The research circle: a possible utopia? Possibilities and constraints of knowledge development within a research-circle practice

ABSTRACT

This article highlights the methodology of the research circle (RC), a variation of participatory action research (PAR). The aim was to examine and discuss possibilities and constraints for knowledge development based on democratic ideals within the RC practice. An empirical example of a Swedish RC project, including fifteen teachers and two academic researchers, was analysed through the theory of practice architectures. Interconnections between the aspects of participants’ actions, conditions or practice architectures and surrounding practices emerged. How these were interlinked appears coherent but contradictory, and these interconnections seem to be both enabling and constraining for knowledge development based on democratic ideals. Space was created for experienced-based knowledge and scientific knowledge to interact. Equal relationships between the teachers and academic researchers and power asymmetries seem to have been created and re-created through different and sometimes conflicting traditions and idea(l)s of research and resources for participation. A question regarding the potential of the RC is whether it can be portrayed as a possible utopia? This article can contribute methodologically to the PAR field by visualising and discussing the prerequisites, power relations, possibilities and constraints of an RC project.

KEYWORDS:

• Research circle
• participatory action research (PAR)
• practice architectures
• knowledge development
• democratic ideals

Introduction and background

This article highlights knowledge development based on democratic values within a variation of participatory action research (PAR), the research circle (RC). Research circles (RCs) have been evolved in Nordic countries on traditions of folk enlightenment; this includes ideals of collective knowledge development and the struggle for justice, equality, recognition and empowerment (Rönnerman, Salo, and Moksnes 2008). Generally, the RC is described as a social practice, in which practitioners (e.g. teachers) and academic researchers meet to jointly and systematically investigate a question which is relevant for all participants (Olsson 2016).

The knowledge process is supposed to be open-ended in that participants initially formulate the mutual question (Andersson 2007). Overall, within the process, often referred to as a democratic knowledge process (Holmstrand and Härnsten 2022), practitioners’ experienced-based knowledge and academic researchers’ scientifically developed knowledge are meant to challenge and enrich each other. Simultaneously, researchers’ participation is, according to Holmstrand and Härnsten, supposed to strengthen the knowledge process, a process for which researchers are given special responsibility. In this regard, researchers’ competence in systematic examinations and their critical attitudes are highlighted. Altogether, the RC is intended to create knowledge that establishes a basis for changes in practices and to bridge gaps between different types of knowledge and researchers and the researched (Rönnerman and Olin 2014).

RCs can be conducted in various ways, and as part of the democratic interaction, the participants are supposed to agree on the procedures (Lindqvist 2019). In line with Freire (1972), democratic dialogues are stressed as dialogical interactions between equal individuals. In RCs, it is thus important that participants make their voices heard and contribute to other voices being heard. Different perspectives on a particular question are thereby meant to be visualised, but these perspectives are not sufficient. They must be contrasted with research, for which the participating academic researcher is seen as responsible (Holmstrand and Härnsten 2022).

Since the RC methodology follows the Nordic tradition of PAR, a collaboration between academic researchers and other participants is seen as enabling both knowledge development and the realisation of democratic values. As researchers (e.g. Forssten Seiser 2020; Nehez 2022; Salo and Rönnerman 2014) argue, collaboration between the participants as equal parties promotes emancipation. However, Kemmis et al. (2014) claims that emancipation is constrained by external researchers, due to hierarchical relationships between researchers and other participants.

RCs have been conducted in various contexts and with different participant groups. In educational contexts, this methodology is relatively new, however, in recent years, it has gained growing interest (cf. Holmstrand and Härnsten 2022). This can be seen as harmonising with researchers’ (e.g. Anderson 2017; Kemmis et al. 2014; Mahon et al. 2017) arguments for the importance of research based on democratic values when teachers’ voices are at risk of being marginalised. The scope of their voices is decreasing due to globalised, neoliberal trends that include increased external control, such as assessments of pedagogical practices, requirements for teachers regarding goal fulfilment and standardised teaching procedures (Anderson 2017).

Despite the RC’s democratic ideals, research cannot be conducted solely on participants’ intentions (Kemmis 2022; McTaggart, Nixon, and Kemmis 2018). Participants might transform their understanding of and actions in educational practices when participating in such a research project, but simultaneously they adapt and transform within educational practices and their surrounding conditions. Consequently, participants’ understanding and surrounding conditions interact and cannot be separated.

