Abstract

This paper explores the influence of the media turn on the dynamics of interpreter-mediated dialogic interactions in the context of the Giffoni Film Festival, a unique and well-known international film festival held every summer in southern Italy. Its theoretical framework brings together the two seminal concepts of ‘footing’ and ‘positioning’, by looking at utterance-level alignments within sequence-level discursive and spatial positions. The discussion pursues a twofold aim: to devise a descriptive typology of interactionally constructed interpreter positions through a qualitative analysis of representative sequences of YouTube live streamed encounters with world-famous cinema stars; and to determine whether and how the presence of live streaming impacts participants’ behaviours. To this end, the types identified in the analyses are compared with non-live streamed interpreted Q&A sessions with members of the film crews collected in the same setting. Since film festival interpreting is here subsumed under the broader category of media interpreting, the streamed data are also discussed against the backdrop of Francesco Straniero Sergio’s pioneering investigation of TV interpreting practices. Live streaming was found to have significant repercussions on physical and interactional positioning in that it limits the agency of both the festival host and the interpreter by imposing a rigid routine on the event; interpreters appear to adjust to the requirements of the new media context not so much out of lack of interactional power – a power which they apparently feel freer to exert in non-streamed events – but because they prioritise the success of the show over their own visibility; differently from TV entertainment, not only are the guests more interactionally dominant than the hosts, but the needs of the remote audience of streaming users are less catered for than those of the flesh-and-blood jurors sitting in the cinema theatre, despite the structural accommodation to the new medium.

Keywords: film festival interpreting, footing, media interpreting, new media, positioning, live streaming, TV interpreting

1. On the shoulders of giants: A preview^[*]

The earliest studies on media interpreting (hereafter MI) date back to the 1980s (Dal Fovo 2020). Since then, the Italian scholarly community has been one of the most productive in the field thanks to the pioneering studies of Francesco Straniero Sergio (1999, 2003, 2007, 2012), and his collection of copious authentic video materials which led to the creation of the unique corpus of TV interpreting, CorIT, hosted at the University of Trieste (Falbo 2012). In particular, Straniero Sergio’s volume on talk show interpreting (2007) opened up multiple avenues of investigation by applying a wide array of theoretical notions to interpreting practice. One such line of enquiry is based on Erving Goffman’s seminal notion of footing. To exemplify interpreters’ orientation to the other participants in interaction in terms of cooperation vs. competition, and involvement vs. distancing, Straniero Sergio (2007: 407–23) conducted a microanalysis of shifts in footings signalled by the use of personal pronouns, highlighting the different capacities projected by the use of singular/plural first vs. third person. This paper deals with film festivals as a sui generis form of media entertainment by bringing together the notion of footing with the theoretical construct of positioning in order to explore how interpreters relate to primary interlocutors in one of the dialogic contexts of the Giffoni Film Festival (hereafter GFF), namely live streamed meetings with famous international guests. An attempt will be made to devise a descriptive typology of interactionally constructed interpreter positions in this context. The detected features will be then compared with those of post-screening Q&A sessions with the crews of the films presented at the festival: events in both categories take place onsite and in-person, with the main discriminating factor being that the latter are not live streamed. The data are analysed also with a view to highlighting similarities and differences between streamed film festival interpreting and televised MI so as to evaluate the influence of the new media on interpreting practices. 

2. From a wide shot of media interpreting to a close-up on film festival interpreting

This paper approaches film festival interpreting (hereafter FFI) as a form of media interpreting. MI currently ‘refers to a very broad and diverse category of cross-language interpreter-mediated communication that falls within the field of audiovisual translation’ (Dal Fovo 2020: 315). The expression has become the preferred one in literature because it’s the most comprehensive and inclusive way of referring to content produced by old and new media (Pöchhacker 2011: 22).

International film festivals are among the current forms of entertainment normally found on online streaming platforms. As first defined by Raffaela Merlini (2017), FFI can be subsumed under MI even though it is neither an activity organised by the media and performed within a studio (unlike talk show interpreting, for instance) nor a situationally displaced translation of events taking place in a remote location, staged and edited for the benefit of a television audience (unlike the live interpreting of breaking news). While the ‘working from a screen’ criterion (Mack 2002) is a side rather than a core condition, interpreting at film festival events that attract media coverage can be defined as ‘interpreting also for the media’, and by extension as MI, on at least three counts: the multiplicity of addressees and communication levels resulting from an additional remote audience; the high degree of professional exposure experienced by the interpreter; and the reference to an ‘ethics of entertainment’. 

