Displaying Inner Experience Through Language and Body in Community Theater Rehearsals

Using multimodal conversation analysis, we investigate how novices learning the “inner body” acting technique in the context of a community theater project share their experiences of the bodily exercises through verbal and embodied conduct. We focus on how verbal description and bodily enactment of the experience mutually elaborate each other, and how the experienced sensorimotor and affective qualities are made to be witnessed and recognized by the others. Participants describe their experiences without naming qualities. Instead, a display of the experienced qualities is made accessible to others through coordinating the unfolding talk and bodily conduct. In particular, we show how grammatical and action projection is fulfilled by interconnected verbal and embodied conduct, with body movement and posture giving off ineffable experiential qualities. The moving body appears both as a source of the experience and as a resource for depicting perceived qualities to others; additional resources (non-specific person reference and gaze aversion) contribute to organizing the subjective and intersubjective layers of the reflection of the experiences. The study contributes to and extends recent research on sensoriality in interaction by focusing on phenomena of proprioception and interoception. The data are two cases drawn from 60 h of video-recordings made in the context of a devised community theater project. The data are in Finnish with English translations.


Introduction
This study investigates interactions between a theater director and participants in a devised community theater project. Using multimodal conversation analysis, we examine ways in which the participants describe their sensorimotor experiences of an acting technique exercise during comment rounds elicited by the theater director. We focus on the participants' use of multiple semiotic resources to tell about their experiences concerning, e.g., effort of movement and affective states associated with particular ways of moving. The analysis contributes to research on sensoriality by showing how participants communicate about experiencing their own moving bodies and, in so doing, make sensoriality part of social interaction.
The participants' accounts of their experiences during the exercises involve proprioception, the perception of movement (also a feature of kinesthesia), body position and balance (Tuthill & Azim, 2018), and interoception, the perception of body-internal sensations, such as fatigue and muscle tension. The significance of interoception in emotional and social life (e.g., pain, anxiety, addiction) is being researched in neuroscience and psychology (Ceunen et al., 2016). In our data, practicing and reflecting on the acting exercises are methods to enhance body awareness (see Keevallik, 2021), which serves as a creative resource in the theater-making and can also serve other areas of the participants' lives (Izarra & Sirviö, 2018).
We focus on how verbal description and bodily depiction of the experience mutually elaborate each other, and how the experienced sensorimotor and affective qualities are made to be witnessed and recognized by others. The more specific interest lies in the observation that the speakers describe their experience of the bodily exercise very vaguely, without naming a particular quality. Instead of descriptive language such as nouns (e.g., "joy") or adjectives (e.g., "tired"), they provide unspecific and fragmented descriptions of the event (e.g.,"when one starts walking then there stays the feeling of like how one continues the walking"), while the more specific experienced quality transpires through bodily conduct. We will show that in coordinating verbal descriptions and embodied conduct, participants make use of projection (anticipation) at the level of action and the level of grammar (Auer, 2005): to unpack a claim (e.g., "it was different"), and to anticipate a certain verbal element (e.g., a complement clause). By coordinating these two types of unfolding projections with trajectories of body position and movement, the participants manage to express experiential qualities without direct linguistic coding and categorization of the experience (see Levinson & Majid, 2014;Mondada, 2021a). The moving body appears both as the source of the experience and as a resource for depicting the perceived qualities to others.
We suggest that the use of unspecific language serves making a re-exploration of the experience witnessable to others, involving the participants' orientation to here-and-now intercorporeality (Meyer, Streeck & Jordan, 2017). We also discuss how and when the participants do arrive at exact categorizations (e.g., "stiff"), and how these cases are different in terms of social action.
First, we describe the theater project and the data, video recordings of exercising the "inner body" acting technique. We then provide an overview of research in gesture studies and multimodal conversation analysis pertaining to our study theoretically and methodologically. In the analytic section, we analyze two cases in which the participants comment on their experience of an "inner body" exercise. In the discussion, we summarize the findings and discuss the emergent character of, and interplay between, the implementation of verbal and embodied means to share bodily, sensorial "inner" experiences with others.

