Understanding extended narrative sensemaking: How police officers accomplish story work

Extended narrative sensemaking consists of (a) agents generating actionable stories when faced with unexpected situations to which they need to respond in real time, and (b) emplotting actions that were undertaken and events that occurred during their response, in order to make deeper sense of them afterwards. When sensemakers revisit critical incidents in which there were involved, they join (a) and (b) through story work. In this article, through the study of stories told by police officers in relation to unexpected, impactful incidents, we show how story work is accomplished. We argue that sensemakers simultaneously enact situations, emplot events, and renew identity. Specifically, we demonstrate that police officers strive to accomplish three different things: first, show how, as engaged responders, they were involved in the ongoing enactment of an actionable story (situated agency); secondly, seek to deeper understand, after the event, what happened to them through emplotting their experiences (complexified sense); and thirdly, update their narrative identity by weaving their experience of the handling of the unexpected situation with the rest of their life story (identity renewal). Our account extends current understanding of ongoing narrative sensemaking by showing how agents construct agency, meaning, and identity at once, and how all three are part of an extended, ongoing sensemaking process.

About a decade ago, the first author of this article conducted ethnographic research into police storytelling. During the fieldwork period, between July 2010 and June 2012, he spent many hours in patrol cars, briefings, and the canteen of a Dutch operational police force. He had countless informal conversations and conducted formal, tape-recorded interviews with 26 officers. After the fieldwork and analysis of the ethnographic data, he published a report and multiple papers on the storytelling practices at the police.
The interviews yielded 289 stories about all kinds of aspects of policing. During the interviewing and the subsequent analysis, the first author was struck by the following realization. Some of the stories, told mostly by the more experienced officers, described incidents that had had a strong impact on them-for instance, drawing their gun for the first time or getting into a violent physical fight. Besides their length and detailed nature, these stories had a particular richness that warranted a closer look (Weick, 2007). One thing that stood out was their phenomenological quality (Gallagher and Zahavi, 2008), as they described not just what had happened in terms of events (e.g. the drawing of a gun), but also in terms of self-shaping experiences of those events (e.g. what it felt like when drawing a gun). How might this experiential quality of story-telling be understood? This question set off an intellectual puzzle, which led to the present paper (the methods section explains the details of our analytical process).
Reflecting on the puzzle, we realized that what is manifested in the stories are different forms of narrative sensemaking. Research has shown that narrative sensemaking includes both in-situ (situated) and after-the-event construction of stories. Studies of in-situ narrative sensemaking stress that sensemaking has a narrative quality even before sense has been completely made, that is, even before it is fully articulated in words (Boje, 2001;Sandberg and Tsoukas, 2015: S9;Weick, 2009;Weick et al., 2005;Whittle and Mueller, 2012: 112). In in-situ narrative sensemaking, agents find themselves in the middle of unfolding incidents, calling for their response. They focus on salient cues and seek to connect them to make a plausible, actionable story on the spot, which will address the questions: "What is a story here? What should I do now?" (Weick, 2001: 462;Weick et al., 2005: 410). An actionable story both accounts for patterns that may have already emerged in what a sensemaker has faced so far and "energizes" the sensemaker to guide further action to deal with the situation at hand (Weick, 1995: 61). In after-the-event (post hoc) narrative sensemaking, agents articulate in stories their sense of past incidents in which they were involved. Stories do not merely report facts but, more importantly, organize them in meaningful plots, "aiming to communicate facts as experience, not facts as information" (Gabriel, 2004: 64, italics in the original). Moreover, while agents frame particular plot lines for particular audiences, they reflexively weave their incident-specific, lived experiences into their ongoing "life stories" (Bruner, 1987;Cunliffe and Coupland, 2012;Haslam et al., 2017;McAdams, 2008;Pratt, 2012).
In-situ and after-the-event narrative sensemaking are connected, making narrative sensemaking an extended process. As has been increasingly acknowledged (Maitlis and Christianson, 2014;Sandberg andTsoukas, 2015, 2020), sensemaking does not stop once an unexpected incident (or event) comes to an end. It rather continues, insofar as agents revisit the incident, trying to make more refined sense of how they handled it. What, however, has been less explored is how narrative sensemaking about unexpected, extreme events extends (and changes) in time, shaping agents' identity along the way. While parts of extended narrative sensemaking have singly been studied, especially those related to emplotment (Abolafia, 2010;Cunliffe and Coupland, 2012;Whittle and Mueller, 2012) and identity construction (Cunliffe and Coupland, 2012;Tracy et al., 2006), the entire extended process has been under-explored in hitherto research.
The research question, therefore, we will address in this paper is the following: how do agents make extended narrative sense of unexpected, impactful events, with what effects on their identity? Addressing this question is important since it will provide insights into the way organizational members' narrative processing of unexpected events, especially impactful ones, over time, modifies how individual sense is created and what it accomplishes. Moreover, while research has shed light on the critical role of agency in the making of sense when agents confront unexpected events, a more nuanced understanding of this agency is required. For example, in in-situ sensemaking, agents strive for making "contextual sense" (Sandberg and Tsoukas, 2020: 11), while in after-theevent sensemaking, they aim at establishing "conceptual" (or "interpretive") sense. Yet, how the two are linked in agents' accounts to make up extended sense, with what implications, has remained under-studied.
We will explore our research question empirically by analyzing in depth three cases of narrative sensemaking in police work in the Netherlands. Although there are national variations, police work is complex and often attracts criticism from the public. It involves officers dealing with situations that are dangerous or become dangerous to people involved, and may get out of control, with humanly devastating and politically explosive consequences, as recent police killings in the US testify (Shear and Fandos, 2021). The cases we picked manifest some of the complexity of police work, especially since they represent "extreme events" (Chen, 2015;Flyvbjerg, 2006;Harding et al., 2002;Jahnukainen, 2012Jahnukainen, [2010), that is, cases in which storytellers were confronted with an unexpectedly overwhelming incident that has typically had a strong impact on them afterwards.
Our theoretical account extends current understanding of narrative sensemaking by linking different dimensions of sensemaking. Specifically, we make the following contributions. First, we demonstrate how, at different moments in time, storytelling does different sorts of work for sensemakers: enacting situations, emplotting experiences, and renewing identities. Secondly, we show the different types of agency are central to it (situated, narrative, and identity-shaping agency, respectively) and what each type accomplishes (engaged responsiveness, complexified sense, identity renewal, respectively). And thirdly, post hoc narration is understood to be driven by, among other things, a teleologically shaped professional identity that fuels the quest for better practice in the future.