The emerging interest in the RC may have grounds other than democratic ideals. As several researchers (e.g. Nehez 2022; Somekh and Zeichner 2009; Stevenson 2014) point out, research projects may be steered from the administrative or political levels within educational systems. Teachers might be expected to become involved in enabling the implementation of reforms by finding solutions for adapting existing practices to new directives. Thus, an RC can be adapted into technical rationalities and used as a suitable tool for governance at the expense of knowledge contribution and a deepened understanding of the possibilities for change (cf. Carr and Kemmis 2005; Rönnerman 2018; Stevenson 2014).

Given this ambiguous picture of the RC, it may seem surprising that research on this methodology as such is scarce. Thus, in this current article, the RC itself is the research object when we examined it as an empirical example of RC project. The aim is to investigate and discuss the possibilities and constraints of the knowledge development based on democratic ideals. The following research question is used:

• – Through what aspects and in what ways does the knowledge development seem to have been enabled and constrained?

This article contributes methodologically to the RC in particular, and to PAR in general.

Previous studies

The increased interest in RCs in educational contexts, can be illustrated by different studies in which various objects have been explored: teachers’ and principals’ leadership (Olsson 2016; Rönnerman and Olin 2014), pedagogical documentation (Lindgren Eneflo 2014), education for multilingual children within the autism spectrum (Andersson 2007), young people’s violent behaviour (Högdin and Kjellman 2014) teachers’ roles in inclusive education (Lindqvist 2019; Siljehag 2007) and teaching and science in preschool (Elm and Nordqvist 2019; Olsson, Lindgren Eneflo, and Lindqvist 2020). Although the RC itself is not the research object in most of the studies, they illustrate that knowledge development underpinned by democratic values is not free from challenges.

Andersson (2007) shows how recurring interactions among participants enabled knowledge development and empowerment. Nevertheless, this RC project can be seen as initiated top down by an academic researcher; it did not emanate from the participants’ jointly, formulated research question but was planned based on the researcher’s interest. However, the researcher strove to invite the teachers to formulate their interests and questions regarding a particular topic.

Rönnerman and Olin (2014) stress the importance of surrounding conditions, such as time resources and composition of participants, for knowledge development, while analysing the processes within two parallel RC groups. One group consisted of teachers from different municipalities with allotted time to participate. The second group, which consisted of teachers from the same municipality, had no such time resources. The planning for both groups was identical, but their processes differed. The first group remained intact over time, and its participants deepened the knowledge of their roles as teachers by comparing different experiences from the municipalities. Participants of the second group dropped out due to difficulties in participating because of a lack of time. Because these conditions were made visible and discussed within this group, there was a basis for joint action to strengthen the teachers’ roles in their practices. According to Rönnerman and Olin, these groups’ different conditions seem to have affected both the generated knowledge and the teachers’ views of themselves as agents for change.

Although knowledge development based on democratic ideals was given positive associations by the participants in the above-mentioned studies, Högdin and Kjellman (2014), show teachers’ resistance to being part of such a process. The teachers perceived pressure to make immediate changes to their practice. Thus, they demanded already researched and functioning models. The fact that the collaborative knowledge process did not suit everyone, indicates limited opportunities for knowledge development. Also, Lindgren Eneflo (2014) reveals limitations in this regard, but on other grounds. In this RC project, the teachers had a desire to act properly in relation to authorities. This implies that the noncritical approach might have limited the perspectives on the examined question and negatively affected knowledge generation.

The increased interest in the RC; however, a lack of studies on the RC, has created a need to further investigate how participants make use of such research methodology for knowledge development. The empirical example examined in this article, is presented below.

The current RC project: background and context

This current RC project evolved from an initiative in a Swedish municipality because of the changed formulations in the Swedish Education Act (SFS 2010:800), which introduced the concept of teaching in preschools. This raised questions among preschool teachers in the municipality. They expressed a need to discuss what the concept could mean in their pedagogical practices. A representative then contacted the first author.

The concept of teaching was introduced simultaneously with an increased focus on children’s learning and knowledge in national policies (Löfdahl and Folke-Fichtelius 2015). Earlier, the special nature of Swedish preschool education, internationally known as Educare (OECD 2006), had been highlighted, in which care of children and children´s development, play and learning are seen as intertwined. This view contrasts sharply with later economic arguments about the importance of preschool education in terms of preparing ‘for demands from the international education economy’ (Persson 2010, 75, our translation). This argument is in line with global perspectives in which education is highlighted as central to creating national and international economic success (Ball, Dworkin, and Vryonides 2010). Thus, global pressures and demands influence national policymaking and the work of local professionals (Lindgard, Martino, and Rezai-Rashti 2013).