Context of Study: Exercises of the Inner Body Acting Technique
Our study draws from 60 h of video recordings made in the "Art as Work and Working Tool" project (Raevaara, 2020;Savijärvi & Ihalainen, 2021;Visakko, 2020aVisakko, , 2020b. The project was a collaboration between researchers and artists around the yearly "theater summer job" project , in which the City of Helsinki together with Kiasma theater hired young adults with no previous acting experience to work for six weeks as artists and prepare a performance for the URB street art and culture festival. Their day-long sessions were video recorded once or twice a week with three cameras. We use recordings from 2016 with six participants. 1 All participants present in the extracts provided written, informed consent to the recordings and their use for scientific publications. The exercises in focus are based on the acting technique developed by the actor and theater director Michael Chekhov (1891-1955) (see Chamberlain, 2018: 1-36). The technique combines body movement and use of imagination to embody a character for an audience (Daboo, 2007). Through the exercises, the performers direct their sensory and cognitive foci in specific ways to actively evoke sensations and "atmospheres" -this use of body-awareness is captured by the notion of the "inner body". After a set of exercises, the director invites the participants to tell what they perceived during the exercise by asking what a particular way of moving evoked in them, whether it was different from a prior exercise, etc. These reflections, organized as comment rounds, serve the participants/learners getting a grasp of the possibilities of the technique and give the instructor a possibility to keep track of the learners' adoption of the technique.
As part of the devised method of theater making, the "summer job" performance is based on materials that the participants produce based on their real life, and from which the professional artists then select and assemble elements for the performance (see Visakko, 2020aVisakko, , 2020b. Through writing texts, recording videos and taking photographs, the participants have worked on how to observe and describe their environment and personal experiences in ways that serve transforming them as material for art. The broader learnables of the project thus involve "sourcing" from experience by reframing and reworking objects of experience for artistic usages, 1 Pseudonyms are used in the transcripts, but the real names of the professional artists present in the excerpts are mentioned at their own request: "Maria" is the theater-maker and director Elina Izarra Ollikainen.
overall familiarization with an artistic attitude, and a focus on the body as a source and means of expression (Visakko, 2020a(Visakko, , 2020b. In an earlier study (Harjunpää et al., 2021), we analyzed how invisible and imaginary components of the "inner body" technique are gradually brought into existence by the instruction of the exercises. The instructor (theater director) increasingly treats certain aspects of meaning attributed to the inner body as known and shared by the participants. In this paper, we explore the participants' own claims and demonstrations of their understanding and adaptation of the technique, delivered as responses to the instructor.

Talk and Bodily Action
Talk and gesture mutually elaborate each other so that often they cannot be understood separately from each other (Goodwin, 2000(Goodwin, : 1400(Goodwin, , 2003. This does not concern only the gesture but also the verbal part, in particular, when using pro-forms that are semantically indeterminate (Kendon, 2004: 161-164, 174).
Gestures accompanying some talk can appear parallel to the meaning of the talk, while in other cases they add, specify, or restrict in significant ways its interpretation (Kendon, 2004: 161, 176). The temporal relationship between talk and associated bodily conduct 2 is variegated, and meaningfully coordinated. Gesturing, body posture, gaze direction, and voice quality project upcoming talk and guide its interpretation in various ways (Enfield, 2009;Heath & Luff, 2021;Mondada, 2005Mondada, , 2007Schegloff, 1984;Streeck, 1995Streeck, , 2009a. Bodily action can also stand in for projected verbal elements (see Keevallik, 2013Keevallik, , 2015 for dance instructions). However, the precise content of gestures often becomes intelligible retrospectively (Kendon, 2004: 170), only once the corresponding talk has been produced.
Gesture typologies distinguish between different semiotic properties of gestures and their relationship to talk (see Müller et al., 2014: part VII). A major distinction is made between pointing gestures (Stukenbrock, 2015) and depicting by iconic gestures (Streeck, 2009b). While pointing centrally serves to locate referents in space, depicting serves to stage, by an analogical embodied activity, some object-property, action, or event for the recipients (Clark, 2016). In addition to hand-gestures, depicting can encompass facial gestures and the whole body (e.g., depicting "driving a car", Sidnell, 2006;Streeck, 2009b).
Bodily depictions can represent (past) events or objects as part of tellings or be used for modelling actions for recipients in instructional activities, but they can also concern hypothetical objects or plans  on proposing and creating scenes in opera rehearsals; Murphy, 2004 on architects planning buildings) and metaphorically represent abstract matters (Cienki & Müller, 2008; Szczepek Reed, 2021 on non-embodied concepts and inner workings of the body 1 3 Displaying Inner Experience Through Language and Body in… in singing training). Bodily conduct does not only show what a body did or should do, but it can serve "'unwittingly'" finding "image schemata (Johnstone 1987) to construe a phenomenon, a bit of content, an experience" (Streeck, 2009b, 151, ital-ics in the original). In the data of this study, the embodied displays of the "inner body's" workings involve construing more abstract qualities (e.g., fluency, ease or confidence), often combined with verbal portrayals of the experience in terms of mental-bodily states (e.g., a "feeling"). 3 Ethnomethodology has underscored that all descriptions are indexical, i.e., context-dependent and reflexively indexing the contexts of their production (Garfinkel, 1967: ch.1;Garfinkel & Sacks, 1970, Heritage, 1984. By the same property, descriptions are always incomplete, vague and expandable. Since each event or state of affairs can be categorized and described in numerous correct ways, descriptions are possible versions, tailored to their recipients and interactional context (Sacks, 1963;Schegloff, 1972;also Edwards & Potter, 1992. Likewise, depictions are not truthful representations but typically partial, stylized, selective, and exaggerated (Günthner, 1999;Tutt & Hindmarsh, 2011). In the analysis of verbal and embodied displays of experience in interaction, attention is drawn to how they come about, what they serve, and how they are tailored for the local context and responded to by co-participants.
Individual sensorial experiences can be brought to an interlocutor's attention through precise coordination of their verbal and embodied display. Studies on the expression of pain, for example, show how gesturing after an initial assessment projects further characterization (Heath, 1989(Heath, , 2002; also on pain, see Stukenbrock, 2008;de Fornel & Verdier, 2014). More recent studies concerning the presence of sensoriality beyond vision in social interaction, including smell, taste, touch, and bodily strain (e.g., Fele & Liberman, 2021;Liberman, 2012;Mondada, 2018bMondada, , 2020Mondada, , 2021aMondada, , 2021bWiggins & Keevallik, 2021) have shown how sensorial activities are made intersubjectively available and accountable using talk and embodied resources. A strained body and its movement can gain prominence through suspended syntax ; tasting activities can be coordinated through precisely timed sniffs and gaze aversion from co-participants while engaging with the object, before delivering an assessment (Mondada, 2018b(Mondada, , 2020. Gaze aversion can contextualize the change of the speaker's footing and the participation framework also in setting off re-enactments from the descriptive parts of narratives (Sidnell, 2006).
How participants coordinate such verbal and embodied conduct to assemble activities (Sacks, 1984) is made observable in the analysis through sequential analysis (Schegloff, 2007) supported by detailed multimodal transcription (Mondada, 2018a, see Appendix for transcription conventions).