The paper is organized as follows. First, we set the conceptual stage by reviewing and discussing the role of narratives in sensemaking, and connecting narrative sensemaking to Riessman's (2008) perspective on narrative analysis. Secondly, we explain our methods. Thirdly, we present and analyze three cases of narrative sensemaking, identifying in them what the agents accomplish. Finally, we discuss our findings in light of broader themes identified in narrative sensemaking research. interruption, they start paying deliberate attention to it, wondering what is going on and what they can do-they are engaged in "involved-deliberate sensemaking" (Sandberg and Tsoukas, 2020). The latter takes place in the action-present: "the zone of time in which action can still make a difference to the situation" (Schön, 1983: 62). An actionable story of what is going on reflects the "contextual sense" (Sandberg and Tsoukas, 2020: 11) made at a certain point in time. Agents construct such accounts in the midst of action, thereby providing themselves with working hypotheses good enough to enable them to continue acting. Thus, to the extent making sense of an unexpected situation continues in situ, agents act based on a story that is emergent (Weick et al., 2005). This means such a story has an under-developed plot and frames situations, at least initially, in terms of scripts (for instance, in policing: "a bar fight-check it out"). Boje (2001Boje ( , 2008; see also Adorisio, 2014) has called such stories "ante-narratives." An antenarrative is a story earlier-than-narrative: it is "constituted out of the flow of lived experience" (Boje, 2001: 3) and is tied to action; it lacks fully-fledged emplotment and coherence; and it is speculative and open-ended (see also Cunliffe and Coupland, 2012: 74;Weick, 2012: 145;Whittle and Mueller, 2012: 112). An ante-narrative is based on the presumption of sense-that the troublesome situation it refers to is, ultimately, sensible. The narrative fragments will eventually make sense, which the sensemaker needs to strive to create. The ante-narrative does diagnostic work in order to propel order-restoring action (Patriotta, 2003).

Sensemaking as narrative understanding
Narrative is most visibly linked with sensemaking when agents explicitly tell stories about unexpected, awkward or extreme situations after they have transpired. In such circumstances, narrative sense is made without an immediate link to urgent action. Most narrative sensemaking research falls in this category. Front-line service-operators, police officers and paramedics, among others, have been found to swap stories from the past (Orr, 1996;Tangherlini, 2000; van Hulst and Ybema, 2020) on a regular basis. More broadly, storytelling has been suggested to be a routine feature of organizational life (Boje, 1991(Boje, , 2008Czarniawska, 1997;Dawson and McLean, 2013;Gabriel, 2000). As well as sharing stories at work, agents tell their stories when they talk about organizational events to interested outsiders, including researchers and inquiry committees involved in "second-order sensemaking" (Sandberg and Tsoukas, 2015: S23; see also Boudes and Laroche, 2009;Currie and Brown, 2003;Gabriel, 2000).
Through after-the-event storytelling, agents seek to establish the meaning of particular events they faced, relate them to similar past events directly or vicariously experienced, and develop further their professional understanding and competence. The incident of concern may be over, but sensemaking does not stop. In after-the-event sensemaking, agents deal with the experiences they have had rather than with the critical incident itself: in situ experiences are transformed into more elaborate narratives (Labov, 1972), even though their rendering might remain or become terse (Boje, 1991). The unexpected or disruptive events that gave rise to situated (in-situ) sensemaking enter individual and/or collective memory, and are turned into discursive objects. What, during situated sensemaking, was an ante-narrative becomes what Boje (2008: 9) calls "a BME retrospective narrative" (i.e. a narrative with a Beginning, a Middle and an End), producing the agent's understanding of "what it was all about." BME-structured storytelling is a "sensemaking process that transforms unexpected events into an ordered narrative account, [. . .] in which single parts are selected and connected among alternative parts and alternative connections" (Boudes and Laroche, 2009: 379). Selecting and connecting parts of the narrated reality, simultaneously involves ignoring or downplaying others. The process of turning events-and the settings in which they transpire -into a meaningful whole, is accomplished through emplotment (Czarniawska, 2004). The emotional arousal that drove narrative enactment is displaced by emotional work undertaken after-the-event, especially if the events handled culminated in a tragic end, accompanied by "grief, pain, fear, anger, and, possibly, guilt and shame" (Gabriel, 2000: 69;Maitlis, 2009). When after-the-event narration occurs under high accountability pressure, the "work" accomplished is likely to be defensive, self-legitimating or self-glorifying (Boudes and Laroche, 2009;Brown, 2000), whereas when it occurs in a non-threatening environment, as in talking to sympathetic peers, mentors or empathetic researchers, it might provide opportunities for reflexivity and, thus, might bring out more of the storytellers' lived experiences (Cunliffe and Coupland, 2012;Gabriel et al., 2010;Rhodes and Brown, 2005;Whittle and Mueller, 2012). 2 Sensemaking as narrative identity work BME-structured storytelling refers not only to the external world but implicates the storytellers themselves-their identity is often foregrounded. As with sensemaking in general, unexpected (especially if awkward or extreme) events are the most tellable (Labov, 1972). Sensemaking is not merely grounded in the identity of the sensemakers as widely assumed (Weick, 1995: 18-24), but, understood in narrative terms, sensemaking also helps "establish or negotiate an identity vis-à-vis others" (Cornelissen, 2012: 118; see also Cunliffe and Coupland, 2012: 79). Whenever agents have experienced unexpected events, especially upsetting ones, the challenge is to weave their particular experiences with the rest of the experiences they have already formed into (professional) "life stories," thus giving shape to their impact (McAdams, 2008;Maitlis, 2009;Gabriel et al., 2010;Sims, 2003).
For sensemakers, the ongoing question is who they are, as professionals. The answer they provide is the "evolving product of conversations within ourselves and with others" (Currie and Brown, 2003: 565). Furthermore, organizational agents tell their stories not simply as individuals but as members of a certain "practice" (Schatzki, 2006) (or "practice world," Sandberg and Tsoukas, 2020: 5;Spinosa et al., 1997: 17), namely a teleologically structured cooperative activity whose members are, in principle, committed to the pursuit of specific "standards of excellence" (MacIntyre, 1985;Moore, 2017;Tsoukas, 2018). As MacIntyre (1985: 215-216) remarks, "there is no present which is not informed by some image of some future and an image of the future which always presents itself in the form of a telos-or of a variety of ends or goals-toward which we are either moving or failing to move in the present." We can, therefore, expect agents to tell their stories with not just the past, but also the present and future ends of their practice in mind (Cunliffe and Coupland, 2012: 81). Storytellers "(re)interpret the past while 'authoring' a shared future" (Maclean et al., 2014: 544).