The literature on the topic defines broadcast communication as talk constructed for absent overhearers (Heritage 1985; Hutchby 1995). As such it entails two discourse levels: (1) ‘reporting interaction’, that is the relationship between the off-screen audience and what is being discussed on screen; and (2) ‘reported interaction’, namely the relationship between on-screen participants themselves (Bondi Paganelli 1990: 45–55). Given the primacy of the former level on the latter, TV viewers may be considered, to all effects and purposes, the intended addressees of media communication. In film festival on-stage talk, complexity increases as the levels of interaction become threefold and are rearranged in order of salience as follows: (1) a primary level made up of the interaction between on-stage participants and the public who is physically present in the theatre; (2) a secondary interaction between on-stage participants; and (3) the relationship between on-screen interaction and off-screen audience. The host, the guest and the interpreter interacting on stage are aware of being broadcast, yet their behaviour is influenced most of all by the reactions of the flesh-and-blood spectators, who are often cast in the role of addressees, as when the host engages directly with them (Merlini 2017; also see Merlini and Picchio 2019; Picchio 2023a). This kind of interacting public is different from TV studio paid audiences, given that the latter’s contributions, if envisaged at all, are carefully scripted (Rizzo 2018).  

The second feature supporting the categorisation of FFI as a form of MI is the interpreter’s exposure and visibility (Jiménez Serrano 2011) which, in FFI, vary according to the degree of media coverage (local, national or international) and the resonance of the event, with the awarding of prestigious prizes to worldwide celebrities attracting much more media attention than short interviews with first-time film makers and novice actors. In any case, media exposure of interpreting mistakes may jeopardise a professional’s image, and have adverse effects in terms of stress levels, as discussed, among others, by Birgit Strolz (1997) and Ingrid Kurz (2002). Media interpreters are also conscious that the quality of their performance is likely to shape public perception of the whole professional category (Pöchhacker 2011: 23). Sergio Viaggio (2001: 29) explicitly refers to the media interpreter’s ‘heavy burden of incarnating the profession before the general public’. 

The third feature is the interpreter’s orientation to what David Katan and Francesco Straniero Sergio (2001) perceptively identified as the ‘ethics of entertainment’, making interpreting behaviour consistent with the playful macrofunction of TV communication. Talk is not so much aimed at seeking and giving information: it goes beyond the phatic function of maintaining social contact and is instead specifically designed to keep up the interest of viewers while ensuring that all participants – both on- and off-screen – feel comfortable with what they are doing. In compliance with this overarching ‘comfort factor’, interpreters are called upon to contribute to the entertainment goal of the media event, with their renditions being often exploited to create ad-lib vignettes. Form thus prevails over content, with oratorical style making for a telegenic performance (Pignataro and Velardi 2013).

3. Remakes of footing and positioning

As originally put forward by Goffman (1981: 128), footing was defined as ‘the alignment we take up to ourselves and the others present as expressed in the way we manage the production or reception of an utterance’. Goffman developed the construct into a ‘participation framework’ made up of a production format accounting for the speaker’s potential alignments as ‘animator’ (articulating utterances), ‘author’ (scripting the lines), and ‘principal’ (taking responsibility for the words spoken), and two broad categories of recipient roles, namely ‘addressee’ and ‘bystander’. Whereas addressees are fully ratified and active participants in interaction, bystanders are adventitious listeners who happen to be in the perceptual range of the encounter, and whose access to it, however minimal, is perceivable by the official participants (Goffman 1981: 132). Being unratified, bystanders should politely disavail themselves of any opportunities to listen and should act in such a way as to maximally encourage the fiction that they are not present. In her critical review of Goffman’s conceptualisation, Marta Dynel (2011) points to a classificatory problem concerning the role of ‘bystander’, arguing how the distinction between its two subcategories of ‘overhearer’ and ‘eavesdropper’ is based on a very elusive criterion, namely inadvertent/unintentional vs. engineered/surreptitious listening. As will be addressed later in this paper, whether a bystander is to be regarded as a ratified participant – a view to which Cecilia Wadensjö (2015: 167) subscribes –, a non-ratified participant – as in Goffman’s work –, or even a non-participant – as ventured by Dynel (2011: 459) – is still an open question with an interesting potential for further elaboration. 

Adapting Goffman’s framework to interpreting, Wadensjö (1998) complements his notion of production format with a breakdown of the hearer’s reception format into three modes of listening, ‘reporter’, ‘recapitulator’, and ‘responder’, which differ on the basis of the interpreter’s increasing degree of agency, with the last of these three footings signalling full interactional autonomy. Inspired by both Goffman and Wadensjö, the model of interpreter footings developed by Merlini and Favaron (2005) with reference to the specific setting of speech therapy seminally connected the ways in which the interpreter is addressed or not addressed by primary parties with his/her speaker alignments as expressed through the use of personal pronouns and direct/indirect speech. In this model, seven types of footing are suggested (‘reporter’, ‘narrator’, ‘recapitulator’, ‘responder’, ‘pseudo-co-principal’, and ‘principal’) with the supposedly canonical one being the alignment of ‘reporter’, where the interpreter is unaddressed by primary speakers, and his/her renditions are in first-person direct speech. Pöchhacker (2012: 57) later noted how a turn-by-turn analysis of this kind, which links the interpreter’s rendition back to the preceding original utterance, may be too mechanical, in that alignment shifts are often embedded in a more enduring footing that is shaped at the level of the communicative event. His contribution thus suggests a need to broaden the scope of footing analysis.

