1 Introduction

In the research on children’s storytelling, two issues that are relevant to the study to be reported have provoked disagreement. The first is the question about when children start to initiate stories. A cursory search of the literature places the emergence of these practices at different ages—anywhere between 19 and 30 months (Engel, 1995; Filipi, 2017a; Heller, 2019; Miller & Sperry, 1988). The disparity can be explained both by the different lenses used to analyse the data as well as definitions about what constitutes storytelling or narratives. Typically studies that report the emergence of the practice as occurring earlier are concerned with the interactional properties of storytelling using the microanalytic methods of conversation analysis (CA) (e.g. Hutchby & Wooffitt, 2008; Schegloff, 1991). Increasingly these methods include the study of embodied action, crucial to the study of very young children’s interactions as it permits a different level of understanding about how very young children participate in interaction (Filipi, 2009; Heller, 2019). The outcome of applying these analytic methods is likely to show that very young children initiate action in talk-in-interaction rather than merely respond to a parent’s initiation. It is exactly this approach that Heller (2019) adopted to examine early storytelling practices which led to her findings that a child as young as 19 months was initiating storytelling.

The second more controversial issue is to consider what constitutes storytelling (on this controversy, see Nicolopoulou & Ilgaz, 2013). Research mainly from within developmental psychology has long held the view that pretend play is a form of enacted storytelling (Nicolopoulou, 2016; Snow et al., 2001). According to Nicolopoulou (2016), pretend play involves “narrative activity” which has led her to conclude that it complements storytelling modes so that they eventually converge developmentally.

The study to be reported in this chapter holds that pretend play is an early form of storytelling. It aims to show the ways in which episodes of pretend play are recognisable as the earliest forms of child-initiated storytelling, which I argue is an important finding. This will be done through analysis of two episodes of pretend play between Rosie, aged 15 months and 3, interacting with her mother. By focussing on the two ages through a microanalytic, multimodal lens, rich findings will emerge to show the changes in how the interactions are organised while continuing to be a co-produced achievement of mother and child.

The chapter begins with an overview of prior research followed by the research design. Next the analysis and discussion of the two selected episodes will be presented. The chapter ends with a general discussion about the uniqueness of the findings and concludes with a set of suggestions for the ways in which findings might be translated into practical applications for families.

2 Background

2.1 Storytelling in Early Childhood and Their Purposes

Through the practice of storytelling in the home, very young children become socialised as members of society. The practice begins through the simple actions of parents reading to their children, engaging with them in pretend play and inviting them to share accounts of the events in their lives. These storytelling activities provide a foundation for the practice to continue when children go to school.

Storytelling fulfils a variety of “social projects” (Mandelbaum, 2012). It can satisfy affective purposes where children enjoy listening to and/or (re)telling stories heard (Bateman, 2020; Cekaite & Björk-Willén, 2018). It can provide children with an opportunity to imagine the experiences of others as they take other people’s perspectives through playing a variety of social roles (Heller, 2019). Participation in storytelling, in turn, creates opportunities to build identity (Goodwin & Kyratzis, 2011; Theobald, 2019). Storytelling can also play an important role in children’s mental health when they are invited to share traumatic events in everyday conversations (Bateman & Danby, 2013; Bateman et al., 2015). Through each of these forms of telling, children develop both interactional and linguistic competence, which are fundamental for literacy and success in school (Bateman, 2018; Snow et al., 2001). They create opportunities and experience for children to build confidence as they engage socially with a range of others including family members (Blum-Kulka, 1990; Busch et al., current volume; Filipi, 2017b; McCabe & Peterson, 1991; Waring, current volume), peers (Bateman, current volume; Theobald, 2016; Theobald & Reynolds, 2015) and teachers (Bateman & Carr, 2017; Theobald, 2019; Theobald et al., current volume).