While each instance of sensemaking (as narrative enactment, narrative understanding, and narrative identity work) has been studied independently, they have not, so far, been brought together. We, thus, lack a multi-dimensional picture of how extended narrative sensemaking is accomplished when agents conversationally revisit their handling of unexpected incidents. In the rest of the paper, drawing on rich interview stories, we aim to make the three instances of narrative sensemaking speak to each other and, thus, shed more light on what is going on in extended narrative sensemaking. To do so, we will draw on Riessman's (2008) work on narrative, since it maps onto the three instances of narrative sensemaking identified in the literature, thus offering a multi-dimensional understanding of narrative (see also Maitlis, 2012). Riessman (2008) has proposed for the human sciences at large, and Maitlis (2012) for organizational research methods in particular, three lenses (or readings) of narratives: thematic, structural, and performative. The thematic lens focuses on the content of the story, "the told." It identifies the key themes within the story (Maitlis, 2012) by examining "primarily what the content of the narrative communicates, rather than precisely how a narrative is structured to make points to an audience" (Riessman, 2008: 73). This content-focused analysis is consistent with the focus of agents themselves when they engage in narrative enactment -their concern with what happens in a particular incident and how the agent deals with it though the enactment of an actionable story. Therefore, to apply this lens to a sensemaking narrative, the following question is posed: How do agents narratively enact unexpected, impactful events while they are ongoing?
With a structural lens, attention shifts from the "told" to the "telling." A structural lens captures narrative understanding: to look for story-elements, the storyteller selects, forms, and sequences to form a plot. Such an analysis "provide[s] insights into each [story] element's function in constructing the story and show how the elements can be combined in different ways to create different kinds of meaning" (Maitlis, 2012: 497; see also Chatman, 1978). The following question is relevant here: How do agents emplot their experiences of handling unexpected, impactful events after they have transpired?
Finally, with the performative lens, the emphasis is placed on narrative identity work: how the interpretation of events becomes part of the storyteller's identity. Stories are seen as "coproduced in a complex choreography-in spaces between teller and listener, speaker and setting, text and reader, and history and culture" (Riessman, 2008: 105; see also Maitlis, 2012: 499). A performative narrative lens shows how "identities and other narrative accomplishments are produced through certain actors, and in particular settings, performed with and for an audience" (Maitlis, 2012: 499). The relevant research question here is: How do agents weave their experiences of unexpected, impactful events into their broader professional life stories?
In the rest of the paper, we will subject our narrative materials to all of Reissman's lenses. Table 1 summarizes the framework through which we have organized our inquiry (we will discuss column 6 in the penultimate section, since it is connected to our empirical findings). In the next section, we will discuss our methods in more detail. Then, we will explore in depth three stories, told by Dutch police officers, and will show what story-work accomplishes with regard to narrative sensemaking. We will proceed to theorize our findings and will end with conclusions about our contribution, limitations of our approach, and suggestions for further research.

More about methods
To explore the three dimensions of narrative sensemaking (see Table 1, column 1), we examine the narrative sensemaking by police officers. Like other fast responders, police officers often have to make sense of unexpected, novel, troublesome or confusing events in the moment (i.e. situated sensemaking). That is, on a regular basis, they almost simultaneously have to make sense of what has just happened, what is happening, and what might be happening next (Bechky and Okhuysen, 2011). During the 26 interviews the first author had (with officers from different ranks, years of experience, and gender), he first asked officers to tell him how they got to do the work they do and what they (dis)like about it, before proceeding to invite officers to recount memorable events they had told others about.
As we explained in the Introduction, the value of a sub-set of the 289 elicited stories struck the first author while he was interviewing and analyzing stories. As we then jointly and abductively moved back and forth between reading data and theory (Cunliffe and Coupland, 2012: 71;Locke et al., 2008;Sims et al., 2009), we came to look at them in the light of current sensemaking theorizing. We noted that the officers we talked to had lived through extreme events in which they had to make sense on the spot (narrative enactment); they had told others about them on various occasions (narrative understanding); and the events experienced had impacted on them professionally or even personally (narrative identity). Some of the officers reported extensively on those events, while others did not get into detail or might have decided not to share some more impactful stories. (The first author had met the interviewees in the course of the fieldwork and often had gone on the beat with them before interviewing them. He presented himself as a empathic listener, someone trying to understand police work as officers experience it (Mishler, 1991)). We looked in our interview database for "rich" stories (i.e. stories that would enable us to understand in some detail the sensemaking that was going on in the stories), using two criteria. First, we selected the longer stories in the data set, on the assumption that they would be richer. The stories in the interview database (289 in total) ranged from a few sentences to six pages. Second, we chose first-hand accounts of unexpected events with significant impact, in which narrative sensemaking was vivid. To understand in greater depth the three dimensions of narrative sensemaking, we decided to focus on three stories. The reasons are as follows.
Firstly, the stories involved detailed descriptions of sensemaking. The three stories refer to incidents that lasted for a relatively short period (10-30 minutes), whereas some of the other rich stories covered several hours and offered, in comparison, less dense descriptions of narrative enactment. Secondly, the three stories displayed the basic elements of well-formed BMEs and-in combination with data from the wider interview-informed us about the narrative identity as well (Cunliffe and Coupland, 2012). Thirdly, the stories included a recollection of the officer's actions, interpretations and feelings of what was going on while it was going on. This juxtaposition of thought, emotion and action allowed us to see how these elements work together to create sense, thus restoring, to some extent, "the past to its own present with all its incoherence, complications, and 'might-have-beens'" (Weick, 2007: 17). Finally, in the three stories, the storyteller was confronted with an unexpected situation in which he or she had, largely, felt powerless and that culminated into a tragic incident. The stories refer to extreme cases, experiences which put pressure on the storytellers to seriously extent their sensemaking into the second and third dimensions. Not surprisingly, the stories were told as part of an ongoing sensemaking effort by the interviewee (two of the interviews lasted two-and-a-half hours, one lasted an hour), during which the officers placed the story in the context of their own unfolding narrative identity.
By focusing on a small number of cases, we join a distinct tradition in organizational (and, more generally, social scientific) research (Cunliffe and Coupland, 2012;Gabriel, 1991;Hopkinson, 2003;Maitlis, 2009;Rhodes, 2000;Sims, 2003;Weick, 1993). In particular, in discursive and dialogical psychology, ethnomethodology, conversation analysis, and deconstructive analyses, using a few cases (even a single case) or narratives tends to be the norm (Cooren, 2007;Edwards, 1997;Llewellyn and Hindmarsh, 2010;Martin, 1990;Rhodes, 2000;Sawyer, 2003;Whittle and Mueller, 2012). The benefit of focusing on a small number of cases is that it preserves contextual richness as well as "sequencing and interrelatedness" (Maitlis, 2012: 507). Moreover, as Antaki and Horowitz (2000: 157) note, focusing on single interactional events narrated in stories, enables researchers to track "the development of a piece of social action as it accumulates over the length of an episode" (see also Whittle et al., 2009: 430).