Moving on to a consideration of the MI literature, in Straniero Sergio’s work (2007: 34–7) the approach is twofold; while at a microlevel his investigation of footing shifts focuses once again on the use of personal pronouns and choice of address, at a theoretical level the concept is used as broadly encompassing all the moves that interlocutors make to regulate their dialogic exchanges. These include instances of initiating/responding, acknowledgment of understanding, metalinguistic comments, affiliation and so on, through which participants construct and negotiate their relations to one another, as well as to the contents and forms of their contributions to the speech event. Straniero Sergio’s multifaceted discussion demonstrates, on the one hand, how interpreters tend to adopt footings that display an independent participation status and, on the other, how their conversational alignments may be triggered by the interactional behaviour of the TV host. Wadensjö’s (2008) study of footing shifts in a talk show interview reveals a different state of affairs: in her data, the interpreter resists the host’s attempts to involve him personally in the show. More recently, Englund Dimitrova (2018) has analysed the case of a TV host-interpreter with experience in both professions, finding that conversational alignments are oriented more towards the showbusiness ethics of entertainment than to the typical conference interpreting one.

Going beyond the notions of role and footing, the theory of positioning espouses a quintessentially dynamic view of interaction (Davies and Harré 1990; Harré and Van Langehove 1999) and gives prominence to the joint construction and constant renegotiation of individuals’ identities. Positions stem from the discursive practices that interlocutors engage in and make available to one another during the unfolding of an encounter, and as such, they are non-predetermined – in other words, no fixed or generally applicable categories can be devised beforehand. In shifting between different discourses, people project a multiplicity of selves, each of which may contradict the selves located in past story lines, as well as those located in alternative ones. If ‘reflexive positioning’ (also referred to as ‘self-positioning’) is the process through which someone positions him/herself on the basis of the cultural, social, and moral systems and the emotional experiences embedded in their personal story, in ‘interactive positioning’ (or ‘other-positioning’) what one person says positions another. With reference to the latter, Davies and Harré (1990: 50) observe that interlocutors may be found to conform to the speaker’s interpretation of their story line either because they share it, or because they see some advantages in doing so, or simply because power asymmetries leave them no other choice; conversely, interlocutors may resist being positioned and may pursue their own story line, quite blind to the one being proposed by the speaker.

Mason (2005, 2009) and Merlini (2009) pioneeringly applied positioning theory to public service interpreting research in order to portray the fluid, locally managed and power-dependent process of identity projection.  Their analyses revealed how interpreters’ and primary interlocutors’ utterances influence one another through a complex weaving of discursive moves and countermoves. More recently, in her analysis of mental health and court interpreting encounters, Delizée (2021) has successfully combined the constructs of footing and positioning, looking at utterance-level alignments within sequence-level positions. The present study follows this same line of inquiry, turning its attention to the film festival setting. It builds on a preliminary investigation of discursive positions and audience design in FFI (Merlini and Picchio 2019), which revealed how interpreters tend to step into non-normative positions only if instructed or authorised to do so by either the host or the film festival guest.

An additional level of analysis is introduced here drawing on a further meaning of the term ‘positioning’, which ‘in a more concrete sense […] is used to describe the physical position of the interpreter in face-to-face encounters’ (Pokorn 2015: 312–13). The significance of positioning in space has been problematised in particular in legal and healthcare interpreting settings. In the latter field, the widespread assumption is that a triangular seating arrangement enables the patient, the healthcare provider, and the interpreter to keep eye-contact, allowing for unobstructed direct interaction between the two primary speakers. Such a configuration has been shown by Wadensjö (2001) to ensure the equality of all participants present in the shared visual radius, and to favour both the interpreters’ impartiality and their active integration in the communicative process. Little is still known about physical positioning in other contexts of practice. To our knowledge, the physical positioning of interpreters in MI has never been the object of in-depth scrutiny, even though its repercussions on such aspects as visibility, entertainment and comfort are all too evident. The few cursory observations which can be found in Straniero Sergio (1999, 2007), Wadensjö (2008), Sandrelli (2015) and Merlini (2017) all point to the importance of proxemics for a successful spectacle-oriented performance. As will be discussed in the following sections, our data show that the participants’ positioning in space varies significantly between streamed and non-streamed events.