Storytelling in childhood takes many forms. The most obvious is storybook reading which occurs both through hard-book and digital formats. An important feature of storybook reading is the discussion that takes place. This starts with naming pictures (Chiong & DeLoache, 2012; Fletcher & Reese, 2005; Potter & Haynes, 2000) and eventually leads to richer discussions not only of the stories read but also of second stories (Sacks, 1992) that are generated as the participants draw comparisons with events in their own lives (Filipi, 2017a, 2019; Reese, 1995; Takada & Kawashima, 2019). Another form of storytelling is found in relating personal news through updates (Searles, 2019) or through recounts of shared events (Burdelski, 2019; Farrant & Reese, 2000; Morita, 2019; Takagi, 2019; Waring, current volume). Also pervasive are spontaneous and planned imaginative storytellings which children share with peers, teachers and parents (Bateman, 2020; Theobald, 2019; Waring, current volume). These can entail enactment where stories are brought to life in pretend play (Bateman, 2018, current volume; Nicolopoulou, 2016; Snow, et al., 2001).

2.2 Pretend Play

The importance of narrative or storytelling to the development of literacy and future success in school (Snow & Tabors, 1993) has provided a strong incentive for the study of children’s early pretend play. Kavanaugh (2006) and Ilgaz and Aksu-Koç (2005) maintain that there are synergies between the structure of pretend play and children’s storytelling and comprehension such that there is a striking relationship between the two. This has prompted Nicolopoulou (2005, 2016) to suggest that pretend play sits on a developmental continuum where the discursive properties of storytelling complement the enacted narratives or “scenarios” in pretend play. Nicolopoulou (2011, 2016) also brings attention to the importance of the social context while at the same time adopting the view (based on theorists such as Bruner) that narratives entail cognitive processes in the construction of reality. Taking a developmental perspective and a position that pretend play contributes to children’s development, Nicolopoulou (2005, 2016) and Nicolopoulou and Ilgaz (2013) maintain that providing experiences in pretend play contributes to children’s development as a range of cognitive processes are implicated. These include memory, the ability to take the perspectives of others (associated with theory of mind (Lillard & Kavanaugh, 2014)), skills in preparing for play and commenting on action (Boyd, 2009), skills in developing the story through a plot and characters that involve understanding that negotiation and adjustments to storylines may need to be made in concert with others, and opportunities to explore identity. According to Nicolopoulou (2016), these are skills that develop over time.

As far as I’m aware, from a CA perspective, there is only one study by Bateman (2018) that has explicitly examined pretend play as a storytelling event rather than from the perspective of children’s moral and social orders in resolving disputes (e.g. Cobb-Moore, 2012) or in negotiating roles for themselves (Björk-Willén, 2012). Bateman’s (2018) study, located in preschool and in the first year of school, was concerned with showing how children co-created spontaneous stories through ventriloquism (or voice projection, Harris, 2000) and embodied actions. Through these resources, the children created stories that were recognisable and coherent, and that were sensitive to progressivity and changes in plot direction. The co-created, enacted stories acted as harbingers of written storytelling.

In synthesising the relevance of the above reviewed studies, this chapter will build on prior research to show how pretend play begins through embodied interaction with a parent who supports the child to direct and develop the enacted story. Interactional competence over time (as described for instance in Filipi, 2019) is also focussed on in comparing the episodes at the two selected ages.

3 Data and Method

Two episodes depicting Rosie interacting with her mother have been selected for this chapter. They come from a large data set collected fortnightly in Australia for 30-min sessions over a  27 month period involving four child–parent dyads. The researcher was sometimes present during the recordings as was the case in the second episode selected for analysis. Ethical procedures were followed for data collection, recruitment and with respect to consent.

Collection of data commenced when Rosie was 9 months old and continued until she was aged two. A further one-hour sample was collected when Rosie was aged 3. In selecting the episodes for analysis, I sought pretend play sequences where the play appeared to be clearly initiated by Rosie rather than the mother, and where the actual story enactment involved co-production rather than imaginary self-play. These started to appear when Rosie was producing one-word verbal utterances at the age of 15 months but was participating in rich interactions largely through embodied actions. (See Filipi, 2009, for extended analysis and discussion of the quality of the interactions in general at this age for the same data set.) For the purposes of comparison in order to capture change, an episode from the final session when Rosie was aged three was also sought. The episodes thus provide a window on the social and sequence structural qualities of early enacted storytelling as displayed through the collaborative actions of parent and child as they unfold turn-by-turn, where the enacted story is clearly begun and developed even if not completed. Both episodes occurred in the home.