There are various ways to do narrative analysis. From the several types of narrative lenses available for detailed analysis (e.g. Boje, 2008;Czarniawska, 2004;Gabriel, 2000;Labov, 1972), we chose to work with Riessman's (2008) because it offered us a clear distinction between three fundamental narrative lenses that, as mentioned earlier (see Table 1), connect well to the narrative sensemaking literature. To be sure, using Riessman's three lenses side by side as readings of the same material is not what Riessman (2008) proposed, although Maitlis (2012) hinted at it. However, our narrative reading of the police officers' stories is meant to work as a "thick description" (Geertz, 1973), using rich, detailed text to elucidate the layers of meaning that can be read from human action. To analyse each story through the three lenses, we started with identifying the themes in it that constitute narrative enactment and then moved sequentially to applying the structural and performative lenses.
Regarding our reconstruction of narrative enactment, we are aware that there is a limit to the accuracy of this lens by using "mere" interview data. Theorists of interviewing (Alvesson, 2011;Silverman, 2011) argue that interviews offer limited proofs of things that happened in the past (which would allow us to reconstruct in situ sensemaking), as interviewees engage in post-hoc representation of events. Interviewees may not know or remember all facts concerning the situations they acted in. They might also misremember. They might lie about aspects of their actions, thoughts or feelings at the time, and they might exaggerate.
Be that as it may, the stories we selected describe sets of events (actions, but also thoughts and feelings) in a manner that is relatively unproblematic for what we want from them. In line with the most authoritative work on it (Weick, 1995(Weick, , 2001Weick et al., 2005), the descriptions culled from interviews reflect what one can typically learn about in situ sensemaking. 3 Moreover, our interviewees selected the stories they told us. It is not very likely that they selected incidents they felt they had to lie about. The stories are in a sense "confessional tales" (Van Maanen, 2011[1988), told to an empathetic listener, providing inside accounts in how police officers dealt with surprises on the spot. The officers in the stories come across as vulnerable agents, who partly legitimize their acts, but also show their inabilities and limitations-they are hardly incidents one can brag about.
Finally, our analysis does not aim to throw new light on each dimension of narrative sensemaking alone, but to explore the three dimensions of narrative sense making as they manifest themselves side-by-side. Since we aim to comprehend how extended narrative sensemaking is accomplished, we are methodologically interested in the simultaneous availability of different narrative sensemaking dimensions in a single text (Rhodes, 2000). The post-hoc reframing of events in the cases is thus not ignored, but, on the contrary, it becomes part of the analysis as we move from the thematic lens to the other two lenses (structural and performative) (Rhodes and Brown, 2005: 171). Sensemaking "after the event" (i.e. telling researchers what happened) provides additional interpretive readings of the same interview data, since it invites an analysis not just of the events-focused story, but also of the telling and the storyteller as presented in the interview. In that sense, the structural and performative lenses can be used to scrutinize the thematic lens. The weakness of interviews (they "merely" reconstruct events, "shaped by the speakers current place in the world," Maitlis, 2009: 53) is, thus, simultaneously their strength-stories require storytellers to construct a perspective on events and on themselves (Rhodes, 2000).
In the next section, we first present each one of the three stories (text in italics), followed by an analysis, in which we apply the three lenses. After that, we theorize the accomplishment of sensemaking story work.

Findings
The day I drew my gun A male patrol officer (born in 1975), who had joined the police some 8 years prior to the interview, told us the following story: Thematic reading: Sensemaking as narrative enactment. The initial information and assignment formed a part of the first (official) diagnosis of the situation, activating "immanent sensemaking" 5 (Sandberg and Tsoukas, 2020: 9): "a family row, involving a possibly psychologically unstable person; go out and check." This is a stylized "embryonic narrative fragment" (Gabriel, 2004: 70), guiding initial action, which is open to further narrative elaboration as the situation develops. Upon entering the scene, the officers used their professional routines (walking in a single file, addressing a person) to enact the situation. Police presence turned the in-house family row to a public situation with legal implications. However, officers' initial routine actions caused reactions, thus altering the situation.
As officers' involved-deliberate sensemaking develops, the answers to the questions "What is a story here? What should I do now?" (Weick, 2001: 462) keep changing. The officers' ante-narrative is redrafted as the situation evolves: from (a) "we are about to check a family row involving a possibly psychologically unstable person"; to (b) "we are under attack; suspect is threateningly running toward us with big knives"; and to (c) "we seek a rescue exit by running for the lift; guns are out, situation is dangerous." Moreover, what is interesting to notice in this thematic reading, is the "help" the police officer got from the (trigger of the) gun and the elevator (cf. Latour, 1994). These pieces of equipment -materials that make their presence felt -were critically involved in the third act of enactment, saving the officers from firing their guns and the suspect from being shot.
Structural reading: Sensemaking as narrative understanding. While the thematic reading focuses on the narratively-enabled ongoing enactment in the sociomaterial context of the incident, and is tied to the particularities of the developing situation, a structural reading treats the situation as a discursive object that is done and over with: the storyteller emplots his story, tying events together, poetically elaborating on his narrative material by highlighting certain aspects of what happened in order to communicate an account that provides insights into his experience of the incident. He now has more narrative control, since he was able to find out more about the events and knows how it ended. The set of events have turned into an incident he can (narratively) master as a teller. That the suspect was running at them holding big knives, for instance, seemed plausible in the action-present, but proved to be untrue in the investigation after the events. The structural reading, however, does not just likely bring in more accuracy with the benefit of hindsight, but, also, allows the storyteller to reconstruct the events as he personally experienced them-a BME narrative is, thus, generated.
In the BME narrative, one particular event is highlighted (the moment the officer drew his gun) and a certain notion of agency is projected. At this point of the story, the storyteller slows the story down, drawing his audience's attention to what the event he highlights meant to him. He says he felt the weight of the trigger, thus realizing that he had spontaneously drawn his gun. The realization of what he had done makes him engaged in a moment of reflection-in-action: "What am I doing now?" The officer had surprised himself as he confronted the enacted situation he had helped create (third episode of enactment). He checked with his colleagues to intersubjectively find out whether he had just lost his mind or had indeed entered into a dangerous situation, which would permit him to defend himself with his gun. In Weickian terms, it is as if the officer had said to himself: "How can I know what is going on until I see what we are doing?" (Weick, 1979: 5).