4. The Giffoni Film Festival: Setting the scene

The GFF, one of the most important Italy-based international film festivals, has been taking place since 1971 in Giffoni Valle Piana, a small town near Salerno in the Campania region of southern Italy. Its uniqueness lies in the fact that the audience and jury are entirely made up of children and young people coming from Italy and abroad, which is the reason why extensive dialogue interpreting between Italian and English (the GFF official languages) is made available. Differences between the FFI setting studied in Merlini (2017) and the GFF context are twofold: (1) GFF jurors are primary interlocutors rather than mere spectators as they directly address the guests with questions and comments; and (2) some of the events are broadcast in live streaming on YouTube.

Referring to the digital era, Bolter and Grusin (1999: 55) argue that ‘all current media function as remediators’ in the sense that they respond to, redeploy, compete with, and reform other media. Each new means of communication remediates some aspects of preexisting media, by introducing innovations in the cultural and mass media ecosystems, and by drawing upon old structures and models which are yet transformed and recombined. As television remediated both radio broadcasting and cinema, web-based communication remediates television. In a non-linear historical progression, however, older media can also remediate newer ones: television, for instance, has refashioned itself into smart TV to resemble the World Wide Web. In shaping one-to-one, many-to-many and one-to-many communications (Jensen 2021), the multimodality and interactivity (Cosenza 2014) of the new digital media have altered public-private, space-temporal dynamics so that today’s people-users live an ‘onlife’ (Floridi 2015) in a ‘real virtuality’ (Castells 2009). By the same token, the concept of audience has become more participatory, fragmented and interconnected (Bentivegna and Boccia Artieri 2018). Drawing on the concept of remediation, this study will look at the live streamed film festival data under analysis in search for significant interactional, interpreting and entertainment shifts from linguistically-mediated TV talk shows.

The excerpts discussed in the paper are taken from the GFFIntD multimodal corpus (Picchio 2023a; forthcoming), which includes 23 audio/video recordings for a total duration of over 920 minutes. For the purposes of this paper, the qualitative analysis will focus on five videos of live streamed encounters with famous celebrities (so-called Meet the Jury, hereafter MJs – see Table 1). The characteristics identified will be subsequently set against the interactional features of non-live streamed post-screening Q&A sessions (hereafter DPPs)^[1] discussed elsewhere (Picchio 2023a; see also Merlini and Picchio 2019; Picchio 2023b) and used here as a foil. Both data sets come from the 47^th edition of the GFF which took place in 2017, and involve the same configuration of participants:

1. an interpreter (I) working in the consecutive mode;
2. an Italian host/chairman (H);
3. either one foreign guest (G) or more than one;
4. Italian and foreign members of the juries (J) making up the in-theatre public; the jurors are aged from 13 to over 18.

Table 1: MJ videos

On the other hand, the two contexts differ on the following counts:

1. the live streaming: the MJ encounters were followed by the additional remote audience of live streaming users, whereas the DPPs were not made available by the festival management to wider distribution^[2];
2. the types of guests: in MJs, they are famous celebrities who are invited to receive an award or present a film première, whereas in DPPs guests are members of the film crew who answer the jurors’ questions following the screening of the movie in competition;
3. the status of jurors: in MJs, they form an audience of fans who meet their favourite stars, whereas in DPPs they interact with the guests in their full capacity as jurors who will eventually cast their votes;
4. the interpreting mode: consecutive without note-taking in MJs, with note-taking in DPPs.

The data were transcribed using ELAN (2021), currently considered by various scholars (see among others Davitti 2013; Bao-Rozée 2016; Vranjes and Brône 2021) as the best tool for the transcription and analysis of dialogic multimodal data. Besides enabling the alignment of transcripts to their corresponding video recordings, its horizontal layout allows for the annotation of verbal, interactional, and multimodal features within an easily searchable tier structure. One of the analytical tiers, named FootPos, was created specifically to annotate the interpreters’ footings and their discursive as well as spatial positions. 

5. Lights, camera, action… onto data analysis and discussion

Following an illustration of participants’ positioning in space, in this section the data analysis will focus on discursive moves described in terms of the interconnection between footings and positions. To remain within the movie metaphor, representative sequences of interaction will be played in slow motion to capture the interpreter’s participation status that emerges from the relational configuration of reflexive and interactive positioning moves in any given sequence. Four categories will be discussed, namely ratified bystander, recapper, primer, and sidekick performer. The suggested typology is to be seen as an attempt at drawing the contours of varying degrees and modes of interpreter agency in the specific context under scrutiny, rather than a systematisation into widely applicable categories.