Given the intention of collaboratively discussing and exploring the teachers’ questions about the concept of teaching, in which their professional experiences were brought to the fore, the RC was found to be an appropriate approach. Even though the planning was carried out in collaboration with the municipality, the researchers formulated the overarching questions to be investigated in the RC: What could teaching in preschool mean and what could teaching not mean? However, based on the researchers’ emancipatory interest, their ambition was to jointly, in collaboration with preschool teachers, explore these questions.

The RC project started in 2017 and continued for two years. Two of the authors (academic researchers) and fifteen preschool teachers were involved.

Theoretical framework

To examine and discuss the knowledge development based on democratic ideals within the RC project, the theory of practice architectures (Kemmis 2022; Kemmis et al. 2014; Mahon et al. 2017) was used. This theory made it possible to analyse what happened within this RC and under what conditions. Such an analysis is relevant when it comes to the possibilities and constraints of knowledge development.

In line with this theory, we understand the research object of this article, the RC, as a socially situated practice (Mahon et al. 2017). Thus, the RC practice is understood as a communicative space in which its participants interact at a given site and time. Hence, this practice cannot be separated from its participants’ actions: their ‘sayings’, ‘doings’ and ‘relatings’ are considered to constitute the practice (Kemmis 2022; Kemmis et al. 2014). For example, regarding this current RC practice (i.e. the RC used), sayings may occur in the forms of the participants’ expressions, descriptions and arguments concerning the concept of teaching in preschool. Doings can refer to how the participants were placed during meetings and how they interacted through body language, sighs, laughter, etc. relatings can appear through participants’ relationships with each other and with nonhuman objects, such as scientific texts and educational policies.

These kinds of actions hang together in a purpose (Kemmis 2022; Kemmis et al. 2014), which of this RC practice can be formulated as the construction of a communicative space for shared knowledge development concerning the concept of teaching in preschool. However, how participants’ actions are interlinked is not always free from contradictions, and sometimes the purpose is not realised (Kemmis et al. 2014). Moreover, actions are influenced by earlier processes and events, within a specific practice and other surrounding practices (Kemmis 2022; Mahon et al. 2017). However, as Mahon et al. (2017) put it, the point is not that actions hang together, but how specific sets of them come to do so.

As mentioned above, actions are not shaped solely by participants’ intentions or experiences but are conditioned by circumstances beyond the involved individuals, that is, practice architectures: cultural–discursive, material–economic and social–political arrangements (Kemmis 2022; Mahon et al. 2017). These arrangements emerge in different spaces. In the semantic space, cultural–discursive arrangements enable and constrain what is relevant to say (and not to say) about a specific topic. In physical space-time, material–economic arrangements (space, time and other resources) enable and constrain participants’ doings. For example, in accordance with the traditions of the RC, meetings should be held so that everyone is giving and is given a voice to raise questions (Holmstrand and Härnsten 2022). This might affect what, how, when and by whom something can be done (or not be done). In social space, social – political arrangements enable and constrain how participants relate to and form (or do not form) bonds of solidarity and a sense of belonging to one another. That is, how participants position themselves and are positioned by others, in terms of power dynamics. In addition to site-specific conditions, a certain practice in a specific site, as the RC practice, is connected to other practices (Kemmis 2022; Kemmis et al. 2014; Mahon et al. 2017), for example, practices of pedagogical activities with children, steering and leading and research (see e.g. Rönnerman 2018). Thus, different practices are interconnected and influence each other’s practice architectures and actions (Kemmis et al. 2014).

Practice architectures prefigure but do not determine participants’ actions, as they are additionally affected by the participants as active agents (Kemmis et al. 2014). For instance, material–economic arrangements may change if participants within an RC practice extend their timeframe for meetings. Thereby, participants’ possibilities of expressing themselves might be affected by and influence cultural–discursive and social–political arrangements. As Kemmis et al. (2014) underline, practice architectures are not fixed and enduring but shift and change over time; however, practices may vary. While some appear to be reproduced through participants’ adaptation to former conditions, others may be transformed by changed actions and conditions. Whether practices are characterised by stability or changeability, they seem to be contested over time and are never neutral sites.

Although the above-described arrangements and their linked actions occur as intertwined and fluent in practice (Mahon et al. 2017), they are supposed to be analytically separatable. To grasp a certain nature of a practice, its specificities and interconnections between sayings, doings, relatings, and arrangements, need to be distinguished (Mahon et al. 2017). Since the research object in this article is the RC, as a specific practice, we understand such distinguishing as important for examining knowledge development within this current practice.

Methods

The data used in this article were generated within the current RC project and collected for two years (2017–2019). As mentioned earlier, this project involved two academic researchers and fifteen preschool teachers (in the following, referred to as ‘teachers’). During this project, 10 meetings were held. In accordance with the teachers’ wishes, the meetings took place after their regular working hours at a central location in the municipality.