Although his description fits quite well with what we know about the social nature of sensemaking and the interrelation between thought and action in it (Weick et al., 2005: 419), we cannot know for certain that all of this took place in his mind. From a structural narrative perspective, however, what is important is not literal truth but experiential truth-what the experience of drawing the gun meant to him afterwards. The officer emplots his narrative in such a way as to highlight the unusual experience he has had and, critically, to project reflective agency (Yanow and Tsoukas, 2009). It is as if he is saying: "I immediately noticed how unusual what I had done was; I became aware of the weight of the trigger; I sought confirmation with my colleagues; in short, I acted reflectively-inaction." From a structural narrative perspective, "drawing the gun" is not seen as a spontaneous response to an unexpected situation (as it was in the thematic reading), but as a discursive object to be reflected on and poetically elaborated (Gabriel, 2004: 65). The emphasis is on the telling of a story and, through it, the accompanying images of agency projected. Simultaneously, the storyteller legitimizes his actions, implicitly blaming the anxiety he had felt and what the arousal had done to his senses. The situation is turned into a story about "the day I drew my gun" (an unusual event for this Dutch police officer and to most of his Dutch colleagues).
Performative reading: Sensemaking as narrative identity. Drawing a gun is a typical example of a "small action with [potential for] large consequences" (Weick et al., 2005: 419). Even if no shot is fired, drawing a gun is a meaningful and consequential act for those who undertake it. Symbolically, the gun (and similar arms) is the most tangible representation of the state's monopoly of violence. An officer who fires his gun may kill someone. From the thematic reading, we can understand the arousal that must have played an important part in the way the officer acted. The structural reading helps us to understand the emotional labor that must have taken place afterwards. These emotions lead from a focus on the events and the acts of the police officer himself to his identity and what his actions tell us about him as a police officer. A performative reading helps us further understand that, by telling the story about the day he drew his gun, the officer is performing a more generic professional life story. In the interview, the officer told us about various experiences he had had over the course of his professional life and connected those to who he is and aspires to be, as a professional. In his own words:

"[On the streets,] I am really a 'talker,' so to speak. I am more into mediation. If I can make somebody see, through a conversation, that his actions were wrong, I do not give a fine. If I am in a conversation with somebody and I see that what I say, my appearance as such, the fact that I address a person,. . . if that makes an impression [on that person], I do not write a ticket. Just as I say, I like giving aid more than law and order".
Here the officer describes his professional identity as an officer-what his personal style of policing is. The situation in which this officer felt compelled to draw his firearm had been the first and only time, and it had made a lasting impression on him. Through his story (the BME narrative), the officer wants to convince us of his character, as molded by official policies and professional experience: an officer who prefers helping people over maintaining, through potentially harsh means, law and order. This identity is not a merely personal matter but reflects a broader espoused value of Dutch policing: use as little force as necessary to handle situations. 6 At the same time, through the story, the officer warns us for the precariousness of that identity, stemming from the dangers of police work: the setting might make even a benevolent officer do evil things.

Driving in reverse
A female team manager (born in 1958), who joined the force in 1978, told us the following story from the time she was a patrol officer:

Then, a man comes up to me and says: "Misses, would you allow me to tell you what happened?" I heard remarks "Ha, the police again, was driving way too fast!" Well, that was totally on my mind. He says: "My car was on R-street and my briefcase was in it. I am approaching my car and see this bloke in my car. He grabs my briefcase and starts running. I ran after him. While running, looking over his shoulder, he did not see you coming." Hé, that is a totally different starting point of it all.
And then it turned out that this boy was also in the system, he was a drug addict who hadn't been in contact with his parents for years. The family was notified [of the incident] and they said like: "Boy, this sounds awful, but we are happy he is gone, because he only caused us trouble over the last years." We went to the funeral. We had something like, 'should we go to the funeral?' Because, you do kill somebody. There was nobody there. That I found the sad part of the story. You think, hey, so this is the way our society works. Thematic reading: Sensemaking as narrative enactment. The call the officers responded to involved a situation that had been going on and they went there to check it out. Immanent sensemaking and the associated routine was activated: "riots; go and check out". They were driving the one-way street in the right direction, but with the car in reverse (first scene of enactment). When they unexpectedly run into a person, they shift to involved-deliberate sensemaking (second episode of enactment). There is no time and no need to find out what could have led to the situation they were facing. The two officers are confronted with the collision and its result: a badly injured person lying behind their car with what might plausibly be a carotid bleed. In line with their role as defined by the Dutch Police, the officers instantly enact their role as first responders (third episode of enactment), dividing their attention between instantly helping the victim and asking for more help through the car radio. The officers' ante-narrative keeps morphing: from the routine understanding "we have a riot to tackle", through the unexpected twist "we have run down a person", to the new understanding "we've got to help the victim".
Structural reading: Sensemaking as narrative understanding. After the narratively-mediated enactment stops (at least for the officer involved), she is confronted with an unknown citizen-storyteller. He is somebody who volunteers to assist in that other role that police officers and their colleagues have: establishing as accurately as possible what has happened in the situation under investigation. According to the officer, the storyteller asks for the conversational floor: "Would you allow me to tell you what happened?" He then starts the story with the victim in the collision. Now, we see the person who we met as the victim of a car accident as a suspect of a car theft. But the storytelling does not even stop at this point. Yet another layer is added when the officer tells us about finding out more about the victim-suspect (a drug-addict), the reaction of the family to the events that had taken place (oddly, indifference), and the funeral at which they were the only guests. The officer emplots the events she narrates to show the moral complexity of police agency: she provides a deeper insight into police work as experienced on the streets, by showing how a professional can be confronted with complex cases, in which right and wrong are hard to establish. The officers drive on a one-way street illegally (in reverse) for the sake of a worthy purpose (to find out suspects for a riot). They unexpectedly run onto and kill an unsuspecting citizen (tragic), who, however, turns out to be a known drug addict and, here, a likely thief running away from a crime scene. The victim's family do not even bother to go to the funeral, but the officers responsible for his death do. Police work is portrayed as paradoxical and complex: tragic sometimes, punctuated with unintended consequences, aiming to protect the common good (law and order) but, paradoxically, inflicting individual death along the way.
Performative reading: Sensemaking as narrative identity. A performative reading takes us beyond the story proper (the selective recounting of events surrounding a particular unexpected situation), to see how the narrated events fit with the broader professional life of the teller and his/her selfunderstanding of good practice. This reading turns our attention to the impact the tragic incident had on the storyteller and the way it shaped her understanding of her work. Regarding the first, it is deeply unsettling to have your "hands covered in blood." The officer appears as a victim of events and their aftermath: "In the long run you learn to live with it, you give it its proper place," she says. In other words, such events become part of who you are as a professional and a person, getting emotionally embedded into your identity ("it always stays with you"). We should also notice that, at some point after describing the action that took place, she says "I became emotional-I still am." In other words, the impact was not just that it made her emotional; she feels the impact still 30 years later-it is ongoing. This impact led her to develop a perspective on the way officers should treat their colleagues in similar situations (with care).
