Formative intervention is a participatory methodology to support a group of professionals, within a work activity system, in the analysis, design, and implementation of new ways of working (Engeström, 2011). In this model, innovations are negotiated with the practitioners, and data are collected to explore the functioning and potential of such groups in their own context. The process of qualitative improvement and change pursued by this methodology is conceived as a progressive increase in the co-construction of new meanings and new work practices (Zucchermaglio and Alby, 2006). Change is approached as an emergent property of the participating organizational group, and the competencies that are needed to develop such a process are considered as located and distributed within it.

The methodology of formative intervention is implemented and developed, in practice, by means of “a toolkit called the Change Laboratory” (CL) (Engeström, 2015), inspired by the work of Vygotskij (1934/1990) and based on the activity theory (Engeström et al., 1996; Virkkunen and Newnham, 2013). This methodology has been widely applied in Europe (see Engeström, 2005), both in healthcare and educational organizations (Stoppini et al., 2009; Ivaldi and Scaratti, 2020) and is at an early stage of development in Latin America (see Montoro and Brito, 2017; Pereira-Querol et al., 2019; Brito, 2020; Vilela et al., 2020).

The CL, which is essentially based on the guided activation of “self-managed” improvement processes, consists of a series of sessions in which practitioners from an organization are supported by means of a bottom-up participatory approach in analyzing the history of their activity system, including sociohistorical tensions and contradictions. One of the main objectives of a CL process consists of unlocking the transformative agency of practitioners to become protagonists in the creation and implementation of a new model of practice (Engeström, 2015; Laitinen et al., 2016). The CL methodology does not provide prepackaged solutions “from the outside” but aims to bring out and develop in the participating group the ability to act in order to transform the activity system in which they work.

To this end, participants are actively and directly involved in the analysis, sensemaking process, and interpretation of their own activity relying on the data collected by means of individual and group interviews, microanalysis of significant and/or problematic observations, narratives, and/or video interactions (Ivaldi and Scaratti, 2020). Such a process involves several sessions and might lead to finding new solutions to problems and possibilities for organizational “expansion.” This type of intervention requires the contribution and experimental attitude of the researchers who play an active role that goes beyond the traditional perspective of a casual observer and facilitator (Engeström, 2015). In this sense, for participants to take the lead in the design and experimentation of new working practices, researchers introduce their own ideas and intentions with the aim of provoking and sustaining a cycle of expansive learning.

The role of the researcher is, therefore, to expand the conversational space during the intervention in order to support participants in examining and reflecting on the current state of their activity system, as well as in modeling, shaping, and experimenting with innovations. Researchers have to sustain dialogic sensemaking when creating socially useful knowledge, acting based on a sort of anticipational fluidity (Cunliffe and Locke, 2020) as a way of relating with others when working with differences. Such differences are experienced through the unfolding living flow within the moment of conversation.

The contribution of researchers is based on the collaborative introduction and application of new tools and ideas through “determined and systematic dialogue” (Engeström, 2015). Consequently, the dynamism of the intervention stems from the interplay of ideas and intentions between researchers and participants. During a CL, sessions are videotaped for the collection of longitudinal data to analyze the intervention process and its development, including the interactions between researchers and the participating group.

In this study, the data collected from the videotaped sessions are employed to examine the discursive practices of researchers during the intervention.

Studies carried out on intergroup relations in theories of social identity (Tajfel, 1981) and self-categorization (Turner et al., 1987) have constituted one of the richest areas of experimental social psychology. Tajfel’s theory explains how social identity influences intergroup behavior, whereas Turner’s theory of self-categorization focuses on the psychological nature of belonging to a group and on the socio-psychological basis of group behavior. Most of the empirical research carried out in this area has concentrated on intergroup behavior between large social groups (distinguished by race, nation, ethnic group, and so forth) or in experimentally created groups (using the technique of minimal paradigms; Billig and Tajfel, 1973). But what happens in small groups, characterized by multiparty interactions between members? For discursive social psychology (Billig, 1987; Cole, 1996; Edwards, 1998), social (and self) categorization processes are situated results of negotiation discursive practices occurring in interactions with others. In this view, identity is “something that people do which is embedded in some other social activity, not something they are” (Widdicombe, 1998:191).

Sacks (1992) highlighted how social identity choices and moves are both indexical (defined by the terms used to mark the belonging social categories to give salience to) and occasioned (there is a particular social context where the categories should assume some relevance). Each group member has different identities to show and to give salience to for positioning the self and the other rhetorically (Hester and Eglin, 1997). The choice between these possibilities of positioning the self and the others (Harré, 1989; Muhlhausler and Harré, 1990) is guided by social factors such as the relationship between member, their roles, and the content and scope of the group. Identity is a resource that participants are using rhetorically and strategically during social interactions, but the possibility of using different social identities to negotiate and build ingroup-outgroup categorizations is also context-related (Zucchermaglio, 2005; Fantasia et al., 2021). Social identities can become an important negotiation content between members of a group rather than a stable characteristic and an a priori of discourse in interaction. The social context itself plays an active role in allowing some possible identity choices but also in defining the access of each group member to the identity negotiation process.