Two Cases of Conveying Experience of an "Inner Body" Exercise
In this section, we analyze how two participants portray their sensorimotor experiences of engaging in exercises of the "inner body" technique. Their responses to the instructor begin with a statement of a propositional attitude that evaluates the task more or less explicitly. The turns unfold as complex, embedded clausal structures, in which the experience of doing the task is conveyed through the coordination and combination of verbal and embodied conduct. We focus on the following aspects: • The temporal coordination between verbal description and bodily depiction, • The syntactic design of the turns and the multimodal conduct when descriptive content has been projected, • Repetitive, cumulative gestures and body movements, • Speakers' gaze shifts, and non-specific person reference.
We first focus on how a participant (Miina) comments on an exercise of two movements, opening and closing. We then analyze how another participant (Tiia) comments on an exercise that makes use of symbolic imagery.

First Case: Miina
The representation of experience analyzed in this section is produced in response to the elicitation by the instructor and theater director, Maria, of the participants' reflections after exercises of the 'opening' and 'closing' tasks. While closing is executed by adopting a contracted, squatted position ( Fig. 1), opening is performed by taking a step to stand astride, opening the arms in a wide curve to the sides of the body, and ending up in an upright position (Fig. 2). After adopting the position, the participants withdraw from it and start walking, exploring how the initial position affects their walking. The participants' comments on this effect are organized in terms of differences between the two tasks.
Miina is the first one to comment on the exercise. She ends up describing her experience three times, as the instructor requests some clarifications. We analyze the first and third occasion (Excerpts 1.1 4 and 1.2).

