the topic from the colors to describing the tree element (58–60). Similar interaction between the guide and this visitor has occurred previously during the tour. The guide glanced at the visitor while she was receiving tactile description on her back. Although we cannot discern from these data how this interaction affected the guide’s planning of her AD, it can be observed that she orients to it repeatedly, as if synchronizing her description and talk with the tactile description.

In SAAD1, the guide inquired explicitly after each AD phase about the mental image that the visitors might have gained. In SAAD2, the guide sometimes invited the visitors to make questions, but it was mostly (some of) the visitors who initiated the joint meaning-making. In general, people can resort to versatile interactive practices of joint meaning-making and the co-construction of meanings (Deppermann 2020): in our data, we found requests, statements and proposals as initiating practices, and confirmations, elaborations, and interpretations as co-constructing practices (see Table 2). The co-participants typically affiliated with the interpretations and assessments made by someone, or at least non-affiliating perceptions were not voiced.

The analysis of the practices of interactive, joint meaning-making shows how the group engages in giving meaning to art at all levels: from minute visual details to the emotional effects of art as well as to their (art-)historical contextualization (see the column on the furthest right), from inspecting artworks (examples 1, 2, 7) to interpreting, assessing and experiencing them (examples 3, 4; 7) and analyzing them (examples 5, 6). When provided with AD and descriptions, also visually disabled people can be active participants in the shared meaning-making, not only asking questions about visual information (examples 1, 2), but also assessing the art (examples 3, 4, 5) and even assisting the guide in describing it (examples 1, 3, 6, 7).

3.2. Participant experiences of the two AD formats

In addition to analyzing the video data, we explored how the two AD formats were received by interviewing participants. Nearly all interviewees had personal interest and experience in fine arts and AD. One blind visitor had engaged with fine arts for decades and goes to art museums and exhibitions. Another blind visitor had always enjoyed drawing and coloring and has a particular interest in colors. The third blind visitor had work experience in the field of visual culture. The fourth blind visitor was an art professional herself. The partially sighted visitor had no background in fine arts, but in music. One of the two sighted interviewees answered that she is mainly an ‘art consumer’ who tends to visit art exhibitions. The other sighted interviewee had taken part into some painting retreats and visited art exhibitions.

Several visitors talked about having visual memories and how this helps them to relate to fine arts today. The partially sighted interviewee explained the relevance of AD to her experience by describing that, without AD (‘explanation’), she would not be able to discern what is visible:

But audio description … it is very important first of all cause even from a totally clear painting I wouldn’t … without that explanation, that there is, I wouldn’t perceive straight away … if I looked at it, I need the explanation that what there is, what is seen

[a partially sighted visitor/SAAD2]

The sighted interviewee in SAAD1 had enjoyed the AD also from a pedagogical perspective:

I think I learnt that I am really capable of *laughs* that was an example of, when you’re with a visually impaired person in an exhibition, then, what to say.

[a sighted visitor/SAAD1]

How did the participants experience the two AD formats? All four interviewees responded positively to SAAD1. They praised its informative aspects and the accuracy of description, but also welcomed the ergonomics of the tour (the use of headsets and portable chairs). The blind visitors appreciated the description of composition in the artworks and verbalized their experience as if they were able to ‘sense the atmosphere of the painting’ or ‘see the picture’ and have a sense of ‘looking at the pictures.’

I must say this is … this was by far one of the best I’ve participated in. I mean like, the most detailed and … so that I felt like I was constantly looking at the pictures

[a blind visitor/SAAD1]

In both cases, several interviewees considered the ‘facts’ in the guide-talk (information and background to the artist and artworks, see Neves, 2015, p. 71) to be valuable content. The excerpt below from one interviewee in SAAD1 summarizes many of the answers:

To me what matters is precisely that story behind it, and the atmosphere

[a blind visitor/SAAD1]

Some visitors are keen on knowing details about visual art:

well I’m always interested in those … the paintings’ techniques, compositions, the motif, and such, horizon … like how it’s constructed the whole thing

[a blind visitor/SAAD1]

In SAAD2, all interviewees regarded the strategy of intertwining AD with guide-talk as positive or neutral. Two of the blind participants emphasized that information about the artist’s background keeps them interested:

I at least felt that I gained a lot … I think it was nice and it kept us interested because there was simultaneously this information about the artist as well as of the paintings

[a blind visitor/SAAD2]

This had- I’ve always been- I’m always fascinated in the guided tours in that, in the background and all

[a partially sighted visitor/SAAD2]

One blind participant in SAAD2 was critical of the style of presenting AD, rather than of its interactional organization or spontaneity per se. However, this participant, too, evaluated the AD as ‘fine’ because the collective experience accepted it (‘the audience was quiet and happy’):

It [the intertwining structure of AD and guidance] wasn’t really disturbing in this exhibition that much … can be that in some other it would have but here … it was suitable here even though it was a little bit subjective even if we go to, to minute details but well … all in all fine I think because the audience was quiet and happy, or relatively quiet

[a blind visitor/SAAD2]

This visitor had participated in both of the guided tours and was asked to compare his experiences. The visitor preferred the detailed, informative style of SAAD1 over the loose and subjective style of SAAD2. Specifically, this was because SAAD2 lacked details about the artwork itself and the inconsistent use of spatial directions (right, left) at times resulted in an unclear spatial positioning of the visitors, the guide, and the artwork. However, another blind participant referred to the SAAD2 guide’s skill in mediating the artist’s personality and considered this as a decisive factor for the good experience:

well at least the guide was so good and … like she was enthusiastic about the topic herself that it came through … and then it came so powerfully the artist’s personality that like … one could go inside into his world

[a blind visitor/SAAD2]

4. Discussion: toward an accessible museum experience

In this paper, we have described two formats of live AD in guided museum tours. By studying naturally occurring video data, we demonstrated how the AD formats are structured as interactional situations between the guide and the visitors and how they enable or disable audience participation. We also analyzed retrospective participant interviews in order to understand how the two AD formats were experienced.

Our analysis showed, first of all, that verbal descriptions of the museum objects (artworks) appear in various discourse contexts, as separate phases but also intertwined with or included in the guide-talk. This finding reflects observations from previous research on the interaction between blind and sighted people: blind people use versatile sources to receive visual information, given directly or indirectly by sighted co-participants (Everts 2012: 228–229). Second, the analysis found versatile interactive practices of joint meaning-making and demonstrated the active participation of (blind and partially sighted) visitors in this. Not only does the guide engage in bodily activities that enable her to orient to and interact with both the artwork and the visitors, but also visitors initiate meaning-making and participate in describing, assessing, and interpreting artworks. The format constrains these activities to the extent that reading AD out loud from a script does not allow for rich interaction simultaneously because the speaker is attentive to the text. Spontaneous talk enables co-participation because pauses and hesitation markers are created while speech is being planned. Even though hesitations can connote negatively about the describer’s insecurity or ignorance, they in fact lay the ground for the ambiguity of perception (see De Coster & Mühleis, 2007) and prepare for co-participation, both of which are advantages to AD: ambiguity creates more opportunities for individual imagination and feeds the joint meaning-making. The interview analysis shows, perhaps more unsurprisingly, that using live AD enhances multiple aspects of (art) museum visits, from learning something individually to creating a collective experience.

We conclude our article by discussing the dimensions of accessible art museum experiences in terms of which aspects become realizable in scripted and in spontaneous AD. The guided tours we analyzed seem to benefit each dimension of experience defined by Pekarik and colleagues (1999): object experiences, cognitive experiences, introspective experiences, and social experiences (Pekarik et al., 1999, pp. 152–156). Both AD formats created object experiences for/in visitors, which is proven by their verbal and bodily feedback and specifically by the detailed accounts of visual perceptions in the processes of joint meaning-making. In the interviews, visitors commented on their ability to ‘see’ or ‘be looking at’ the artworks thanks to AD; in the group interaction, requests by blind participants revealed that they had formed a mental image of the artwork and now needed to complete the picture with some information. Similarly, the visitors’ cognitive experiences, or gaining information or knowledge of the artworks, were reflected in both data. In particular, the scripted AD was praised by the interviewees for its accuracy and informativeness. Cognitive experiences even seem to work in favor of introspective experiences: as some visitors explained, they remembered aspects of the visual world (like the artist’s style) or even could see better (as a partially sighted person) thanks to AD. The guide’s behavior enabled visitors to gain introspective experiences: her enthusiastic presentation style enabled them to feel a sense of belonging. The joint meaning-making within the groups certainly had a similar effect, and the guided tours were social experiences overall. Finally, the joint meaning-making of art, during which visitors could ask questions about the artworks or elaborate on the AD, generates a kind of collective experience – a joint construction – which in future is likely to transform to an introspective, individual experience: a memory, a feeling, or the like.