This RC was shaped as a space for interactions between the teachers and the researchers in which dialogues were central. The teachers were divided into two even-sized groups to enable everyone to talk and be heard. In line with the described responsibilities of academic researchers with this methodology (Holmstrand and Härnsten 2022), the researchers acted as facilitators in each group, while being involved in dialogues with the teachers (i.e. asking questions and listening). The dialogues were mainly based on the teachers’ narratives of their everyday pedagogical activities with the children. These narratives were related to scientific texts and policy documents (see Olsson, Lindgren Eneflo, and Lindqvist 2020).

All meetings were recorded and transcribed by the researchers. The empirical material consisted of recordings (27 hours) and verbatim transcriptions (380 pages).

Ethical considerations

This current research project followed the ethical requirements of the Swedish Research Council (2017) concerning information, consent, confidentiality, and use. Participation was voluntary, and the teachers could withdraw from the project at any time.

Analysis

To answer the research question (Through what aspects and in what ways does the knowledge development seem to have been enabled and constrained?), the analysis was done in two steps:

First, we identified practice architectures in terms of cultural – discursive, material – economic and social – political arrangements (Kemmis 2022; Kemmis et al. 2014; Mahon et al. 2017). They emerged, for example, through the participants’ opportunities to talk (sayings), interact (doings), and how the participants were positioned (and positioned themselves) to each other and to national policies (relatings). Thus, interconnections of arrangements and actions (sayings, doings, relatings) were visualised. During this analysis, we noticed traces of surrounding practices.

Second, in the search for surrounding practices, we identified and categorised pedagogical practices, research practices and steering practices. Pedagogical practices refer to the participants’ experiences and perceptions of education in preschool. These practices emerged, for example, through how the teachers talked about their everyday experiences in pedagogical activities. Research practices refer to traditions for how research is conducted and the participants’ relationships to research-based knowledge. For example, these practices appeared through how the researchers designed the RC according to research traditions and how the participants commented on previous research. Steering practices refer to how participants reflected on the requirements that national authorities prescribed and mandated via policies. How these practices interconnect with the practice architectures (cultural – discursive, material – economic and social – political arrangements) is exemplified in Table 1.

Table 1. The RC practice and its interconnections between arrangements and surrounding practices.

	Pedagogical practices	Research practices	Steering practices
Cultural–discursive arrangements	What was possible to say and not say regarding teaching in preschool	What was possible to say and not say regarding how research is ‘done’ in the RC	What was possible to say and not say regarding policy documents
Material – economic arrangements	Opportunities for teachers to participate in the RC in terms of, e.g. time and place	Opportunities for researchers to participate in the RC in terms of, e.g. time and place	Educational policies introduced by national authorities
Social – political arrangements	Relationships between teachers and researchers in the RC and the preschool community	Relationships between teachers and researchers in the RC and the research community	Relationships between the participants and educational policies

In the following results sections, the nexus of the participants’ actions (sayings, doings and relatings), arrangements and surrounding practices, is presented in more detail. How the arrangements and actions are interconnected is first briefly described. In the following two sections, interconnections of actions, arrangements and surrounding practices are illustrated. We understand these interconnections as important for emphasising possibilities and constraints regarding knowledge development within the current RC practice.

Results

This RC practice evolved from collaboration between a municipality and a local university. This illustrates how material–economic arrangements enabled the initiation and maintenance of this practice and, consequently, the interactions (doings) between the participants over time. Thereby, space was shaped by and for the participants’ sayings (cultural–discursive arrangements) and relatings to other’s sayings, doings and nonhuman objects such as changed national policies (social–political arrangements). This exemplifies how arrangements and actions in terms of sayings, doings and relatings were interconnected.

How interconnections between arrangements, actions and other practices emerged, are exemplified in the following sections. As several aspects appear to be intertwined, we first concretise connections between arrangements, actions and pedagogical and research practices. Thereafter, these connections are extended through steering practices. These sections include excerpts translated by the authors . Here, the letter ‘T’ represents a teacher and ‘R’ represents a researcher. The expressions ‘T1’, ‘T2’ etc. illustrate the speech order and are not linked to certain individuals.

A web of connected aspects

The nexus of actions, arrangements, pedagogical practices and research practices can be concretised by how dialogues were usually initiated in the RC practice. They were characterised by a special order to speak, which seemed to have enabled everyone to be heard and listened to. Such dialogue during the initial meeting is exemplified below.