3
Displaying Inner Experience Through Language and Body in… After initiating the elicitation of comments with generic questions (line 3), Maria produces a polar question concerning differences evoked by the opening as compared to the closing movement (l. 5-6). Miina confirms a feeling of difference in the first clausal unit of her response (l. 8) (n:y kyl siinä tuntuu että… 'well it does feel there that…'). She starts unpacking the statement about the feeling with a temporal 'when-then' structure (l. 8-9). It sets 'walking' as the kinetic context of the sensation to be reported ('when…walking'). This part of the turn includes a second reference to feeling/sensation, the noun tunne ('then…feeling'). Miina unpacks it with a clause-formatted modifier expressing that the feeling concerns the manner of moving (l. 10). Using the interrogative proadverb miten ('how') leaves the manner of walking open, and the particle niinku ('like'/ 'sort of') further frames the description as an approximation. The description of the experienced quality is still verbally unspecified (see Szczepek Reed, 2021: 6) when the turn ends in et(tä) (l. 10). Used turn-finally, it works as a particle that leaves the turn's import for the interlocutor to infer (Koivisto, 2014), here on the basis of the embodied cues.
During her turn, Miina touches her shoulders with her hands, and pushes twice her bent arms from a position with the elbows directed forward (Figure 3) to the back/sides (Figures 4a and 4b). In light of our understanding of the inner body technique, and the subsequent clarifications (see l. 14-21), this bodily display is associated with a sensation of the body being led effortlessly forward in a stream of continuous movement. The upper body movement is portrayed as central to the reported effect through the way the bodily display unfolds with Miina's verbal turn in the lines 8-11. When Miina begins to describe the walking in line 9, just before saying the feeling 'stays' (jää), she starts to raise her hands to touch her shoulders. During jää se tunne päälle, 'stays the feeling ongoing', she moves her elbows to the front (Figure 3), preparing a backwards stroke. After the complementizer et (which projects unpacking the 'feeling'), at the onset of miten ('how', line 10), her elbows reach the most retracted back position (Figure 4a/b) for the first time. The embodied trajectory from miten ('how') is summarized below (Scheme 1).
The first stroke back occurs when defining the feeling has been projected. During the talk, which identifies where the feeling applies but does not further describe it, her gesturing progresses to reaching the back-position for the second time. The second stroke back occurs without simultaneous talk. It is preceded by the turn-final että, which, together with Miina's gaze shift to Maria, appeals to the recipients' inference, while Miina holds the body position visible far into Maria's subsequent turn (not shown in transcript). The unpacking of the sensory experience, which has been expectable in particular slots of the unfolding turn, thus co-occurs with a precise movement phase.
In addition to the verbal content, the type of movement, gaze and person reference play a role in conveying the sensorimotor quality. Simultaneously with the first stroke back and reference to the manner (miten, 'how'), Miina turns her gaze to middle distance (see Heath, 1989; Figure 4b) away from Maria, whom she has been gazing at before. Miina's continuous self-touch points her shoulder and chest area as the approximate locus of the sensed effect. Together with the gaze aversion, they frame the ongoing action as remembering and re-sensing the bodily source here and now (see Mondada, 2018b on gaze aversion while tasting cheese).
Moreover, the repetitive, cumulative change of posture together with self-manipulation by hands can be regarded as "exploratory" (Streeck, 2009b: 70). Exploratory procedures are "epistemic actions" of the hands with the purpose of gathering Scheme 1 Multimodally unpacked projections in Miina's turn information, performed to discover features of an object that are not visible but revealed to active touch. The audience viewing the hand-movements can infer those features from the touch (e.g., hand sliding on a smooth surface vs. one with friction). This typically involves repetition and rhythmicity, or prolonged tactile contact. Gestural repetition in general can convey 'plurality', 'iterativity', and 'durativity' (Bressem, 2021). In Miina's case, her dynamic self-touch with repetitive movement makes a process of accessing an enduring sensorimotor quality observable to others, portraying the body as the source of its reactualization.
The effects of the prior exercise are presented not as distinctly subjective but to some extent generalizable by various verbal resources. The present tense (tuntuu 'feels', jää 'stays', lähtee 'starts') works to generalize, while the demonstratives in siinä tuntuu 'in it/there feels' (l. 8) and siihe jää 'to it/there stays' (l. 9) refer anaphorically to the event and treat it as recognizable for the recipients (see Laury, 1997). Miina uses for person reference the zero-person construction, in which the verb occurs in third person singular but without an expressed subject (ø in translations of the transcripts), as in lähtee, 'starts', l. 8, and jatkaa, 'continues', l. 10. This non-specific person reference form is used in Finnish to establish a viewpoint for the recipient or anyone to potentially identify with (Laitinen, 2006;also Raevaara, 2020). According to Laitinen (2006: 209), zero-person constructions serve as crystallizations of human experience that express "changes of state, emotions, perceptions, receivings or the experiences of human beings in particular situations". With the verb tuntuu 'feels', the zero is not in subject position, which makes its status as a zero construction less clear, but the experiencer/observer is nevertheless left implicit (Huumo, 2006). Indexing with the zero that anybody in similar circumstances could experience the same -while the speaker is through gaze and bodily conduct engaged in exploratory action -appeals to the viewers' potential access to similar experiences. Directed at the instructor, it demonstrates recognition of the learnable, the sought-for results of the exercised technique.