Secondly, the storyteller claims that her view on the people she works with ("suspects," in particular) has been profoundly reshaped by the incident. Having experienced the moral complexities of this particular case helped the officer to develop a perspective on the treatment of the people officers encounter in the line of duty (i.e. with respect). What matters in the wider context of the practice of policing is the morality the police officer upholds. The officer argues that the police should start from, or look for, the good in people. As she says in her interview: "We are all people. And in all people there is something good, I always say. In some you have to look a little harder, but it is true. Well, that drives me." As in the first story, the shaping of one's identity is not a purely individual accomplishment: it is impacted upon by personal experience and official policy and culture. According to official Dutch Police policy (well-known by police officers) the police is expected to not just uphold the law, but also "to give help to those who need it." 7 To pass a judgment, the officer says, that is not up to me, but to a judge.

Holding a dead child
A team manager (born in 1964), male, who had started working at the police in 1986, told us a story from back when he was a patrol officer: Thematic reading: Sensemaking as narrative enactment. The officer's description starts with immanent sensemaking: reporting routine observations ("a woman on the sidewalk") and what he made the things out to be ("the mother, clearly"; "a van" becomes "a delivery van"). It was clear from the start that the emergent story was emotionally shocking. But what is also clear is that from the moment the officer came into contact with the situation, there was no "solving" of the problem or restoring an interrupted activity as is the case when immanent sensemaking is interrupted. The action was mostly past. There was not much more to enact there and then. The only thing he could do was to make some overall sense of what appears to be a tragic, senseless situation: what happened, by whom, how, when. The officer must have felt quite helpless. Involved-thematic sensemaking and the associated narrative enactment of an unfolding situation are minimal here, since the damage had already been done-the girl could not be rescued.

Structural reading: Sensemaking as narrative understanding.
A lot of what the officer describes is not so much his own situated sensemaking at the time the incident took place, as the sense the driver must have made before and after hitting his daughter. The reconstruction the officer provides became possible through the police investigation. The detailed, objectifying nature of the reconstruction makes the officer seem an outsider to the situation. This fits with the nature of his job and the organization he works for. After an incident has transpired, the police need to determine not just plausibly, but with as much accuracy as possible, what has happened in specific situations. When moving to the structural reading, we realize that the officer's story is a BME narrative with a particular emplotment. That is, the detailed description of the accident itself does not just give us the facts. It does more. The description helps us see what happened as a tragic plot (Gabriel, 2000). The man always parked his car in a way that other cars would slow down and the kids would be safer. It is tragic that this safety measure became a dangerous move-the father running over his own daughter. Through his BME narrative, the officer portrays his work through the lens of tragic agency-a witness to tragic events calling for his attention. That sense of tragic agency is more sharpened by an awareness of his own personal circumstances.
Thus, the story emplots the events by starting out with a description of the personal situation the officer-in-action had been in when he arrived at the scene: "I just became a father the year before." Although this is an important part of one's personal life, it does not necessarily differentiate one officer from another at work and, therefore, it is usually not mentioned. The reason why the officeras-storyteller highlights this personal detail is because, for him, it became highly relevant in this particular situation. Implicitly it is as if he is saying: "Of all the things I could say about my situation at the time, this is what you must know to understand what I experienced back then. As a policeman, I witnessed a most tragic event that was accentuated by my own parenthood." Performative reading: Sensemaking as narrative identity. Moving to the performative reading, we ask ourselves how the storyteller fits the events experienced in with his broader professional and personal life. In the first instance, the police investigation and the reporting seem to have helped the officer to get a clear understanding of all that had happened before he arrived at the scene. His detailed description shows this. But more than that, in the story itself, he tells us how he entered the situation as a police officer and how his personal identity -being a father -got triggered as well. Put differently, when the call came in, his fatherhood was not a point as such, but it became relevant later on. His identity as a father got foregrounded as he involved himself in the moment and afterwards. In addition, he found value in his own identity as a father to engage the father who had just accidentally killed his child. This enabled him to have compassion.
If we zoom out further, to the meaning the situation acquired in the years that followed, it is important to note that the accident took place at the beginning of the 1990s. The Dutch Company Care Team, whose task is to help its members talk to their distressed colleagues who have just experienced a high impact incident, did not yet exist. The accident and its aftermath, however, were decisive for the officer's involvement in the Company Care Team some years later. In fact, the connection is clear from the way he frames his story, starting it with "If people ask me why I joined the Company Care Team. . ." The meaning he gave to what he had experienced became part of his own life story and career, so much so that he found a new mission. To this day, he is active in the Company Care Team as a coordinator.

Theorizing the making of extended narrative sense
In this paper, we set out to address the following question: how do agents make extended narrative sense of unexpected, impactful events, with what effects on their identity? The stories we selected included critical moments when officers' immanent sensemaking and their associated routine actions had become problematic, in the face of unexpected, extreme events. 8 We have explored here the making of extended narrative sense. Our interest was not only in how unexpected, extreme events have been narratively dealt with in situ, but also in their afterlife-namely, how these events were handled as discursive objects and might lead to changed selves. Our stories are instances of an ongoing process in which narrative enactment occurred during in-situ sensemaking, followed by after-the-event BME storytelling that emplots experiences and connects them and/or shows how they have been connected to the broader professional or personal identity, thus renewing it, which then becomes the basis for engaging new situations later on, and so on.
A common pattern emerges through the stories, concerning the way storytellers present their situational agency: they project an image of engaged responsiveness to an ongoing situation. They portray awareness of their role in helping co-produce the evolving situation they face and appear willing to revise their ante-narratives on the spot, to enable them to enact an orderly resolution of the problem at hand. Ability to think on their feet is primed. At the same time, other, somewhat unexpected agents or actants, already on or entering the scene, shape and inform the development of the emergent actionable story as much as the officers themselves: a trigger and an elevator (Story #1), a citizen-storyteller (Story #2) and a dying girl (Story #3). Whereas police officers typically have crucial resources at their disposal to control the situations they face (i.e. weapons, training), our cases exemplify situations in which the apparently powerful officers do not gain control over events. Their in-situ (involved-deliberate) sensemaking is as much shaped by other agents or actants and what is happening as it is by their (forceful) enactment. Their assumed episodic power, in that sense, is failing (Schildt et al. 2020).