As Hutchinson & Eardley (2018) assumed, experiencing art and museums with the help of interactive AD involves much more than merely informing the recipients on the content and style of artworks. The social and collective dimensions of experience seem particularly relevant, which is in line with the finding that all museum experiences are highly interactive, whether guided or not (vom Lehn & Heath, 2016). In future, the study of AD experiences should combine different analytical approaches to empirically demonstrate how these experiences are formed. For example, the foci of discursive functions (Holšánová, 2008) seem to correspond to the categories or dimensions of experience (Pekarik et al. 1999). These foci should be observed in multimodal data, which makes it possible to analyze live AD in terms of multisensorial experiential categories.

In this study, we applied a microethnographic methodology to collect and analyze data from two cases of live AD. Although our study is exploratory and the data limited, it found significant phenomena that have not been described in AD research previously but clearly invite more research in future – with more data and purposive sampling. Even if ethnographic research does not aim at generalizing its findings beyond the context of the study, it is possible to contemplate how the ‘site-specific findings have relevance beyond the site and [that] particular patterns of behavior resonate with larger social orders’ (LeBaron, 2005, p. 278). We extended the relevance of our findings ‘beyond the site’ by identifying and describing patterns of behavior that are generalizable to other contexts of social interaction (e.g., the features of spoken language, the physical boundaries of scripted speech, and the use of multimodal resources). Our findings are applicable to the practice of AD, but rather than prescribing AD norms, we wish to enable practitioners with analytical perspectives to explore different interactive practices and to evaluate themselves which of the two formats works better in what context and for what aim.

Finally, our analysis of spontaneous AD is but the first step in a hopefully emerging line of research into other situations and contexts in which AD is provided as situated assistance or paraprofessional action (see Koskela, Koskinen & Pilke, 2017) rather than a professional service, and in which AD users and producers engage in joint meaning-making. Such forms of ‘everyday AD’ remain scholarly largely untapped (see, however, Márquez Linares, 2007; Quereda Herrera, 2007), whereas their societal prevalence and scientific relevance are important.

Acknowledgments

The authors are indebted to the research participants involved – with you we now understand much more about audio description! We also thank the research assistants who helped preparing the data and a few friends and colleagues who we consulted about art vocabulary or who participated in data sessions. Finally, we express our gratitude to the two anonymous reviewers whose feedback helped us to concentrate on the most relevant aspects of these rich data and analytical paths.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Academy of Finland.

Notes on contributors

Maija Hirvonen

Maija Hirvonen is Associate Professor of German language, culture and translation at Tampere University. Her main research interests are human-made audio description and machine-made video description, accessibility, intermodal translation, multimodality, and interaction and conversation (especially between blind and sighted people). 

Betta Saari

Betta Saari holds an M.A. in German language and is pursuing a PhD in the same field at Tampere University. Saari’s dissertation project studies the accessibility of visual art and art experience from the viewpoint of blind and partially sighted people. She currently works as e-commerce customer service at Luhta Sportswear.

Notes

1 This work was supported by the Academy of Finland under Grant # 295104. Due to the nature of this research, the data cannot be shared publicly, so supporting data is not available. Part of the analyzed corpus (SAAD1) is archived in The Language Bank of Finland, from which it can be requested for research use under specific conditions (data identifier: http://urn.fi/urn:nbn:fi:lb-2019062002).

2 We use ‘scripted AD’ for AD that has been written first and is reproduced in a spoken format, and ‘spontaneous AD’ for the format that seems unscripted (a script is not used) and that is characterized by features of spoken language.

3 https://projects.tuni.fi/mutable/

4 After submitting this paper, the MUTABLE corpus has grown by one case.

5 After the data were gathered, the researcher lost contact with her, so this information can no longer be acquired.