The micropolitics of conflicts in total institutions – The case of special approved homes for youths in Sweden

This article focuses on conflicts between youths and staff at special approved homes in Sweden. We direct a special focus at the institutional micropolitics within which these conflicts arise and which the conflicts also contribute to form. Drawing on the work of Emerson and Messinger, our point of departure is an interactionist analysis of the micropolitics of trouble. One focal aspect in our study is the recurring patterns of conflicts – a pattern we have chosen to label the ‘conflict script’. The conflict script is a process set in motion when the staff explicitly state that they have ‘had enough’. Once started, it becomes an imperative and is therefore, in a sense, a consistent micropolitical measure. The conflict script generates immutable positions – the staff cannot back down, since their authority is at stake, and the youths know that resistance will result in the use of coercion. However, what leads to the staff having ‘had enough’ varies between interactions, which thus produces inconsistent micropolitics. The conflict script is central to understanding how trivial breaches of the rules, or other forms of disturbances, can escalate into situations that involve the use of force in the form of physical restraint and isolation.

be taken on a voluntary basis. This legislation is a so-called 'protective legislation', striving for the youths' own good. However, its coercive elements are extensive, especially for youths placed in special approved homes. This is often perceived as a kind of punishment by the youths themselves -a prevalent opinion in the interviews conducted in this study. Barker (2013) has argued that in order to understand the penal regime in the Swedish context, and its inclusion of both mild and repressive elements, it is necessary to consider not only the criminal justice system but what happens in different areas of the state's exercise of control in relation to citizens. One factor of special significance for the system as a whole is what happens, in parallel with criminal justice, in the area of the social services. The control exercised here can be every bit as invasive in relation to the individual as the sanctions imposed by the penal system. The same is true of the rehabilitative elements found in both prisons and youth institutions (c.f. Crewe, 2011a;Scharff Smith and Ugelvik, 2017). The social services area, with its focus on producing change in young people, is of particular importance for the institutions on which this study is based, and the power that is exercised may be compared to what Crewe (2011b) has termed soft power, with a focus on bringing about self-governance in the youths via the opportunities that staff have to exercise discretion.

Special approved homes as total institutions
Perhaps the most important distinguishing feature of the total institution is the confinement to which inmates are subjected, which is symbolised by the (often physical) barriers that separate inmates from a social life outside the institution (Goffman, 1961). Total institutions involve a loss of social roles, depriving individuals of 'certain stable social arrangements in [their] home world' (Goffman, 1961: 14). The youths admitted to special approved homes are expected to realise that they need to change in line with what staff and social workers have identified as desirable behaviour. This can be described as a power struggle regarding the youths' identity, where the youths are instructed to 'self-direct themselves' in a way that the staff view as 'manageable'. This kind of order requires that both desirable and undesirable behaviours are defined and recognised by means of rewards and punishments (Enell, 2017;Goffman, 1961;Liebling, 2008).
Rules thus assume a central role in the organisation of the total institution, and there appears to be no limit to which aspects of life in a special approved home that may be subject to regulation (Franzén and Holmqvist, 2014). In order to live up to institutional demands for adaptation, inmates adopt different strategies. This, in turn, contributes to a situation where such institutions rarely achieve their rehabilitative goals (Franzén and Holmqvist, 2014: 71). In the context of special approved homes, Pettersson (2017: 88f) has described a process of superficial adaptation, whereby youths adapt themselves to institutional rules and conditions rather than undergoing any 'real change'.

Social control as institutional micropolitics
The constant surveillance that characterises total institutions means that subtle shifts in the youths' behaviour can produce a sense among the staff that something is 'off'. This may result in irritation regarding a certain youth's behaviour, but it is only when agents of social control -in this case staff members -react to the problem that it crystallises and becomes clearly defined. The problem is thus intimately linked to, and formed by, the measures that define the problem. As Emerson and Messinger (1977: 123) put it: 'naming something a problem has implications, prefiguring some solutions and removing others'. These institutional micropolitics provide a basis for a more complex and profound framework for interpreting conflicts and social control in total institutions. This framework emphasises the significance of power, influence, control and above all human interaction.
A review of the research on conflict management in youth care (Wästerfors, 2009b) discusses a number of ethnographies that have analysed social control at such institutions in terms of micropolitics. One important factor that this research emphasises is the way that the strict and allencompassing rule system becomes an object of constant negotiation between youths and staff (Franzén and Holmqvist, 2014;Kivett and Warren, 2002;Price, 2005). It follows that social control in these institutions is in a constant state of change and renegotiation and that the same 'problem' may result in very different responses. Wästerfors (2009a: 48) found that refusing to clean one's room resulted in a warning from the staff that the youth's planned home visit might be cancelled (cf. Goffman, 1961: 13). In another case, the staff instead help the youth to clean the room. Staff may ignore rule breaking in order to avoid a situation escalating or might show indulgence when an otherwise well-behaved youth breaks the rules (Kivett and Warren, 2002;Price, 2005). Thus, the way a 'problem' is dealt with, and thereby defined, is dependent on a number of factors: who is causing the problem, the nature of the relationship between this individual and the staff and the individual's prior history of troublesome behaviour (Emerson and Messinger, 1977: 125;Wästerfors, 2009b).
Directing a focus at conflicts on the basis of the micropolitics of trouble means that the interactions in the institution, the institutional context and the actions of all its participants (including the audience in the form of other staff members and inmates) become part of the analysis and the understanding of these conflicts: whether they escalate or de-escalate or whether an incident is even defined as a conflict in a given situation (and not in another). This perspective on conflicts fits well with Goffman's interactionist perspective regarding what happens in the context of life in a total institution. Both perspectives view power, which is often unequally distributed but not entirely determined by different positions, as a central part of the interaction. A further contribution made by this perspective is that definitions of both those involved in the conflict and the conflict itself may vary and shift over time. One example is that if a measure taken by someone with power to govern the definition of the situation is clearly directed at one of the participants in the conflict, such as when the staff choose to isolate one of the youths involved in a conflict, the other party is transformed into the victim and it becomes clear to everyone involved who constituted the 'problem' (Emerson and Messinger, 1977: 121).