Clarifying Response
Miina's comment in the previous segment concerns particularly the 'opening' task. Excerpt (1.2) shows that it is somewhat ambiguous to Maria what kind of comparison Miina has made in relation to the closing task. She displays her understanding (see l. 18-19) that Miina could not sense an effect of the starting position in the closing task. Miina disconfirms that (by repeating the finite verb in a grammatically positive form twice, l. 20), explains how the sensations were different in the two tasks, and depicts the opening sensation again, in a way that emphasizes its effects. (Another clarification sequence, between Ex. 1.1 and 1.2, is not shown for reasons of space).

3
Displaying Inner Experience Through Language and Body in… Excerpt 1.2 U373_E_sisäinen liike_5: 30-5:40 When explicating the difference between the opening and closing, Miina describes the effects of the opening movement for the third time. Self-repairs and restarts (l. 20-21) suggest that she is searching for a description to remedy Maria's wrong impression. The overall trajectory of moving the arms from front to back (l. 21, Figures 7-9) corresponds to the earlier instance, yet, Miina parses it into highlighted subphases: She now pushes her shoulders back in an arc-shaped trajectory, resulting in a more extensive opening of the chest (Figure 8). When reaching the back-position, she again slightly pushes her shoulders with her hands, then throws her arms down on her sides as if forcefully swiping something away, and finally lets her arms swing back and forth (l. 22, Figures 9-10). She steps from crossed legs into a parallel position (l. 21, see Figure 5 vs. Figures 6-10), taking a more consistent "open" posture.
Like in Excerpt. (1.1), Miina uses here a 'when-then' format. The first part sets the kinetic circumstances (ku avaa 'when ø opens', l. 21), followed by an unspecified formulation of the manner of walking (nii sit sä jäät niinku kävelee sillee et 'so then you remain like walking in such way PTC', l. 21-22). However, the emphasized backward stroke in (Excerpt. 1.2) starts earlier: the preparation phase occurs already with naming the kinetic context (after ku avaa 'when ø opens', l. 21). In addition, Miina produces (l. 22) new movements during the 0.5 s pause between the verb jäät niinku 'stay like' and its infinitive complement kävelee, 'walking': she releases her arms and swings her palms 5 . The highlighted gesturing and its timing in relation to the unfolding descriptive turn emphasize the display of the effect of opening and serve to repair Maria's understanding of Miina's previous comment. The turn to the bodily source of the experience is again displayed by Miina gazing at middle distance or diagonally away, from the moment of announcing a difference (erilaista, l. 21) until ending the movement trajectory with swinging arms. Together with the demonstrative sillee 'such way', the utterance-final et once more appeals to the recipient's interpretation.
Miina's description of the opening also becomes retrospectively clarified in the continuation of her turn, when she describes the effects of the closing task as being in contrast with the opening.