Once agents have moved beyond handling the unexpected situation, they are able to tell a "full" (i.e. BME) story of what happened to them. The storyteller then places events in a setting and describes the actors involved, highlighting some elements of the situation while ignoring others. When we compare the emplotment of a BME narrative told to an empathetic outside listener with the enactment of an ante-narrative as reported to have developed during in-situ sensemaking, there is an important difference. Clearly, the storyteller now "knows" the story. When tackling an unexpected event, officers-in-action do not know what will happen next. With the story of the incident emerging in front of them, they have to work with what is plausible and actionable (Weick et al., 2005). That is a big problem the officers-as-after-the-event-storytellers do not face. Of course, afterwards, storytellers may remain in the dark concerning certain aspects of what happened in action, but, typically, they get to know things they did not know when they were in the midst of the unfolding situation. Such knowledge enables officers to tell a relatively coherent BME story. Through it, they ascribe meaning to their experience, in a way they could not do when engaged in situated action.
The after-the-event assignment of meaning clarifies for the officers what it was all about, thus creating a more complex sense of their experience. When they were engaged responders (that is, when they were engaged in involved-deliberate sensemaking), their attention was pragmatically directed at the action-present-responding to the contour of a co-enacted, evolving situation was more important than ascertaining discursively, in a detached manner, its meaning (Weick et al., 2005). Situational agency is necessarily circumscribed by immediate events and urgent concerns. However, stepping back from their action-present concerns allows officers to revisit their experience of the events and, thus, recover portions of the unfolding experience they had originally not been explicitly aware of. Their tacitly understood situated agency is now foregrounded and new meaning is created (Tsoukas, 2011). As after-the-event storytellers, they can see what they were only implicitly aware of as engaged responders: their own role in situated sensemaking. Thus, after the event, through BME storytelling, the significance of the incident becomes apparent to them. In the context of the telling we relied on in our research, like other after-the-event contexts in which officers are asked to recount their experiences, storytellers can regain some of their sensemaking powers. They could, in the context of our interviews, choose what stories to tell and how to tell them -they had narrative control and the chance to exert important episodic sensemaking power (Schildt et al., 2020). Our three storytellers, however, decided to share the tragic stories we have reported here rather than stick to epic stories that would make them look good, or comic or, more likely, cynical stories about their work (Dean et al., 1998;O'Leary, 2003). Looking back at the incidents, they see themselves as: surprised by their own deeds (Story #1); carriers of paradoxical and morally complex agency (Story #2); tragic witnesses (Story #3). Through BME storytelling, then, narrative agency takes place, which enables sensemakers to complexify their sense by reviewing their own situated agency (Shotter and Tsoukas, 2011; see also Maitlis, 2009). Weick et al. (2005: 419) write: "To deal with ambiguity, interdependent people search for meaning, settle for plausibility, and move on." True as this is for agents engaged in involved-deliberate sensemaking, organizational members may have a hard time moving on afterwards, because of the impact of what happened to them or others in the incident they handled. While the normal work of police officers includes dealing with what to many members of the public appear as unusual or extreme events (Granter et al., 2015), for many officers (if not most) it also entails coming to grips with rare events that have significant impact on their professional lives (and possibly their personal lives as well). In line with this, our stories show how typical emotional arousal found during involved-deliberate sensemaking (Maitlis and Christianson, 2014), was replaced by elaborate narrative descriptions of inner thoughts and feelings (Story #1: "What am I doing now?"; Story #3: "As if it were your own kid."). The initial sense of the incident morphs over time, as those who experienced it relive it narratively afterwards. As feelings last and are re-felt, the authors of Story #2 ("It always stays with you") and Story #3 ("I have made my peace with it, but I do carry it with me all of the time") use metaphors in their stories that stress the bodily nature of lived experience (Cunliffe and Coupland, 2012;Sandberg and Tsoukas, 2020).
Interestingly, although critics of interviewing warn about interviewees distorting and possibly lying to make themselves look good, the storytellers here depict themselves more as victims than heroes (Story #3: holding a dying child) and even as possibly wrong-doers (Story#1: drawing a gun when not needed; Story#2: killing somebody with a car), thus bringing out the moral ambiguities of policing (Dick, 2005). Through BME storytelling, we come to see, agents are enabled to not only connect the relatively distinct episodes that composed their experiences in order to create meaning concerning their role in this process but, also, to connect their BME stories with the rest of their life stories, in search of "narrative unity" (MacIntyre, 1985: 205)-a coherent account of a unified "narratable life" (op.cit., p. 217). Through reweaving their life stories to coherently include BME stories, agents renew their identities.
Agents do not merely register the happening of things but are subjects constituted by their experiences (Mesle, 2008). Experience reconfigures an already established pattern; an impact-full story reweaves an already formed life story. The reweaving can accomplish various things: demonstrate the precariousness of professional identity (Story #1); sharpen police-work values, such as care and respect for colleague and suspects (Story #2); redirect one's career or outlook in life (Story #3). The telling of these stories in an interview should be seen, partly at least, as an act of self-empowerment (the teller decides which story to tell and how; the agent decides how to continue living his professional and personal life, following his/her experience of highly impactful events; Mishler, 1991: 117-123). It is the effort for the agent to regain power by re-narrating past events in a way that they cohere with the teller's identity, which, in turn, evolves as it re-weaves these events into the agents' life story.
As MacIntyre (1985) has noted, agency and narrativity are intimately connected: "I can only answer the question 'What am I to do?' if I can answer the prior question 'Of what story or stories do I find myself a part?'" (MacIntyre, 1985: 216). Professional agents, be they police officers (van Hulst and Ybema, 2020), pilots (Fraher and Gabriel, 2014) or musicians (Beech et al., 2016;Maitlis, 2009), understand their identity in terms of "a narrative web that nestles in the identity narratives of each member of the community, blending elements from their profession's past successes and trials, present challenges, and future aspirations and hopes" (Fraher and Gabriel, 2014: 927). The narratives agents live out are characterized by open-endedness and driven by teleology.
Our police officers do not know what will happen next in their lives (open-endedness). Nonetheless, their lives already have a form, which projects itself toward a shared future. How police officers do their work is critically shaped by what their practice world is for (teleology -the ends they collectively pursue); the affective dimension of their work (pursuing policing matters to its practitioners, as the stories indicate); and the standards of excellence they are bound by. For officers in the Dutch context, the telos of police work is to uphold the law by avoiding unnecessary force; be decisive, helpful, and respectful.