The participants were placed around a table that allowed them all to make eye contact. The researcher asked for the teachers’ memories of their initial working periods in a preschool. The researcher assumed that different recollections could draw attention to changes in pedagogical activities. Such an approach could promote teachers’ distance from and critical reflections on the current topic (the concept of teaching in preschool). This dialogue was initiated by the researcher’s questions: ‘When you started working in a preschool, did you talk about teaching then, as you remember it? If not, what did you talk about?’

The teachers used a whiteboard to mark the year of their starting point on a large timeline and told their stories:

T1: It was a bit like you [nodding and turning towards another teacher], say, these themes [certain content in pedagogical activities] you’re going to do: When it´s Easter you should do this, and week by week, you do that. That’s how it was planned/ … /

R: Not the concept of teaching then either … Do you recognise [turning towards another teacher]?

T2: Yes, I do, the same period, thematic approach (several teachers were laughing) … You could check last year’s calendar, so it was almost the same, and not based on children’s interests./ … /Then you end up in a preschool already shaped in a way. I think you get caught up in it, so it’s hard to come up with new [ideas].

Through this dialogue, the teachers, regardless of their scope of education and experience could talk freely; the focus was precisely on the teachers’ recollections of a certain time and not on ‘correct’ answers. The sayings were commented on and confirmed by laughing and nodding. Affirmations via body language were enabled by the participants’ placement in the room. Through this dialogue a narrative of exemplified preschool experiences over five decades was co-produced. At times, when most of the participants did not know and could relate to each other through earlier acquaintances, this narrative appeared important for the relatings of the RC practice. All the teachers could participate in equal ways when their professional experiences were sought and were the focus of this dialogue. All the formulated experiences seemed to be listened to and recognised. The interest that the storytelling seemed to arouse might have enabled the participants’ equal positions.

Despite the above-illustrated space for the sharing of teachers’ professional experiences, the design of the RC and how the topic (i.e. the concept of teaching in preschool) was approached was decided on and formulated by the researchers. However, this topic was formulated as interesting and as a reason for the teachers’ participation. In this regard, the researchers’ actions can be understood as more traditional in terms of how research is usually initiated. Thus, the researchers struggled with different kinds of research traditions and idea(l)s, which were reflected in contradictory actions and arrangements within the RC practice.

While the cultural – discursive arrangements enabled extensive communicative space, the teachers were expected to talk about teaching. In the excerpt above, the teachers’ sayings were interpreted and concluded based on what the researcher saw as essential: ’Not the concept of teaching then either’. This illustrates the role of the researcher as facilitator. Mainly, it was the researchers who interpreted, summarised, and referred sayings to what they perceived as relevant. The division of roles was not discussed, but the researchers took on this role as a given fact. Although the researchers had ambitions for enabling equal positions between the parties, academic researchers and teachers, the division of roles may, nevertheless, have promoted the researchers’ understandings at the expense of the ones of the teachers.

This asymmetrical relationship between the parties can also be related to conditions and resources for research. The funds for this current project generated only resources for the researchers, in accordance with conditions of research practices in general. Unevenly distributed time resulted in space for the researchers to plan and follow up between meetings, while the teachers were occupied by regular activities at their preschools. Such resources exemplify not easily changeable material – economic arrangements, which constrained the non-hierarchical relationship between the parties, despite the researchers’ efforts.

Nevertheless, the researchers’ struggles to navigate and possibly bridge research practices’ contradictory traditions and ide(l)s, also seem to have enabled an extensive space for sharing teachers’ professional experiences. Such a space was created by recurring dialogues based on teachers’ narratives of pedagogical activities in their preschools. In this regard, traces of pedagogical practices became significant in RC practice. However, these dialogues were usually based on the researchers’ questions, such as: ’What can be understood as teaching?’, ’What cannot be understood as teaching?’ and ‘What are the reasons for one or the other?’ Over time, the dialogues were also based on the teachers’ questions, which were usually formulated similarly, such as ‘Can this be understood as teaching’. The openness of the researchers’ questions was taken for granted by the researchers, but it proved easier to formulate what teaching could mean than to formulate what teaching could not mean. The teachers might have filled the current concept of teaching in preschool with meaning because they were expected to do so. Despite ambitions of enabling openness to various sayings, constraining cultural – discursive arrangements and the dominance of research practices at the expense of pedagogical practices emerged.

The web of connected aspects expands

In this section, illustrations of connected aspects within RC practice are expanded by adding traces of steering practices. For instance, such traces appeared in participants’ reasonings about national policies and pedagogical activities:

T1: Actually, I think it´s obvious that we use the same concept [teaching] as in school. We’re part of the same [educational system] … the [Swedish] Education Act also applies to us.