3
Displaying Inner Experience Through Language and Body in…

Excerpt 1.2 continued
Miina does not make the contrast explicit but shifts to talk about closing (l. 23) by using the modal deictic tällee 'this way' pointing at her simultaneously begun movement (Streeck, 2002): she draws her shoulders up to her ears and tenses shoulders and arms (see Figures 11-13). Tension and the cupped hands (l. 23, 26) contrast with the effortless quality of the earlier loose waving of the arms (l. 22, Figures 9-10). Overlapping Miina's talk, Maria still displays her understanding of Miina's prior description of opening by saying #joo# 'yes' (l. 24). Maria straightens her back and sweeps her arms across her chest to the sides (Figures 12-13). This coinciding with her slightly strained voice  gives the impression of bodily resonating with Miina's report of the sensorial effect of the opening.
During the overlap with Maria's joo, Miina cuts off her turn and releases her posture (l. 25). When resuming the verbal turn from tuntuu 'feels' (inserting the hedge vähä niinku, 'a bit like'), Miina lifts her shoulders again, now reaching a more contracted shape (Figure 14) than on the first time ( Figure 13). Even if occasioned by the overlapping talk, the resumption and redesign of the turn enable Miina to secure the visibility of her bodily action and emphasize features of the 'closing' effect by performing them more intensively.
Co-occurring with the verb tuntuu, 'feel', the contracted and tensed body shapes display Miina's sensation well before she arrives at the explicit description of the experienced quality with the adjective jäykkä, 'stiff' (l. 26). When Miina arrives at this categorization, she is already withdrawing the hand-shape and returns her gaze at Maria, suggesting that the description is ripe for the instructor's uptake. 6 Miina comes to produce the label 'stiff' only to contrast her description with the opening task. For the opening, the main target of reflection, she did not specify the experienced quality verbally. The sensorial experience is rather conveyed as "ineffable" (see Gülich, 2005;Levinson & Majid, 2014). Its description relies centrally on embodied resources to stage a bodily and sensorial event for the recipients, which allows them some intercorporeal access to the qualia of the experience.
Miina's depiction of the effortless movement (opening) and the tensed body (closing) co-occur, again, with gazing away from Maria and using the zero person construction, the non-specific person reference that invites recipients to identify with the experience. It could be translated here as 'one' (l. 21-22 'when one opens'; l. 26 'one is'). 7 It works together with gaze aversion during bodily action to portray the sensorimotor qualities as experienced by an individual sensing body, yet intersubjectively available to others. Miina's gaze to middle distance (or diagonally away) is coupled with projection of an element to unpack a propositional attitude (l. 21-22 'it is like different so that', Excerpt 1.2) or a feeling (l. 23-26 'it fee-a bit like feels that', Excerpt 1.2 continued); the 1 3 Displaying Inner Experience Through Language and Body in… ongoing sensory exploration is part of fulfilling these projections. Returning to gaze at the instructor towards the end of the descriptions achieves a shift from sensing to inviting the recipient's uptake of the described experience.
In sum, in both comments on the opening exercise (Excerpts 1.1 and 1.2), Miina states that there is a feeling and describes the context in which it can be felt, but does not identify the sensorial quality beyond using pro-forms and approximating expressions ('in a way/sort of', etc.). The experienced sensorimotor quality is, instead, given off by her embodied conduct. Moreover, her redoing in Excerpt 1.2 is designed to emphasize the key features of movement and posture in response to Maria's misunderstanding.
The combination of multiple resources displays "doing being" sensing. It casts the reflection of the experience as located within the body, reactualized here and now. Although the experience is not straightforwardly describable, its reflection is designed as recognizable for others who may have had similar experiences based on participating in the same exercises. The descriptions emerge incrementally: the bodily action is coordinated with complex, recursive syntactic structures, during which some embodied searching or exploring occurs. Gesturing is not strictly linked to the ongoing formulation of the description in terms of specifying "meanings" or lexical affiliates. Instead, elements of the movement trajectory (e.g., the stroke or preparation phase of parsed, upgraded redoings) occur at precise moments of projection for a substantial turn element, thus giving off bodily displays of the experienced quality to enrich and guide the interpretation of the unfolding, unspecifying verbal turn. Exploratory action can also result in verbal identification of a quality, as seems to be the case with Miina's categorization 'stiff'. Yet, it was produced only to clarify the initial description of the opening through a contrast to the sensation in the closing. Miina's experience of the opening exercise is methodically conveyed through complex syntactic structures that enable, through projections, to combine and coordinate unspecified lexical content and embodied conduct as a display of sensorial, "ineffable" qualities.

Second Case: Tiia
We now analyze the comment by another participant (Tiia) concerning a later exercise. In Excerpt 2, Tiia describes her experience of an exercise that included symbolic imagery, such as imagining a crown or a marshmallow hanging above one's head while walking. Maria asks for comments (kommentteja haluaisin kuulla 'comments I'd like to hear'), and Tiia provides hers after two other participants (not shown).