3.1 Analytical Methods

Analysis of the data follows the methods of CA as outlined in Chap. 1 of this volume where interaction is understood as a set of multimodal actions. These actions are organised through turns that are designed with reference to speakers’ interpretation of the prior turn(s) and their epistemic positions (Heritage, 2012). In this way, talk is built to establish, display and achieve intersubjectivity. It also affords an opportunity to examine how membership to a group or category is achieved (Evaldsson, 2007; Kyratzis & Goodwin, 2017; Pomerantz & Mandelbaum, 2005). The latter is important in exploring children’s identity shifts in storytelling.

The turns themselves are organised in sequences (Schegloff, 2007), and sequences are organised into larger episodes (Schegloff, 2010) where a single activity (in this case, enacted storytelling) is constructed through stretches of talk and sustained through multiunit turns across sequences (Heritage & Sorjonen, 1994). The analysis is conducted through a microanalytic lens launched to capture both verbal and nonverbal minutiae of talk, including embodiment and prosody and not just words. With respect to the analysis of children’s interactions, attention to both what is said and to how something is said and done through the body and voice uncovers both how interactionally competent very young children can be, and the complexity of the interactional resources they deploy to engage in interaction (as reported for example in Filipi, 2009).

Transcription notations used in the extracts are consistent with each of the chapters in this volume that follow the Jeffersonian system (Jefferson, 2004). The following additions from Filipi (2007) are used to denote nonverbal features: --- → to indicate gaze, TU to indicate turning towards, P--- → to indicate pointing to, and the curly bracket { to indicate the onset of a nonverbal action.

4 Analysis and Discussion

4.1 Pretend Play at 15 Months

Extracts 1 and 2 from the same episode provide an analysis of an early form of storytelling that occurs through enactment and therefore is relevant at an age when the child is largely achieving successful interactions with those around her through multimodal resources (Filipi, 2009). This episode of early pretend play is typical at this early age when the child is producing single word utterances in turns that are filled with physical actions and vocalisations. Socially, these early pretend play episodes provide opportunities for the child to draw on and replicate or play out her daily experience in concert with, in Vygotskian terms, more competent others (see Vygotsky, 1978).

Like the recounting of daily events, in this episode the pretend play draws on one of Rosie’s routine experiences, that of being taken out for daily walks in her pram by her parents; so she is reimagining and re-enacting the experience by taking on the role of the parent or the responsible adult.

As the episode is very long, the transcript will be broken into two parts. Extract 1 is concerned with preparation for the pretend play as well as initiating the pretend walk. The mother has just brought a doll and stuffed toys into the room. Rosie abandons the blocks she has been playing with and picks up Teddy and Clown. She moves towards the pram. Important to the analysis is the fact that Rosie and her mother move in and out of play as they manage the scene through setting up the objects that are important in creating the story. To be noted is that throughout the episode, play is interrupted. Interruptions of this nature are part of these early episodes and offer displays of the ways in which the mother supports the child in play but also the means by which the child herself recruits the mother’s help.

Extract 1:

Getting ready for the pretend play and for the walk

figure afigure a

Rosie initiates the enacted story through the actions of picking up Teddy and Clown, and moving towards the pram. The mother then launches a series of action descriptions through a question format which propels the story (lines 1, 3, 5, 8). It ends with an assessment in line 10 oh lucky teddy. Each of her questions work as format tying (Goodwin, 1990). They tie one turn to the next in sequentially relevant or expected ways to achieve cohesion; in this way they provide a verbal unfolding of the enacted story. The questions are therefore sensitive to Rosie’s actions (of walking towards the pram with Teddy and Clown and then focussing only on Teddy by putting him in the pram in preparation for a walk), and they are designed to fit verbally as appropriate actions so that they are sequentially implicative (Schegloff & Sacks, 1974). Rosie also produces a series of responses through format tying. She does this through vocalisations (lines 7 and 9), actions (placing Teddy in the pram) and gestures (pointing and nodding). Her participation in the enacted story is therefore facilitated and made relevant through the mother’s questions. Through these actions the plot for the story (a walk with the pram) and the characters (Rosie, Teddy and Clown) are collaboratively established.