Thus, the stories police officers told us are not only descriptions of sequences of events or even plots, but also performances of a "purposive" professional identity (Chia and Holt, 2006: 644): they are told from the perspective of the telos police work aims at, with officers manifesting their (sometimes agonizing) quest for better practice. Even though institutions like the police and their working culture can bring officers to celebrate heroic action and cover up mistakes, their telos also exerts "systemic sensemaking power" (Schildt et al., 2020) over officers, as their identity is shaped by the extent to which they believe their actions are driven by the collective purpose and they, thus, are worthy members of the profession. The plots in the three stories make a point beyond the events narrated. The notions of agency revealed through our three readings (situated agency, narrative agency, identity-shaping agency) are oriented toward accomplishing particular tasks (respectively: engaged responsiveness, complexified sense, identity renewal) (see Table 1, last column).
Finally, the telos of policing endows the stories with a going-forward quality. The stories analyzed here are shaped by the officers' ongoing participation in the practice world of policing, which makes officers reflect on the adequacy of their standards of excellence when narrating critical events of their practice. Over time, even if a BME story officers tell about an incident can remain quite stable, it might have substantially altered a professional's life and self-understanding, as Story #2 and #3 in particular attest. In this, standards of excellence provide a benchmark in terms of which current actions are evaluated. That is why our interviewees do not neutrally describe events and actions, but do so within the evaluative framework of their practice world. And, then, we come full circle: insofar as after-the-event storytelling reshapes agents' identity (Cunliffe and Coupland, 2012), the latter helps shape the actionable stories agents will work with in situated sensemaking, in the future.

Concluding thoughts
The narrative sensemaking literature has, by and large, looked separately at three distinct yet interrelated processes: narratively-shaped in-situ sensemaking; post hoc narration of the handling of unexpected incidents; and evolving selves, in the aftermath of first-hand handling of unexpected incidents. While all are important, they have hitherto not been studied together to show the key features of extended narrative sensemaking. In this paper, we sought to develop a more sophisticated notion of extended narrative sensemaking.
Connecting three ways of theorizing narrative sensemaking and, accordingly, using three readings of our empirical materials, we have sought to study how agents make extended narrative sense of extreme events, when they revisit them in an emphatic context. Our contribution is threefold. First, we expand the scope of narrative sensemaking in organizations. Specifically, we have demonstrated how, at different moments in time, storytelling does different sorts of work for sensemakers-enacting situations, emplotting experiences, and renewing identities. Secondly, our theoretical account of extended narrative sensemaking shows the different types of agency involved in it (situated, narrative, and identity-shaping agency) and what each accomplishes (engaged responsiveness, complexified sense, identity renewal, respectively). And thirdly, post hoc narration is understood to be driven by, among other things, a teleologically shaped professional identity that fuels the quest for better practice in the future.
All and all, studying extended sensemaking enables us to see the loop that connects ante-narrative, BME narrative, and teleology. Appreciating this allows a more comprehensive understanding of narrative sensemaking as an ongoing, extended accomplishment. The meanings of unexpected incidents develop as temporal distance grows, situational requirements change, and agents seek to get a deeper sense of their experiences through after-the-event storytelling. The latter leads them to refocus from the exigencies of the unexpected situation to their role-related experiences, and their professional selves.
Our concluding thoughts are directed at our methods, limitations and future research. Firstly, regarding our narrative methods, Riessman (2008) and, to some extent, Maitlis (2012) have presented the three lenses to study narrative materials mostly as distinct analytical options to choose from. These readings, we admit, are a small selection of kinds of readings possible (Rhodes, 2000). We have, however, demonstrated here how the three we use here could supplement one another if applied simultaneously. Such use of the three readings offers the promise of a wide range of applications. Moreover, being aware of the limits of interview data (Alvesson, 2011;Silverman, 2011), we propose research in which more and different kinds of data are used to further explore the three dimensions of narrative sensemaking surrounding one particular set of events. Observational, archival and diary data, as well as videotapes of narrative enactment in which agents make sense of events and their identity in relation to those events, would strengthen analyses.
Secondly, although we opted for a research design that sought to bring out the analytical benefits from simultaneously applying three lenses, other research designs would also be beneficial. Thus, a longitudinal research design focusing on the same incident would have revealed how police officers repositioned their own role and identity over time, depending on who the account was for. Yet another research design could be comparative: to compare the accounts of police officers to different incidents, seeking to inductively capture theoretically significant similarities and differences. Moreover, a research design that focuses on stories told in a high-accountability (and, therefore, judgmental) context, such as public inquiries (Boudes and Laroche, 2009;Brown, 2000;Dwyer et al., 2021), would likely yield different stories and highlight different types of agency to the ones we studied.
Thirdly, our study was at the level of individual agents . It would be interesting to explore how teams and other collectives (Bechky and Okhuysen, 2011;Dawson and McLean, 2013) make narrative sense in the three dimensions we distinguish. How would team members tell about the same situations from their own individual perspectives? How might the same situations renew the identity of a team differently? And how is collective narrative sense made through time (Brown, 1998)? We also advise investigating the way sociomaterial practices, such as police work, influence how individuals and groups make sense. Engaging such questions would allow scholars to explore how team sensemaking is narratively constructed and show more of the contested nature of narrative sensemaking and of the power struggles that go with it (Abolafia, 2010;Brown et al., 2008;Schildt et al., 2020).
Fourthly, we picked extreme cases in which the storytellers were seriously prompted to engage in elaborate after-the-event sensemaking because of the powerlessness they had felt. Such cases allowed us to study the three dimensions well but leave open the question in what manner less extreme cases could lead to a renewed identity. At the same time, this might also allow for a more thorough study of those aspects of the narrative that are important from a sensemaking perspective, which, however, our present data hinted at but was not revealing enough to allow us to delve deeper into, such as the emotional and institutional dimensions of the narrative sensemaking (Maitlis, 2009;Schabram and Maitlis, 2017).
Finally, police work, like that of other first responders and frontline organizations, entails dealing with crises of which the action-present is rather short and distinguishable from its afterlife. In other cases, however, narrative enactment and after-the-event narrative understanding might be overlapping as the action-present stretches to include moments of reflection on it (Yanow and Tsoukas, 2009). As in organizational change projects (Boje, 1991) or policy making (Abolafia, 2010), a story might be emergent for a long time. The study of such a case does not prevent the use of our framework, but would be a new challenge. In line with this, through our framework, other kinds of narrative material, like reports (Boudes and Laroche, 2009;Brown, 2000), could also be studied. 7. See https://www.politie.nl/en/themes/about-the-netherlands-police.html 8. We want to distinguish extreme events from extreme work as Turnbull and Wass (2015) have studied it: a constant high workload, responsibility and pressure. The daily work of most police officers "is not replete with car chases, armed robberies or public disorder" (Turnbull and Wass, 2015: 513). Extreme events, therefore, are more an exception than a rule in regular police work.