Displaying Inner Experience Through Language and Body in…
Scheme 2 Multimodally unpacked projections in Tiia's turn Tiia starts her response by evaluating the current exercise as 'a lot easier' (paljo helpompaa l. 23) than a prior one, in which they spoke lines of the performance while making throwing movements (l. 24). The gist of her upcoming comment is that the current exercise is more holistic and effective than the throwing exercise, making it easier to speak the lines.
Like Miina, Tiia expresses her propositional attitude towards the task (directly evaluating it), and unpacks it multimodally (Scheme 2). Tiia starts to respond by using a comparative structure (l. 23-24), initiated with a place-holding expletive subject (se o [-] helpompaa ku… 'it is [-] easier than…'). The object compared to ('than to only…throw') is established multimodally, including swinging the arm. Next, the comparative construction projects unpacking the object of comparison, i.e., what was easier. Tiia expresses it using an infinitive subject (ottaa…, '(to) take…' l. 25), preceded by related embodied conduct. During the niinku 'like' and the 0.4 s pause at the end of l. 24, Tiia gazes away with her eyes almost closed, creating a space of sensing (Mondada, 2018b). She softly rolls her shoulders from front to back (l. 24, Figures 15-16), followed by a small bouncy movement (l. 25) echoing the bigger shoulder roll. The shoulder rolling first develops into straightening her posture, and then dissolves through smaller, repetitive, bouncy movements (l. 25-26). Together with Tiia's facial expression they create an impression of enjoyment and come across as an exploration of a here-and-now developing sensation. Thus, even before Tiia establishes the object of comparison verbally, the experienced effect of the "more holistic" exercise is made available to others by bodily conduct that conveys much more than a gloss of the verbal expression for 'taking a role completely'.
Tiia's response involves a gradual shift to accounting for her view. She reaches the utmost straightening of the posture (Fig. 17) at the onset of the word kokonaan, 'completely' (l. 25). Labeling the exercise as 'complete' role-taking makes a claim about why it works, and thus accounts for the claimed difference to merely doing the throwing movement (see l. 24).
After this, Tiia specifies that the effect of taking the role applies to speech, using (like Miina) an interrogative proadverb (miten sen sanoo 'how ø says it', l. 26) that refers to a manner but does not identify it. The specification has been projected by the demonstrative siihen (DEM.ILL) and the complementizer et (l. 25) (here translated as 'take a role… regarding') (see Keevallik, 2011 on demonstratives as projective devices in Estonian). Meanwhile, Tiia turns her gaze again to middle distance, and the embodied trajectory dissolves through a small final weight shift.
Tiia's embodied actions locate the body as the source of the experience, while the design of her talk (the infinitives and the zero person) does not designate who makes or experiences the movement, e.g., who feels the degree of effort in the actions. The effect is something anybody in those circumstances could experience. The bodily conduct is coordinated with verbal projections to identify the object of comparison (what was 'easier') and where/how the effect applies ('how one says it'). Through this, Tiia's claim about the effects becomes enriched by displays of turning to sensory experience and a dynamic body posture which can be characterized as indexing fluency, enjoyment and preparedness.
In the latter part of Tiia's commentary, from l. 27, she shifts to explain and account for her reported experience. She now analyzes the effects of the task in the past tense and categorizes the feeling (olo, l. 29 8 ) as 'more self-confident'. During this talk Tiia gazes mostly at Maria, who takes up Tiia's reflection by nodding. From here on, Tiia describes her experience by quoting her prior thoughts (i.e., verbally depicting, Clark, 2016) (l. 30-31), contextualized as a quotation by prosody, facial expressions, and a deictic shift to the enacted moment (mä 1sg; tää lause 'this sentence'). Using the first person instead of the zero person now highlights herself as the speaker and experiencing subject. With these resources, the effect of the exercise is verbally explicated and accounted for to the instructor, rather than made available for recipients' visual and intercorporeal recognition as in the beginning of the comment.
Both Miina's and Tiia's repetitive embodied conduct depicts a sensation that develops and endures when they are under the influence of the exercise (as in 'take completely', 'stay/remain'). The bodily enactment is coordinated with projections to elaborate the verbal turn; yet they are not tied to a precise lexical affiliate. Both speakers arrive at verbally specified categorizations only as clarifications or accounts of the earlier displays. These more reifying categorizations thus result from the re-exploration and intersubjective display of an experience, which in itself succeeds without such verbal elements.