5.1 Framing in on the interpreter in live streamed Meet the Jury events

The participants’ seating arrangement on stage is fixed throughout all the MJ encounters and is decided by the festival management. It is thus a clear instance of spatial other-positioning aimed at accommodating the event to the requirements of live streaming. The guest sits between the interpreter and the host (Figure 1).

Fig. 1: Positioning in space

The three chairs face the cameras and the jurors in the theatre stalls, and are placed in such a way as to form a triangle (Figure 2). While this arrangement is clearly designed to facilitate interaction between on-stage participants, differently from the more traditional dialogue interpreting triangular arrangement here it is the guest rather than the interpreter who sits in the middle. As the main attraction, the guest thus occupies the focal point of the stage and is always visible on screen, often in close-up view for the benefit of the remote audience.

Fig. 2: The triangular seating arrangement

Camera shots often include the host as well, testifying to his^[3] functional role as ‘gate-keeper’: it is he who welcomes the guest, presents the jurors, introduces the awarding ceremony, and sets the pace of the encounter. Unlike TV talk show hosts (Straniero Sergio 2007: 105–106; 117), however, the GFF host does not decide the topics of conversation, which depend on the jurors’ contributions.

As for the interpreters, they are instead rarely on screen, mostly becoming visible only when directly addressed by the guests. The discussion that follows will bring them back into the analytical picture. The four positions identified below are presented in order of increasing interactional involvement on the interpreter’s part, which does not necessarily reflect a corresponding degree of autonomous agency.

RATIFIED BYSTANDER 

In excerpt 1, the host addresses the guest to introduce the jurors and present the welcome video they prepared for him. The interpreter sits with them on the stage but is not introduced.

Excerpt 1 (MJ_2_47)^[4]

After being excluded by the host throughout the opening sequence, the interpreter takes the floor in turns 8 and 11 to translate the guest’s thanks for the welcome video, overlapping partially with H1 (turns 10–11). His subsequent routine involvement in the speech event follows upon one juror’s question in English. The exclusion of interpreters from the interaction typically occurs when the host interacts directly with the guest in English without asking for a translation. These moments precede or follow the Q&A session proper and are a common practice at the GFF^[5]: interpreters expect them, and jurors, who are equally familiar with these introductory sequences, are supposed to be able to grasp the general meaning – even if Italian jurors (especially the younger ones) are not likely to speak English fluently. On the other hand, those live streaming users who either do not understand English or do not usually follow the GFF events are denied access to comprehension. This situation is completely different from TV talk shows, in which host-guest monolingual exchanges are always translated for the benefit of the remote audience (Straniero Sergio 2007: 425).

To describe this type of interpreter’s other-positioning by the host, the label ‘ratified bystander’ has been coined. Following Dynel (2011), neither ‘overhearer’ nor ‘eavesdropper’ were thought to be suitable descriptions: rather, the construct of ‘bystander’ was drawn upon to recall the Goffmanian (1981) concept of inactive participation in the communicative event. In contrast to his original categorisation, however, excerpt 1 shows that the interpreters’ presence is ratified by the GFF management: interpreters do not have to pretend they are absent, nor do they disavail themselves of any opportunities to listen, which is why the bystander status is here qualified as ‘ratified’.  This position extends over a sequence of host-guest exchanges, which ends when I1 gets involved in the interaction through two instances of self-positioning as ratified participant.

It is also worth noting that when interpreters are cut out of the conversation, they are virtually invisible on screen. In excerpt 1, I1 features only in wide camera shots: out of 10 shots (the majority of which are close-ups of G), I1 is included only in four very wide ones (Figure 3).

Fig. 3: A very wide shot in which the interpreter is hardly visible

In other MJ encounters, the interpreters are never present on screen. As a result, whereas for the jurors the interpreters are ratified bystanders, for the streaming users they are to all effects and purposes invisible, in some sequences inaudible, and in any case anonymous.

RECAPPER

In excerpt 2, the juror asks the guest a tricky question (turn 1), which results in a dyadic sequence between them (turns 3–10): first, the guest makes sure she has grasped the meaning of the question, and then she tries to answer without ‘spilling the beans on the bad experience’. The interpreter translates only at the end of the guest’s answer (turn 14).

Excerpt 2 (MJ_1_47)

I4 starts her rendition with ‘allora’, an Italian discourse marker that may be used to introduce a summary but also to project tentativity in what follows. The turn provides in fact a recapitulation of the main points of the guest's contribution. As she takes on the footing of recapitulator, her summarized rendition results in omissions (‘I’d like to say a few choice words to certain people’), unclear formulations (‘they don’t teach you but then you learn’)^[6], and one serious mistranslation (‘I can say I won’t do it again’ when the guest argues the opposite). Nevertheless, the interpreter manages to convey the key message, that is, the need to find the right balance between family life and work.