Many studies have shown how, even using minimal lexical choices, speakers mark those aspects of their own (or other) social identity that they intend to present as relevant in specific interactive contexts (Drew and Heritage, 1992; Sacks, 1992; Hester and Eglin, 1997). In particular, pronouns are discursive components that, together with lexical selection (Drew, 1992), could act as a powerful “membership categorization device.”

In this theoretical and methodological framework, we focus on how the researcher, as part of an intervention process, discursively supports change in the group of healthcare professionals of a Mexican hospital, through the construction and negotiation of specific social identities. Specifically, we present both quantitative analysis and qualitative analysis of how, among the various discursive devices at hand, the researcher strategically uses the pronominal markers “we” to evoke social groups and identities that are rhetorically functional to achieve the goals of the organizational change intervention. Our aim is to provide a micro- and discursive analysis of the strength and effectiveness of organizational change interventions, usually and mostly described and analyzed by considering more general dimensions of participation.

Our preliminary analysis on the occurrences of the “we” pronoun was functional for identifying the identity categories that emerged and how often these categories were employed by participants. Table 1 provides a quantitative overview of the identity categories that emerged across the sessions and of the speakers marking them. Among the participants, the researcher results to be the one who most frequently invokes collective identities in every session (except for session n. 3, see Table 1).

The identity categories that were codified from the data and account for the “we” marked by the researchers during the sessions are as follows:

Board of directors. This “we” was codified each time the researcher referred to the group of the board of directors, i.e., area coordinators, heads of department, and hospital directors, both medical and administrative, who could be in the room during the session or not.

Incumbent researchers. This “we” was codified when the researchers referred to themselves as senior researchers in charge of the intervention and responsible for conducting the sessions.

Researchers in the room. This code states the “we” used to point out the research team in a broad sense. It was codified each time the researcher indicated the research team made up of two incumbent researchers and two junior researchers, who were present in the session and jointly were in charge of the setting organization, video cameras placement, preparation of material, and so forth.

All the people in the room. This “we” was codified each time the researcher referred to all the participants present in the session, both members of the board of directors, senior researchers, and assistants.

Participants in the shadowing technique. This group was codified when the researcher pointed out to the participants in a shadowing exercise carried out in the last two sessions (codified just once in session n. 7).

The identity categories made relevant by researchers appeared as broadly linked to the aims of the different sessions of the intervention. We focused here on session n. 1 (Table 2) and session n. 4 (Table 3) since they were particularly central for the successful accomplishment of this intervention. The specific objectives of the first session were to question the state of necessity for the intervention as well as to engage the participants. In the fourth session, the aims were centered on supporting the participants in focusing on improvement areas and proposing alternatives and innovations to improve their work practices. In the following paragraphs, we described and provided examples of how identity categories were used by researchers in these sessions.

During this meeting, the researchers presented the methodology of the intervention, describing and contextualizing its theoretical and epistemological principles. This was the first time they met with the participants and was the initial opportunity to negotiate and inquire them about the state of necessity for the intervention, as required in the formative intervention approach (Engeström et al., 1996).

Excerpt 1 (Table 2) provides an early example of identity categorization employed by researchers, where the “we” indicates the researchers conducting the intervention. There were four researchers in the room but two of them were juniors who observe and do not intervene in the session. Here the “we” refers to senior researchers who represent a university role and methodological expertise (how do we work?). This “we” is fielded to be depowered with respect to the “you” of the participants, who are presented as repositories of knowledge and agency. There is a reformulation of who has the direction of the intervention, from the researchers to the participants, therefore an attempt to propose this role to the participants in order to nominate them to an active and agentic role. The rhetorical intention is, therefore, to explain the intervention methodological approach and describe what researchers do with an informative purpose, but there is also a subtle attempt to persuade the participants by requesting them to get involved in the enterprise and to be its protagonists. Without this adhesion, in fact, the formative intervention model could not have been put into practice according to its main theoretical and methodological principles (Virkkunen and Newnham, 2013).

A different nuance of the same rhetorical intent is found in Excerpt 2 (Table 2), where Researcher A explains to the participants a theoretical concept of the intervention model (the concept of community as a component of an activity system). Researcher A exemplifies such a concept by including the category “researchers in the room” (incumbent and junior researchers) as part of the community of the hospital activity system at that moment. In doing so, the researcher establishes a link between the practitioners and the researchers as “newcomers,” offering a new and broader perspective related to the theoretical concept presented.