After an intervening series of actions involving the stuffed toys and frustrations as Rosie works nonverbally to recruit her mother’s assistance (not shown in the transcript), the enacted story continues through further preparation for the walk. The character Baby is now placed into the pram alongside Clown, while Teddy is removed. Although this is the only action in this section of the episode, it takes place over several turns as the mother works to support Rosie to complete the action but also to engage in producing (unambiguous) verbal responses. This is evident in the mother’s pursuing actions through her yes produced with upward pitch that acts as a confirmation request in line 68. This action follows Rosie’s nonverbal answer to her mother’s yes/no question in line 63. Although it could be deemed to be an appropriate answer, the mother’s pursuit in line 68 is launched to invite a verbal response. Filipi (2009, 2013) maintains that these pursuing actions begin around this age. Parents start with “gentle” pursuing actions just as children start to produce one-word utterances, and they become more persistent when children start to produce a verbal yes. In this extract, we can see that Rosie does actually produce a verbal yes response in line 77 which co-occurs with a head nod. On receipt of this response, we note that the mother produces a third turn closing ok. This action works in two ways. It closes down her linguistic pursuit through a display that it was successful in achieving the expected verbal yes, and it also resolves the action of placing Baby in the pram so that the enacted story can progress.

Extract 2:

Going for the walk

figure b

The next stage of the enacted story involves starting the walk. This is announced by the mother in lines 80–82 (…they look ready for a walk). The walk entails leaving the confines of the playroom to go down the corridor. However, Rosie needs to deal with the closed door before she can begin. Although the mother can see that Rosie is at the door, she does not acknowledge that her yelling is a request for assistance until Rosie turns and looks at her in line 99. The mother is treating this action as a recruitment of her participation, specifically her help, as her subsequent actions indicate: her confirmation check in line 98—do you want to go through the door—which Rosie confirms nonverbally, her subsequent question launched to request her return to the room and finally her assistance by opening the door for her. Soliciting the assistance of others in the accomplishment of everyday tasks is a normal social action (Kendrick & Drew, 2016), and there are both explicit and tacit ways in which speakers do this. They include open requests and offers of help. Pfeiffer and Anna (2021) describe the actions of children aged from 2;8 to 3;6 in achieving this work. They report the use of the “oh + x” format accompanied by gaze as a form of participant recruitment which reports trouble. Rosie’s nonverbal actions accompanied by her cries of frustration (in line 96, and in lines 11–45 and 51–58 not shown in the transcript) could well be an early form (i.e. before the onset of words) of the organisation of sequences where the child reports trouble and recruits assistance.

The solicitation and granting of help remove the final obstacle to the enacted walk which now gets under way. Rosie leaves the room, agrees to come in through her nod and vocalisation in line 103 and eventually returns. At this point (lines 112–116), the mother who has up until this point positioned herself as observer and supporter of action outside the story, enters the story as a character through a greeting sequence and a welcoming of the visitors.

As noted, pretend play has a narrative structure that is related to storytelling (Ilgaz & Aksu-Koç, 2005; Kavanaugh, 2006). Nicolopoulou (2016) also maintains that in the early stages, pretend play has more basic story elements, particularly regarding plots. This is evidenced in the above extract where the plot involves a very simple going for a walk and (brief) visit. Elements familiar to a narrative are present. The action of (co-)creating the plot (going for a walk and visiting), deciding on the characters (Rosie/adult, Clown, Baby, the host), recruiting physical objects (the toys, the pram), and negotiating physical space and boundaries (the door, the corridor outside the playroom), all of which are essential elements of the pretend play. As well, an important part of the pretend play is its preparation or in Harris’s (2000) words, the “stage management”, in which recruitment of assistance is pivotal. Finally, although there is a story beginning, and a story development, the ending is abandoned. This is so often the case in these early interactions where activities are vulnerable to being hijacked by other concerns or distractions (Filipi, 2009).

However, the social aspect of the storytelling is of principal concern here. Two features in this regard stand out: the first is the collaborative actions in developing the enacted story. Noteworthy here are the ways in which the mother both scaffolds Rosie’s participation and also takes the lead from her so that Rosie is genuinely involved in shaping the enacted story; first by initiating it through her actions (which the mother interprets) and subsequently by contributing to its development by soliciting assistance with the physical actions necessary to its progress. The second feature relates to the identities that emerge in the episode. Drawing on categorisation analysis (Fitzgerald & Housley, 2015; Hester & Eglin, 1997), we can see the ways in which Rosie’s play identity shifts as she co-constructs the pretend play. She uses her experience of participating in the real world as herself, the child, to become the parent responsible for taking the “babies” for a walk in the pretend play. However, these identities are constantly shifting from the “role of parent” in the pretend play to herself as she engages in the preparation and removal of obstacles. The mother retains her own identity throughout to support the story creation as an outsider where she observes, interprets and facilitates participation, and Rosie orients to her in this role. However, at the very end, the mother enters the play as the person being visited. The result is an interaction rich in both imaginative play and in the management of its construction.