Looking at the whole sequence rather than at the interpreter’s turn, I4’s positioning is prompted by the guest herself. It is G who interacts directly with the juror without allowing for the interpreter’s renditions (turns 3–10), and then explicitly, albeit politely, asks the latter to wait further and to translate only at the end of her answer (turn 11). To signal a distinction between sequence-level instances of interactive positioning and the traditional footing of recapitulator, this type of other-positioning of the interpreter by a primary participant has been named ‘recapper’. As the interpreter is acting on cue here, the term has been deemed appropriate also to evoke a tool-like functioning; in military jargon, the recapper was an ignition device used in the past to reload firearms and enable them to detonate reliably in any weather conditions^[7]. In excerpt 2, I4 waits to be externally ‘triggered’, thereby accepting G’s interactive positioning, as she apparently realises that the situation is a delicate one and that G may require an extended turn to produce a non-committal answer. 

Technically, excerpt 2 recalls the ways in which talk show interpreters group together several original turns into a single rendition – even resorting to similar turn-opening discourse markers. In the TV setting, however, during extended dyadic monolingual sequences interpreters are seen to actively self-position as translators, given that they repeatedly try to get the floor back, even by means of interruptions (Straniero Sergio 2007: 240–4). A second difference between talk shows and the GFF, as emerges from the analysis of this excerpt, is that in the former interactional coordination is completely in the hands of the host (Straniero Sergio 2007: 215–23), while in the latter it is the guest who gives instructions to the interpreter.

PRIMER  

In excerpt 3, a juror formulates a convoluted two-part question in English (turn 1) which the star does not grasp completely.

Excerpt 3 (MJ_2_47)

As the guest reformulates the juror’s long-winded turn to make it clearer to himself (turn 3), I1 takes an unsolicited initiative (footing of principal – turn 4) and condenses it in one short question. This helps G focus, as he repeats to himself the interpreter’s words literally (turn 5). Treating his contribution as an aside meant just for the guest’s benefit, I1 does not speak into the microphone, thus managing the situation metaphorically behind the scenes. In light of this autonomous contribution by I1, later on in the conversation (turn 7), just as he is about to conclude his reply, the guest turns his gaze to the interpreter, presumably to ask for confirmation that he has fully answered the question. Although I1 is off screen, G’s response ‘yeah he is just part of me now’ coupled with his head nodding would seem to indicate that the interpreter has in fact confirmed. Here, I1 thus adopts the footing of responder.

In excerpt 4, a juror asks the guest a question about Italian cinema (which Italian film he likes the most and why); the star addresses her a request for confirmation to make sure he has understood the question properly (turn 2). The interpreter butts in, acting as unaddressed responder^[8], and his answer (turn 3) partially overlaps with J’s (turn 4). The guest turns to I1, repeating the words ‘Italian film’ (turn 5) and seeking for further confirmation, and the interpreter nods (turn 6) before translating the juror’s initial question for the benefit of the Italian-speaking audience. Later on, the guest tries to remember an old Italian movie ‘about cycling’ and looks again at the interpreter, implicitly asking him for help (turn 10). I1 intervenes as addressed responder suggesting the title of a very famous neorealist Italian movie, ‘Ladri di biciclette’ (turn 11). At this point, the host himself validates I1’s suggestion (turn 12), mentioning also the film’s director (turn 14) and the release year (turn 16).

Excerpt 4 (MJ_2_47)

In excerpts 3 and 4, the interpreter primes the guest by giving him the pieces of information he needs to first understand the questions and then provide suitable answers. Both self-positioning and other-positioning can be seen to be at play; more specifically, instances of the former process occur when I1 answers questions which are not addressed to him, while instances of the latter entail the guest turning to the interpreter for help. Whereas the footings of principal and responder are determined on the basis of single turn exchanges (as in Merlini and Favaron’s model), here the interpreter’s position emerges from the whole sequence and takes into account not only the participants’ verbal contributions but also their body language, including gaze. To describe this position, the term ‘primer’ has been chosen to indicate the interpreter’s information-giving function^[9] and, at the same time, suggest the ‘prime’ part he plays in these stretches of interaction.

SIDEKICK PERFORMER

In excerpt 5, the guest recounts an episode that happened to him while shooting a TV series: covered in bees, he was stung twice. Pointing at his microphone and alluding to the male reproductive organ, he illustrates by means of a gesture where he was stung the second time (turn 9). Mindful of the underage audience, while both are off screen the host intervenes in English saying that G was stung ‘there’ (turn 10), and I2 keeps the double entendre of G’s non-verbal allusion commenting ‘in Italian it’s the same it’s the same thing’ for the benefit of the Italian-speaking jurors and remote public (turn 11). Out of curiosity, the guest leans towards the interpreter – who is consequently brought back into the frame – and asks him to explain what he has just said in Italian (turn 12). I1 acts as responder (turn 13), translating his previous turn into English and miming the guest’s previous gesture.