T2: And we have a national curriculum as well as the [compulsory] school.

R: Well, we have had it [the curriculum] for a longer period than the concept of teaching.

/ … /

T3: Well (several hums), I think it’s hard to use this concept in preschool; it might feel unfamiliar in some way … We’re not so comfortable with it.

T4: Teaching can be related to something more controlled. Well, it might not be like that, but that’s how you think. [Regarding] children’s learning, then we focus on the children, their interests, so they can learn.

T5: Teaching seems more controlled; the children don’t get involved and can’t exert influence … They get less opportunities to play …

As preschool professionals, the teachers, unlike the researchers, had to relate to and handle the concept of teaching, introduced by authorities via policy. Above, contradictory sayings and relatings imbued with traces of steering and pedagogical practices are illustrated: the necessity of adapting to policy versus the risks of such an adaptation. The latter can be seen as resistance to changed educational policies and positioning for traditional values of Swedish preschool education (i.e. child-centred activities in which children exert influence). In this excerpt, traces of steering and pedagogical practices also appear through ambivalence; teaching can relate to more controlled activities but not necessarily, by implication, if traditional preschool activities are given space.

In this regard, the teachers seemed to position themselves both as passive recipients of and as active agents vis-à-vis policy. Additionally, traces of research practices appeared in terms of the researcher’s questioning: ’We have had it for a longer period than the concept of teaching’. However, such questioning can be understood as contradictory in that the teachers were possibly encouraged to fill this concept with meaning. Over time, the language used by the teachers had similarities to that of the policy documents. Repeated sayings appeared to be adapted to an increased focus on fact-oriented teaching content, which could result in measurable knowledge. Such sayings might reflect subordination to the changed views of preschool education formulated via national policies. In this case, steering practices seemed to dominate pedagogical practices.

In the constructed meanings of the current topic (i.e. the concept of teaching in preschool) cultural – discursive arrangements seem to have been shaped and reshaped, which constrained critical reflections. Although the researchers can be understood as co-creators of such arrangements, they attempted to challenge assumptions that appeared obvious and taken for granted:

R: But you rarely illustrate that [norms and values regarding the concept of teaching]. Or? Have you thought anything about that? Are there any examples of when you teach about norms and values? You often talk about lettering and…

T1: It’s more tangible.

T2: We forget [to mention] care, though we don’t forget to show care [in preschool] but we don’t mention it [here].

T3: We don’t formulate it [here] … We’re required to document [in preschool] and then we often document the product and not so often the process. How to do that? It’s difficult.

T4: Democratic values

T2: It’s elusive!

R: But in relation to the concept of teaching, this matters.

T5: We carry along old, traditional views of teaching [relates to traditions of teaching within the compulsory school that differ from traditional values of how preschool activities should be carried out].

R: But you don’t want them?

T5: We want to expand them.

External control via documentation requirements of pedagogical activities seems to have influenced the teachers’ sayings and non-sayings. However, the teachers argued that aspects that were not easily formulated, such as care of children, were put into action in preschools. Nevertheless, traces of steering practices tended to become significant in RC practice.

Questioning assumptions perceived as taken for granted, however, concerned the teachers’ actions, but the researchers also acted in such ways. As mentioned earlier, the researchers assumed that certain questions created space for different sayings, but these questions did not become the subject of critical reflection. Solely, the teachers’ assumptions were challenged. This illustrates social – political arrangements that possibly reproduced the established academic researchers’ superior position vis-à-vis the teachers in the RC practice.

The fact that the teachers as preschool professionals had to relate to policies does not mean that their language use solely reflected uncritical adaptation. Over time, the space for criticism seems to have become more extensive. This critique can be illustrated by the articulated resistance to the definition of the concept of teaching in policy (SFS 2010:800) as goal-driven processes. This definition conflicted with teachers’ experiences of interacting with preschoolers:

T1: It just feels rigid to have a set goal from the beginning, which you stare at blindly. And then you miss a lot of input from the children along the way/…/Sometimes I may not know exactly where it will end up/…/

T2: Maybe the formulation [goal-driven processes] is unclear. We might get too focused on goals – just goals. These processes must be conscious processes, so we’re aware of what we’re paying attention to and why we’re acting in certain ways.

This excerpt reveals how the teachers reformulated the concept of teaching in preschool, from goal-driven to conscious processes. In doing so, policy formulations appeared as possible to adapt to the teachers’ professional experiences and their expressed, advocated ideals. A space for interpretation was created, which possibly strengthened the teachers’ agency. This positioning was embraced by the researchers and supported by scientific contributions, such as articles. Hence, traces of pedagogical practices seemed strengthened. The reformulation of the concept of teaching may also have contributed to the identification of problematic conditions in preschools and the necessity for change.