Discussion
This study has analyzed how participants display their bodily, sensorial "inner" experience of exercises from the Chekhovian acting technique in the context of community theater rehearsals to the instructor and fellow participants.
When giving their comments, the participants engage in a social activity that ultimately orients to a joint, artistic goal. The effects of the inner body technique are somewhat different for everyone, and the participants are the "primary party entitled to assess these effects" on their own bodies (Lindström & Mondada, 2009: 304). Yet, the tools of the artistic work should be interpreted in sufficiently similar ways by all group members to be able to prepare and carry out the performance (Visakko, 2020b). The participants' reflections give evidence of their level of understanding and alignment with the goals of the exercise. They demonstrate skills of self-reflection, drawing on authentic sensorimotor experiences; they also give these experiences meanings that demonstrate their increasing ability to sustain a character (e.g., through awareness of different manners of moving and speaking). The main recipient of these displays is the instructor Maria, who has the knowledge and authority to assess the application of the technique. She coordinates the discussion and displays bodily and verbally her understanding of Miina's and Tiia's actions. The fellow participants show little immediate uptake of each other's comments, but they can acknowledge each other's experiences, for example, by designing their own comments in relation to what others have reported.
We have focused on how verbal and bodily conduct mutually elaborate each other in the participants' reflections. In coordinating verbal and embodied conduct, participants make use of the multi-unit structure of their turns, bodily enactment of movements and postures, gaze-shifts and zero person reference. The analyzed cases thereby exemplify relations between complex syntax, sensoriality and embodiment in interaction (Maschler, Pekarek Doehler, Lindström & Keevallik, 2020).
In particular, emerging trajectories of movement and posture match projections of grammar and action for unpacking elements within the description. The projections have been created by expressions of a propositional attitude, by syntactic design and other projective elements including turn-final particles, demonstrative pronouns and other pro-forms.
There are various relationships between mutual interpretive enrichment and constraint: • Verbal projections of propositional attitudes (e.g., 'it does feel there') constrain the interpretation of the following bodily depiction and the upcoming vague descriptive verbal content (e.g., 'remain like walking in such way so') as being a description of subjective experience and locating them with respect to certain (phases of the) exercises. • The bodily conduct, whose meaning may be rather obscure in itself (waiving elbows back and forth or bouncy movements), becomes contextualized by the talk, linking it, e.g., to particular phases of the described exercise or differences between exercises. Here the retrospective dimension is important, as later descriptions can elucidate the meaning of earlier gestures by explication but also by contrast. • The verbal descriptions, which are at many points incomplete, aborted, inference-rich, unspecifying, and abstract, become enriched and specified by bodily enactment, both referentially and concerning sensory and affective meanings. Earlier onset and emphasized delivery of a depiction on a second attempt can serve to clarify a prior description. • Participants make use of projection to design multimodal descriptive actions: projections create slots where identification or description of a certain element is to be expected. Instead of directly describing it, participants make it available and recognizable to others through visibly engaging in sensorially inference-rich conduct (and without producing direct gesture + lexical affiliate pairs).
The reportings of "inner body" experiences investigated in this study concern observable events-the exercises-that all people present have shared and co-produced in embodied ways. They presuppose that some bodily experience of them is available to the recipients as a background for understanding the descriptions. The reports are intimately tied to their situated context and its pragmatic relevancies, i.e., practicing the acting technique. In this context, a body is enacted that is in the process of sensing, being affected by the movement task at the moment of describing it. The inward-orientedness to the current, re-actualized bodily experience (see Arano, 2020) is displayed by gaze withdrawal, self-touch and -manipulation, and repetitive, stepwise composition of movement and body posture.
Miina and Tiia do not bodily depict a "quale" (Levinson & Majid, 2014: 408) in itself, but they make observable its bodily source. They do not repeat the exercise itself; rather, they make (the exploration of) its sensational effect visible bodily. They achieve this by a selected, highlighted and modified use of elements from the actual exercises (e.g., a posture or a direction of movement). Their repetitive gesturing involves a stepwise composition of a larger trajectory, such as bouncy movements in gradually taking an upright posture. It can also involve active stimulation of the effect of the body movement by the actor herself (e.g., pushing back own shoulders with the hands, producing the effect of "being moved").
The cases demonstrate the self-exploratory and tentative character of emerging descriptions of bodily experience. Especially in the beginning of the comments, the experience seems to be ineffable in many respects, its qualities becoming only gradually elaborated with the aid of following bodily enactment (Streeck, 1995). While the reliance on bodily displays portrays the experience as subjective, sensory, and to some extent ineffable, although recognizable, the resulting unspecified verbal categorization by (pro)nominal and (pro)adjectival/adverbial structures is an incrementally achieved, bodily supported verbal abstraction. Participants treat it as sufficient, as when Miina and Tiia return, after a moment of sensory exploration, to gaze at the instructor: this is a sign of arriving at a sufficiently recognizable portrayal of the experience (see Thompson & Suzuki, 2014). Returning to gaze at the instructor when a turn is still unfolding is coupled with delivering categorizations and accounts-these contrast with the pattern of gazing away during the moments of doing-being-sensing.
The use of the zero person in the comments portrays a (presupposedly) intersubjectively recognizable experiential response to the technique they are practicing. The response is reproducible and to some extent generalizable. At the same time, the zero person highlights the experience instead of the view of the individual speaking subject. This goes together with how the embodied conduct and gazing away invite the recipients to relate to the sensing body. In the same vein, the fragmented and unspecific language appears to serve the sharing of the sensory experience by leaving it to some extent indeterminate, yet relatable by inviting the recipients to fill it in with their own resonating experience.
Transcription symbols for multimodal conduct. * * Trajectories of embodied action are delimited between two identical symbols.
*--> The action described continues across subsequent lines -->* until the same symbol is reached.

>>
The action described begins before the excerpt's beginning.
-->> The action described continues after the excerpt's end.