Excerpt 3 (Table 2) shows another rhetorical use of the groups’ categorization. In this case, the identity marker used by the researcher refers to “we” as the board of directors, a category into which the researcher puts himself (“if we are the body”). In this excerpt, when talking about “body,” the participants are expressing two meanings interrelated: body and board of directors (cuerpo and Cuerpo de Gobierno in Spanish, respectively). In this regard, the researcher solicits a reflection on the multiple relationships of the board with other organizational actors through the metaphor of the “body” but also solicits the group participation by means of asking numerous questions to the participants. The researcher asks participants a question (“I ask you”) but formulates the content as if the participants had asked it (“if we are the body”). In this way, the researcher models a reflective attitude that consists of adopting the metaphor of the “body” and, in doing so, his rhetorical purpose is to support reflection and leave the participants the role of protagonists and those who have the main agency. The researcher builds such a discursive structure by means of questions and proposals, using a sort of “ventriloquism” (Carter, 2002), to encourage participants to question themselves on certain issues. At the same time, the researcher makes an effort to not be the only one to propose topics for discussion, which would be contrary to the formative intervention methodological principles. It is possible to affirm that the changing identity category of the researcher for a moment (“I ask you, if we are the body”) is one of the ways in which this nondirective and participatory approach is discursively performed, but at the same time, it reveals a commitment and an important rhetorical work aimed at negotiating the engagement of the participants.

In Excerpt 4 (Table 2), the researcher shows a different use of the pronoun “we” referring in this case to the category “all people in the room”. There is a shift from directing the question to board members to phrasing it using a broader and more inclusive identity category (“do you feel comfortable with that definition, or do we change it, modify it?”). Here Researcher A uses the category to activate a reflection that respects the bottom-up rule and he puts himself in the group, the group of those who seek a definition for the management activity of the board of directors. The attempt consists in constructing this enterprise as shared and collaborative, in this way the researcher accomplishes two things, i.e., he reaffirms the participatory approach and at the same time involves and orients the participants in such a defining task.

During the first session, the collective identity markers were used to create engagement and influence in the group without being directive, trying to respect the bottom-up participatory principle through questions and proposals, through the identity categories shown here. These identity markers are used by the researcher to promote reflection and action, but they do so by constructing the procedural directions as shared, as something that comes from the group of participants and not from the researchers. In doing so, the identity markers are employed to serve the purposes of the session and contribute to their achievement, serving mainly on how to achieve such purposes.

The objectives of the fourth session consisted in supporting the participants to identify improvement areas and propose them alternatives and innovations conducive to the development of their work practices. In Excerpt 5 (Table 3), the identity category “incumbent researchers” is evoked to represent the researchers as the only ones with the competence to see what the object of the intervention is from a theoretical point of view (“the hospital as a whole, a collective activity”) (Engeström, 2015). These early “we’s” are in some ways a transgression from considering only the participants as repositories of knowledge, even though later the “we” of the researchers is evoked to empower the board reiterating the bottom-up participatory approach (“we want to put you in the leading role”). It seems that such an approach constrains the way researchers proceed discursively. In this regard, the use of such “we” pronouns by researchers performs a sort of compromise with respect to leaving everything completely open and in the hands of the participants. However, at the same time, the researchers tend to influence toward an outcome and move the intervention forward.

Excerpt 6 (Table 3) is still an example of the identity category “we” as “Incumbent researchers,” used in this case to support the participating group in identifying an improvement area and adopting a change proposal (“this is a model that we are proposing to you”). The rhetorical intentions linked to the category seem to continue with a negotiation process in order to launch resources conducive to change and, in doing so, stimulate participants’ transformative agency. However, the use of “we” as incumbent researchers achieves a distancing from the participants and connotes the researchers as holders of a methodological competence capable of orienting some aspects of the intervention. During the fourth session, such “we as researchers” seems less participatory than in the first session but exerts more pressure not surprisingly in such a central moment of the intervention.

In Excerpt 7 (Table 3), the researcher supports both participation and reflection by remarking on the board-of-directors category. On the other hand, the researcher puts himself as part of such an identity category to start a discussion about how the board members get involved with the local community. By considering himself as part of this group, the researcher opens up the discussion into the topic of patient inclusion, as proposed in the CL methodology (Virkkunen and Newnham, 2013) (“why not, bring a patient, let the patient tell us about a case, that we see from his point of view?”). Establishing a dialogue with the participants in such a direction seems to highlight a barrier observed by the researcher, who adopts this identity category to play a less intrusive role and support the participant group to overcome such barrier (“we call ‘breaking down the barriers’, crossing the barriers, the barrier between us and them”).

In Excerpts 6 and 7, we identified a passage from “we as researchers” (“this is a model that we are proposing to you”) to “we as board of directors” (why not, bring a patient … that we see from his point of view?). This change of identity category seems to push forward the adoption of the methodological principles that in the passage are constructed as shared and applied to the board. In Excerpt 8, the passage continues into the category “all the people in the room” that the researcher uses to support the reflection and discussion on the leadership style topic (“How to lead, as we would like than someone lead to us”).

In summary, during the fourth session, the identity markers are used to frame the focus of the intervention, propose resources conducive to change, and support the collective discussion according to the point of view of researchers. This latter rhetorical function (provide support for the collective discussion) results central to engaging the participants during the first and fourth sessions and can be considered as a specific action that allows the researchers to sustain the reflection process. In this way, researchers used such identity markers to develop the intervention and achieve its methodological goals.