Furthermore, opportunities for change seem to have been strengthened by a clearer positioning towards the compulsory school:

T1: Adaptation [to compulsory school]: Is that how we gain higher status in preschool? Is that why the concept of teaching is needed? Should we become like them? Shouldn’t they pay attention to us instead?

T2: But then you must clarify what teaching in preschool means/…/and argue … There is a leadership team [in the municipality] discussing how the national curriculum should be implemented [in local preschools]. I’ll email one of them, so they include this question in the discussion: What does teaching in preschool mean?

Despite the preformulated topic (the concept of teaching in preschool), the RC practice can, by making conditions visible, be seen as enabling empowerment and changes within surrounding practices.

Thus far, illustrative examples have been presented to concretise the RC practice’s interconnections of actions, arrangements and surrounding practices. Overall, a multifaceted picture of the RC emerged. Influences of pedagogical, research and steering practices seem to have affected the practice architectures and actions, and the ways they seem interlinked appear as both coherent and contradictory. How different aspects seem to hang together is summarised in Table 2.

Table 2. The research circle practice’s nexus of actions, arrangements and surrounding practices.

	Pedagogical practices	Research practices	Steering practices
Cultural – discursive arrangements	Space for teachers’ professional experiences in preschool.	Dialogues in which participants were heard and listened to. Dialogues were based on the researchers’ questions.	The topic of the RC was a concept introduced in educational policy.
Sayings	Teachers’ formulated their experiences of pedagogical activities with children.	Teachers’ formulations were supported by scientific contributions. Researchers questioned teachers’ statements.	Teachers filled the topic with meanings in terms of fact-oriented content versus democratic values.
Material – economic arrangements	Between the meetings, the teachers worked in the preschools.	Research funds created time for researchers.	Educational policy was introduced by national authorities.
Doings	The teachers were less involved in planning and analysing.	The researchers summarised, analysed and planned the meetings.	The teachers’ uncritical adaptation to policy was questioned by the researchers. Teachers’ resistance was embraced.
Social – political arrangements	Relationships between the participants in terms of solidarity.	Equal versus asymmetric relationships between participants.	The teachers needed to handle educational policies. Researchers were ‘free’ of such policies.
Relatings	The teachers’ experiences were recognised.	The teachers’ taken-for-granted assumptions were questioned by the researchers. The researchers’ similar assumptions were not questioned.	The teachers positioned themselves as passive recipients of policy versus active agents.

How the RC practice appears regarding possibilities and constraints for knowledge development based on its democratic ideals, is discussed in the following.

Discussion

Through the lens of the theory of practice architectures, contradictory actions can be understandable and perhaps almost inevitable in any practice. Although the RC was created for a specific purpose (i.e. to construct a communicative space for shared knowledge development concerning the concept of teaching in preschool), the RC practice was not shaped in isolation from other practices. The analysis revealed how actions, practice architectures and other practices were interconnected. This nexus appears as both enabling and constraining for knowledge development based on democratic ideals.

In the current practice, practice architectures enabled practitioners’ experiential knowledge and researchers’ scientifically produced knowledge to interact, which is a prerequisite for and characteristic of knowledge development in an RC (Holmstrand and Härnsten 2022). Thereby, actions in terms of sayings, doings and relatings were confirmed and challenged, which is also essential for the development of knowledge (Andersson 2007; Olsson 2016).

The teachers’ actions were particularly challenged by the researchers when actions were perceived to be influenced by steering practices at the expense of pedagogical practices, namely when education-related ideas in which predefined, measurable knowledge dominates traditional values in Swedish preschool education (Alasuutari, Markström, and Vallberg-Roth 2014; Löfdahl and Folke-Fichtelius 2015; Olsson, Lindgren Eneflo, and Lindqvist 2020). Such questioning appears essential for the development of knowledge underpinned with ideals of change, criticism and emancipation (Holmstrand and Härnsten 2022). Here, the researchers’ actions can be seen as promoting teachers’ capacity for change, regarding, for example, the conditions of pedagogical activities in preschools. However, the object of questioning was precisely the actions of the teachers; those of the researchers were not questioned by either the teachers or the researchers themselves.

Within an RC practice, researchers (e.g. Andersson 2007; Holmstrand and Härnsten 2022) highlight the importance of a dialogical approach to knowledge development in which questions are posed rather than answers formulated. Although the analysis showed the researchers’ efforts in this regard, difficulties in promoting such an approach emerged. For example, the researchers’ questions seemed to have constrained knowledge development when the concept of teaching in preschool was filled with meanings that were taken for granted and easily formulated.