Turning to interaction, a number of interactional properties are visible. As stated, in creating the story, Rosie recruits her mother’s participation, which Drew and Kendrick (2018) and Kendrick and Drew (2016) claim is a resource used by speakers to elicit the assistance of others to accomplish a task. Rosie’s action and the mother’s response here provide a display of collaboration. With respect to the mother, both of her physical and discursive actions are clearly launched to support Rosie in participating. Discursively, she does this through a series of questions launched to co-create the story step-by-step. In line 1, she gives voice to the initiation of the story through a yes/no question (shall we put clown in the pram?) which is heavily tilted towards agreement (Bolden, 2016), received both verbally (yes) and nonverbally (head nods) from Rosie. Agreement is thus the preferred response, but there is also other work being done here through the sequence organisation which is composed of a series of adjacency pairs. These both provide coherence in the larger activity through format tying, but also as stated above, a “gentle” pursuit by the mother of responses that come off as confirmations that her interpretations of Rosie’s actions are correct. Furthermore, there is also explicit linguistic work taking place which is displayed through her pursuit of a verbal yes where attention is drawn to the inadequacy of the nonverbal response (Filipi, 2013). Forrester (2008, p. 124) refers to this as a “sequence implicated” repair. This action leaves the well-researched tripart Initiation Response Evaluation (IRE) (Mehan, 1979) sequence (aligned with instructional or didactic work in the classroom and in interactions with young children (Filipi, 2009; Tarplee, 1996)), open or incomplete because a third turn closing—the acknowledgement rather than the evaluation—is withheld until a verbal response is received. It therefore acts to initiate linguistic work through a form of repair, an “other-initiated self-repair” where the correction comes off implicitly.

In sum, the analysis has shown the ways in which this early form of storytelling is intricately shaped and managed. The progress and development of the enacted story is interwoven with a range of matters: the preparation for the story by mobilising objects and toys, agreement on the direction of the story itself, work on language and the removal of physical obstacles to enactment, all achieved collaboratively.

4.2 Pretend Play at Age 3

In the next extract, we turn to an episode which occurred when Rosie was aged three in order to examine the ways in which pretend play has changed over time. Rosie and her mother have just read a favourite children’s story, Hairy Maclary (Dood, 1983), which is about the adventures of a group of neighbourhood dogs. One adventure involves a run-in with a tomcat, Scarface Claw. Two features are important in this episode: first the way in which Rosie makes a real-life connection with this character in the storybook by inserting Claude, their neighbour’s cat’s name, into the story; second the emergence of a second story (Sacks, 1992), an important feature of conversational storytelling where speakers make sense of and find commonalities with their experience.

As this episode is also quite long, the transcript has been broken up into two parts to facilitate the analysis. The analytic focus throughout is to track Rosie’s initiating actions and the resources she uses.

Extract 3:

Rosie initiates the enacted story (Lines 21–23 are reprinted from Filipi, 2019, p. 132)

figure c

There are in fact several initiating actions in this episode as both Rosie and her mother come in and out of play. The episode starts off with an offer by the mother to read the storybook Hairy Maclary (lines 1 and 3). The offer is not taken up, however, as Rosie instead enters the role of a cat by scratching the chair. The mother reacts by acknowledging the initiation of the pretend play in line 5 (you sound like a pussy-cat) and abandons her initial invitation. She thus orients to Rosie’s action as an initiation of the pretend play sequence. Her subsequent follow-up yes/no question—are you being a pussy-cat—is launched as a confirmation request which receives the voice projected (Harris, 2000) miaow as a response in which the mother engages as well. Retrospectively, it treats Rosie’s action as an invitation to her mother to participate in the play. In line 10, Rosie suspends the play by proffering a suggestion for the enacted story. She does this through an I know + pause + you can be role suggestion format and assigns roles to the play participants—dog to the mother and cat to herself. The mother takes on the role of the dog by whimpering. However, Rosie is expecting more from the character of the dog (line 15) when she explicitly asserts how the dog character should behave—now you can chase me. This is followed by a reason which she formulates as a fact—doggies chase cats—which is previously co-established knowledge about dog and cat behaviours acquired through the repeated reading of stories like Hairy Maclary. The mother supports the veracity of Rosie’s assertion by making reference to the events in the storybook in line 19. This touches off a discussion about the Hairy Maclary story and the local (neighbour’s) cat, Claude, through a Q/A sequence so that the play itself is suspended.