Excerpt 5 (MJ_3_47)

The interpreter is other-positioned as a sidekick performer by the guest who explicitly prompts his ancillary contribution to the show. Sequences like this one differ from the Taormina Film Festival live TV broadcast data discussed in Merlini (2017) where interpreters’ applause-relevant involvement, generally initiated by the host, was found to display a higher degree of agency resulting in the position of host’s ‘co-performer’.

5.2 Live streamed MJs vs. non-live streamed DPPs

The analysis of MJs presented above has revealed how the presence of streaming determines a less dynamic unfolding of the event. Mapping the prevalent interactional flows in a diagrammatic (freeze-frame) form, Figure 4 shows the spatial and discursive positions that characterise the MJ encounters. Single-headed arrows indicate the interpreters being other-positioned as ratified bystanders, recappers and sidekick performers, whereas the double-headed arrow refers to their other- and self-positioning as primers.

Fig. 4: Spatial and discursive positions

The diagram clearly illustrates that interactional flows cluster around the guest and the interpreter, with the guest playing a key role in streamed MJ events: s/he positions the interpreter as recapper and sidekick performer, and solicits or receives information when the interpreter contributes to the interaction as primer. Visually, the guest is also the focal point on stage and is permanently visible on screen. Evidence of the host’s power as gate-keeper of the event is provided by his decision to position the interpreter, on some occasions, as a ratified bystander. Although the topical development of the encounter depends on the jurors as fully ratified participants, they do not appear to exert any direct power on interpreter positioning.

If these results are compared to the findings of the analysis carried out by Picchio (2023a) on non-live streamed post-screening Q&A sessions with film crews (DPPs), the latter reveal a much greater diversity of both spatial and discursive positions. In terms of physical arrangement, depending either on personal preferences or on the festival management’s instructions, participants are alternatively found to be on stage or under it, and to stand or to sit. As for the range of interpreters’ discursive positions, seven different types were identified in the DPP data, ranging from zero to fully autonomous agency:

• ratified bystander: this position occurs at various points during the event – rather than just in the opening sequences – when the hosts take upon themselves the task of interpreting;
• linguistic support (cf. Merlini 2009): the interpreter is required by the host to provide minimal renditions at certain specific moments in the interaction;
• recapper: unlike in the MJs, this position is prompted in the DPPs by the host rather than the guest;
• coordinator of turn-taking;
• moderator: when the Q&A session with the film crew is particularly heated, interpreters autonomously intervene, both textually and prosodically, to mitigate (cf. Caffi 2001) those original turns in which criticism is addressed to the movies in competition;
• performer: interpreters produce jokes, quips and hilarious vignettes entirely on their own initiative;
• host: if a chairperson cannot be present, interpreters are called upon to host the event while also translating their own and others’ turns.

When positioning is compared and contrasted between streamed MJs and non-streamed DPPs, most types are seen to characterise either one or the other. The only two that are found in both settings are recapper and ratified bystander. However, this overlap is only partial, given that the recapper position is guest-initiated in MJs and host-initiated in DPPs, and the ratified bystander position is much more frequent in the latter context. Hypotheses about this stark divergence between the two sub-corpora are formulated in the concluding section.

6. Finale: FFI between old and new media

This paper has presented a qualitative analysis of the spatial and discursive positions that characterise live streamed interpreter-mediated encounters at the 47^th edition of the Giffoni Film Festival. By bringing together a turn-by-turn analysis of footing shifts with an examination of sequence-level dynamics, four other- and self-initiated positions have been identified, for which the labels of ‘ratified bystander’, ‘recapper’, ‘primer’ and ‘sidekick performer’ have been suggested. Taken together these positions provide evidence of both a visual and interactional prominence of the guest in these kinds of events. When viewed against the backdrop of non-streamed Q&A encounters in the same context, one finds that the latter exhibit a much higher degree of variability both in terms of physical arrangement of participants in space and their discursive behaviours. Furthermore, in non-streamed interaction all the identified positions point to a marked conversational dominance by the host and a more autonomous conduct on the interpreter’s part. One may thus safely infer that the presence of a streaming audience tends to limit the institutional participants’ freedom of agency by imposing a more rigid routine on the event. Since in non-live streamed encounters not only the host but also the interpreter presumably feels freer to act, it is also possible to hypothesise that GFF interpreters conform to the positions being suggested by primary speakers not so much because they see themselves as having no choice (hence, out of lack of power; see Davies and Harré 1990: 50), but because they intentionally adjust their conduct to the requirements of the context, prioritising the success of the event over their own professional image. This interpretation gains further strength if one combines the impact of live streaming with that of juror status. Of the four features that distinguish MJs from DPPs, the different capacities in which the jurors participate in the two types of encounters may have contributed – albeit indirectly and to a minor extent – to determining interpreter behaviour. In a situation where the jurors are not wearing their ‘institutional’ hat and are just enjoying the show as any other audience, the interpreters (and the hosts) may feel that keeping a low profile is the best strategy to allow the guest-stars and their fans ample room for manoeuvre. When, on the other hand, the only public is the in-theatre one and the jurors step into their role asking questions and making comments on the films in competition, the interpreters (and the hosts) may feel that a higher degree of agency is required on their part to coordinate interventions, soften criticism, and maintain a cheerful atmosphere. 