As mentioned above, the researchers’ actions were not questioned; thus, the space for collaborative knowledge development created within the RC practice did not seem to enable resistance to the researchers’ actions. Instead, the teachers appeared subordinate to the traditional discourse of research. For example, as pointed out in the Results section, teachers seem to have gradually adapted the researchers’ ways of posing questions. Consequently, the researchers needed to navigate different dilemmas: to be responsible for the knowledge process (Holmstrand and Härnsten 2022) while not pushing dialogues in a given direction and striving for equal relationships while different conditions for the participants to engage were created and re-created. Through unevenly distributed resources of time, the participants did not have equal opportunities for planning, leading and analysing the RC meetings. Such arrangements, which seemed to constrain non-hierarchical relationships between parties, were not changeable during this project. This was not discussed by the participants during the RC meetings.

As Mahon et al. (2017, 20) argue, power asymmetries raise questions about ‘what avenues for acting (saying, doing and relating) are opened up, and closed down, by particular power dynamics at play’. We suggest that all participants within a research practice, for example, an RC, should be involved in such questions. Such dialogues might strengthen participants’ consciousness of what kind of knowledge can be developed and thereby promote knowledge development overall.

As Somekh and Zeichner (2009) argue, ideals of PAR are in danger of being diluted if such research methodologies become subject to the implementation of authorities’ directives. This might be seen as a relevant risk for this RC project due to the close connection between its current topic and changed educational policies. However, we also want to draw attention to such risks regarding the influence of research practices. Holmstrand and Härnsten (2022, 44) argue, that academic researchers, like authorities, may position themselves as ’those who know best and know what is in need of change’ (our translation). Although researchers’ critical perspectives appear relevant to knowledge development, they cannot impose their views on practitioners (cf. Freire 2021). At the same time, researchers cannot pretend that they do not have any views. However, does a research practice create space for practitioners to question researchers’ views?

When researchers (Andersson 2007; Olsson 2016) emphasise the importance of a balance between recognising and challenging perspectives, this primarily concerns practitioners’ perspectives. However, we argue for the importance of critically scrutinising academic researchers’ perspectives and actions. Not least, according to Anderson (2017, 444), research tends to be seen as ‘under the sole control of the researcher and increasingly as a commodity used for tenure, promotion, and research grants’. Is it possible for academic researchers and practitioners, such as teachers, to speak on mutual terms and develop shared knowledge when research is one of the core concepts and the very prerequisite for PAR in general and the RC specifically?

Given the above paradox of RC practice, does the RC appear as utopian rather than a collective force for criticism, change and empowerment? In many ways, the idea(l)s of this methodology can be understood as high-flown and perhaps utopian. However, in line with Freire’s (1972) reasoning, utopian ideas may be important; they bring hope and something for which to strive. By imagining a better situation, we can explore whether it is possible to move towards such an aspiration.

According to Freire (1972), hope may be most vivid in unfavourable times when increased external control risks marginalising practitioners’ voices (Anderson 2017; Moss 2014). The RC, like other methodologies based on democratic ideals, can thus be seen as a resource for identifying and orchestrating actions in the direction of a vision in which ideals may never be fully achieved. Nevertheless, these actions are important. Perhaps, then, the RC can be perceived as a possible utopia. Such an idea does not only require resistance to the negative consequences of increased control and governance. However, in line with Anderson (2017), we stress the importance of academic researchers’ resistance to traditional, narrow ideas and conditions for research, which are contradictory to the ideals of PAR but, nonetheless, surround this field.

Limitations

Regarding the limitations of this article, this current RC practice cannot be seen as representative of such practices in general. RCs can be conducted differently and in diverse contexts (e.g. Högdin and Kjellman 2014; Holmstrand and Härnsten 2022; Lindqvist 2019; Rönnerman and Olin 2014); however, through the use of general theoretical assumptions, questions can be transferred to and discussed in various contexts and further explored. How complexities and paradoxes within research practices can be understood and handled may be such a question.

Outlook

We welcome a continued dialogue on the possibilities and constraints of the RC methodology in particular and of PAR in general. In this regard, theories, for example the theory of practice architectures, can be appropriate resources for academic researchers and practitioners who can become more aware of the prerequisites for, as well as the possibilities and limitations of, an RC, when such a project is planned and carried out. By analysing the architectures of this current RC practice, this article can methodologically contribute to the PAR field.

Acknowledgments

Many thanks to the preschool teachers in the study presented. It would not have been possible to carry out without their participation.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Skolforskningsfonden, Dalarna University [grant number du-HDA2021-00033].