Extract 4:

Rosie resumes the enacted story

(Lines 61–68 are reprinted from Filipi, 2019, p. 132)

figure d

Once talk about the story and Claude, the local cat, is concluded (not shown in the transcript), Rosie initiates the resumption of the pretend play (line 68) by again insisting on being chased. The mother, however, tries to negotiate a different behaviour for her dog character—cuddly rather than chasey, a suggestion that Rosie rejects in line 66. Play is suspended twice more (to confine the physical parameters of the play to the living room and to comment on the roles assigned by drawing attention to the fact that Rosie (the cat) ended up chasing Mother (the dog)) before Rosie again resumes play through an imperative construction you can chase me as she returns to role. In lines 88 and 90, Rosie again insists on being chased, while the mother resists (I’m a sleepy doggy). Rosie’s growling action in line 92 provides the mother with an opportunity to retain her sleepy dog character so that we have a very clear sense of how both mother and child are each either delicately resisting the assigned character features, negotiating them or accepting them. Finally, in line 97, Rosie initiates a change in the characters by again using the I know + pause + you can be role suggestion format which is accepted by the mother so that play resumes.

The pretend play in the above episode shows how a second story develops from storybook reading (see also Filipi, 2017a and 2019 for a more elaborate discussion of second stories arising from storybook reading) that provides an important foundation in interactional competence. The play itself displays how drama around the characters of the cat and the dog is created and adapted as it unfolds in an impromptu way through embodiment and voice projection (Harris, 2000) or ventriloquism (Bateman, 2018) (the barking and miaowing). It also provides a very clear display of what Rosie knows about events in the world: that dogs chase cats but that there are exceptions. This is revealed through the talk about her local neighbourhood cat, Claude, who like the character in Hairy Maclary, is an aggressive cat. Notably this is a character (a Claude cat) Rosie rejects as she differentiates it from a pussy-cat for the pretend play even though the mother suggests that she might be the Claude cat (lines 80–84) after all (as a result of chasing her, the dog).

As well, the episode provides a display of story development as a co-produced event, which is important for socialising. This includes using the I know + pause + you can be role suggestion format in allocating roles, which works to elicit an agreed direction for the play thereby highlighting its collaborative nature from the beginning. It also provides a way of linguistically marking a change in the talk as the speakers come out of the story to talk about the enacted story structure or further development. Further displays of the enacted story as a co-produced event include: working with resistance and with negotiation through assertions about facts and observations about life; knowing when to resume the play, notably, through understanding or judging when a prior sequence has come to an end; and ensuring that both speakers are in the same physical and embodied space (by facing each other) for returning to role. In all these features, the episode clearly points to interactional competence that includes tightly organised embodied and verbal resources.

5 Discussion

This study was concerned with examining the interactional properties of enacted storytelling in two episodes of pretend play between a mother and her child, Rosie, at two ages: 15 months and 3. The analysis pointed to the ways in which Rosie initiated pretend play at 15 months to enact her everyday experience of going on a walk in the pram. The sharing of everyday experience is a recognisable feature of tellings in conversation. The study’s finding is thus significant because it reveals that pretend play is the earliest form of child-initiated storytelling but also that it starts very early, at an age previously not reported in the research. This strongly suggests that the origin of storytelling is in pretend play from which other forms of storytelling evolve. In making these claims, I attempt in part to resolve the issues cited in the introduction pertaining to what constitutes storytelling and to the age when children begin to initiate storytelling.