Given that film festival interpreting has been subsumed here under media interpreting, some concluding considerations on the differences between streamed and TV talk show interpreter performances are also called for. Firstly, the analysis has shown that when positioned as ratified bystanders, GFF interpreters are totally excluded from the conversation, and consequently no translation into Italian is provided to the streaming audience. On the contrary, the comprehension needs of TV viewers are prioritised so that only on very rare occasions and for brief stretches of talk are interpreters marginalised or altogether deprived of their translation space. Secondly, whereas TV hosts have full control of on-screen interaction, management of GFF streamed events is left in the hands of the guest-stars who take centre stage, both conversationally and on screen, thereby becoming the principal performers. Thirdly, while in talk shows people who do not know each other, and/or have nothing in common, and/or do not necessarily have intimate relationships chat as if they were close friends for pure entertainment purposes (Straniero Sergio 2007: 73), the GFF jurors are truly interested in learning about their favourite stars’ careers and private lives; thus, the informative function of communication plays a significant role, and the entertainment one – though always present, as the sidekick performer position demonstrates – is not the only or even primary goal.

The absence from the camera shots of either the host or the interpreter or both is quite common at the GFF, which implies that online users have only limited access to the full interactional context and are therefore not the main intended addressees. Recalling the norms of TV interpreting as identified by Straniero Sergio (2007) – namely ethics of entertainment, comfort factor, and exposure –, this invisibility would be unimaginable on TV where interpreters often step into the limelight as full-fledged performers, thus becoming famous public figures (Straniero Sergio 2007: 186–93). While the requirements of the streaming are seen to impose on the event a more rigid structure which limits the room of manoeuvre (both spatial and discursive) of both the host and the interpreter, the interpreter’s occasional exclusion from interaction and frequent invisibility on screen mean that the needs of the remote audience are not really catered for, or at least not as much as those of the flesh-and-blood jurors who sit in the GFF cinema theatre.

Resonating with mass media literature, this paper seems thus to confirm that live streaming is not, as some would argue, ‘like TV only better’ (Bolter and Grusin 1999: 3). It is rather a new sui generis form of media communication, within which streamed film festival interpreting is found to remediate some interactional and entertainment dynamics of televised interpreting.

About the author(s)

Raffaela Merlini is Senior Lecturer of English language, translation and linguistics in the Department of law, economics, politics and modern languages at LUMSA University (Rome). She previously held posts at the universities of Hull, Salford, Trieste, and Macerata. She has published in the field of Interpreting Studies mainly on dialogue interpreting topics. Her current research interests focus on the interactional and relational dynamics of face-to-face interpreter-mediated talk in the healthcare, media and business settings. Her scientific contributions have appeared in peer-reviewed international journals and in volumes published, among others, by Benjamins and Routledge. She authored the entry on “Dialogue Interpreting” in the Routledge Encyclopedias of Interpreting Studies (2015) and of Translation Studies (2019). She co-edits a book series on translation (Prolusioni, EUM).

Laura Picchio holds a PhD in Humanities and Technologies (Interpreting Studies) from the University of Macerata, Italy. Her PhD dissertation focuses on the impact of digital media on film festival dialogue interpreting performances (in terms of agency, positioning and visibility on screen) and audience reception. Her research interests also include TV interpreting, in particular talk show interpreting. She works as an adjunct professor of English language at the University of Macerata, lecturing within the Primary Teacher Training Degree Course. Among other publications, she contributed to Issue 28 of The Interpreters’ Newsletter (2023) and to Issue 22 of Revista Tradumàtica (2024). She took part in the 11th EST Congress (2025), and co-authored a paper presented at the 2nd Forlì Workshop on corpus-based interpreting studies (2025).

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

•  Previous article
• Next article