The focus across the two ages provided an opportunity to investigate the interactional practices and resources used at the two moments in time (see also Waring, current volume, on changes in children’s storytelling), helpful in bringing to light the ways in which pretend play changes over a period of time as verbal resources replace the largely embodied ones. In the earlier episode at 15 months, the pretend play was achieved with a great deal of support from the mother as she interpreted, gave voice to the events of the story and was recruited to provide assistance in the preparation or stage management. It uncovered a complexity in the various aspects that went into conjoint story enactment through Rosie’s and the mother’s actions as they prepared for the play, interrupted it in order to organise the characters and objects and dealt with physical obstacles. In the episode at age 3, Rosie was initiating the pretend play, assigning roles and suggesting story development with which either the mother agreed or in which she negotiated alternative directions. In initiating the pretend play, Rosie drew on her lived experiences (going for a walk in the pram) and her knowledge of the world (specifically dog and cat behaviours), co-produced in her everyday conversations (Bateman & Church, 2017) and storybook reading, to create recognisable stories. Through the actions in the pretend play episodes, which are associated with early forms of literacy and storytelling (Bateman, 2018; Nicolopoulou, 2016), we are afforded displays about how language acquisition is promoted, how turn-taking and the staking of claims about knowledge are encouraged, and how children make sense of their experience, which is shared and connected to that of others.

The development of the enacted stories in the extracts just analysed is largely achieved in embodied ways at age 15 months so that objects and toys are prominent as life is breathed into them in object personification (Giménez-Dasí et al., 2016). Embodiment in the stage management of the story is also evident. It involves the preparation of the objects for play and their placement as well as defining the physical space and imposing the boundaries for the play (which the mother controlled in both episodes but which needed Rosie’s agreement). Clearly (and expectedly), it is in the verbal resources where the greatest change has occurred at the age of 3, and in the delicate and fluid movements between the story, other matters and story resumption.

6 Conclusion and Recommendations

Through storytelling people make sense of their lived experiences as they share stories with others (Bruner, 1987, 1991, 1997; Ochs & Capps, 2001; Stokoe & Edwards, 2006). Pretend play provides a sense-making practice where children draw on their lived, everyday encounters to make sense of it through play as they take on roles that provide a rich source of identity construction. In developmental psychology, the development of theory of mind (ToM), which starts between the ages of 12 months (Liszkowski et al., 2004, 2009) and 2 (Luchkina et al., 2018), is highlighted as being important to young children’s ability to understand the knowledge states of others. It is defined as being crucial for social interaction and claimed to be developed through pretend play which requires taking the perspectives of others (Lillard & Kavanaugh, 2014). ToM has been critiqued in studies that are grounded in CA (e.g. Filipi, 2009; Jones & Zimmermann, 2003; Kidwell & Zimmermann, 2007) because of the cognitivist and positivist approaches used that are based on experiments to explain the development of mind. The findings and attention to the social actions of the child as she interacts with her parent are important and draw attention to the need to comprehend what the child knows/understands but by attending to naturally occurring practices displayed in interaction rather than through controlled experiments.

A microanalytic CA lens on pretend play has thus brought attention to how we can understand pretend play as an instance of naturally occurring storytelling. It has provided an opportunity to observe the displays of knowledge of a very young child as she took on roles that she had experienced and to act in those roles in scaffolded concert with her mother. Children’s experience of pretend play starts in the home in interactions with family that create important milestones and provide the solid, social foundation on which to build as they prepare to go out into the world to form friendships with peers. Pretend play takes on a central role in building friendships as they participate in and negotiate their own roles and those of others (Cobb-Moore, 2012) just as Rosie has done at the age of 15 months and 3 with her mother. Children also need to steer their alliances with peers by negotiating the rules of play (Karlsson et al., 2017), and they need to be able to steer courses of action in order to influence their uptake (Theobald, 2013). These competencies have been uncovered in the episode at the age of 3 with a mainly supportive parent who only minimally resists an allocated role. They will be further refined as children engage in disputes (Danby & Theobald, 2012) with peers which constitute practices for achieving the functions just described.

In closing, the study suggests a range of possible practices, derived from the analysis, outlined below that parents might engage in to support their children’s early pretend play in the home. Parents can:

  • provide opportunities for extended talk

  • create opportunities to make sense of the stories read by talking about the ways in which the stories resonate with actual events in children’s and family members’ daily lives

  • build children’s knowledge about the world through discussions about characters and events in stories read

  • be sensitive to children’s actions that can act as possible pretend play initiations

  • set up toys in the space allocated for play that can act as triggers for pretend play

  • model pretend conversations with toys

  • provide pretend play conversation starters such as let’s pretend, I know, you can be…

  • provide experience in negotiating roles and in story development.