The concept of the conflict script
An interactionist perspective is characterised by the view that human beings are active agents. Our actions are based on the meanings they have for us, meanings that are constantly being negotiated, as they are reinterpreted and constructed in interactions with other people in concrete situations (Blumer, 1969). Something that is often emphasised in this perspective is the way in which social interactions (e.g. conflicts) tend to follow a pattern, which Goffman (1983) among others has focused on via the concept of interaction order. These behavioural regularities may be described in terms of a script, which Barley and Tolbert (1997: 98) define as 'observable, recurrent activities and patterns of interaction characteristic of a particular setting', with the setting in this case being that of the total institution. The concept of the conflict script regards recurring situations in which the staff have 'had enough' and issue a form of ultimatum: 'go to your room'. The command initiates a process where both the staff and the youths become constrained in their alternatives of action -every form of resistance from the youth following the command will be met with increased force from the staff. Situations escalate to physical force more commonly at units with newly placed youths, which is in line with social psychology research (c.f. Schank and Abelson, 1977). This indicates that the youths learn the consequences of the conflict script in time and instead choose to go to their room of their own accord to ease the consequences. On the surface, this can be interpreted as the youths themselves actively de-escalating the situation. However, given the ultimatum the conflict script entails, the youth has no other choice whether she/he wants to avoid being subject to physical force. The staff's alternatives are similarly constrained, because they need to carry out the conflict script to maintain their authority.

Setting, data and method The setting
Youths who are placed in special approved homes are among the most vulnerable young people in Sweden, and in comparison with a reference population of youths of the same age, they present significantly higher levels of problems across a range of areas (Ybrandt and Nordqvist, 2015). Youths may be placed in order to conduct an assessment that will result in a treatment plan. Alternatively, youths may be given what are referred to as acute placements, while waiting for further care in a special approved home or in a non-institutional context. The majority of youths in special approved homes are subject to acute placements. The different types of placement involve youths being assigned to different units, and the study includes all forms of placements and unit.
The youths at special approved homes reside in relatively small units, usually comprising six to eight youths, with high staffing levels. The youths have their own rooms and there are common spaces where they can spend time, for example, watching TV, playing games and eating. Rules relating to where the youths can spend their time, at what hours, when they must go to bed and get up in the morning, if and when they can play computer games, watch TV and so on vary both between different institutions and between different units in the same institution.
The youths included in this study are still required to participate in compulsory schooling. Schooling is provided at the special approved homes but takes place on other premises than those in which the youths live. During the time that is not devoted to schooling, the youths participate in treatment and in possible organised leisure time activities. The youths can, in restricted forms, spend time outside the institution and may be given leave to visit their families. The majority of their time, however, is spent at the institution, and in this sense, their everyday lives are primarily characterised by the conditions that prevail at a total institution (Goffman, 1961).

Data
This article constitutes part of a larger research project. Its objective is to study the use of a tool for conflict management, in the compulsory institutional care of youths and adult substance abusers. The material comprises qualitative interviews with youths/clients and staff at the relevant institutions. The material employed in this article is restricted to two institutions that are responsible for the youngest youths placed in youth care institutions, one for girls and one for boys. The interviewed youths were aged between 12 and 16, while the staff were between 28 years and 60 years of age. 2 In total, 18 youths (9 girls and 9 boys) and 9 staff members (4 women and 4 men) were interviewed at the two institutions. The institutions include both locked units and open treatment units. While the doors of the open units are still locked, those living in such units have more opportunities to spend time outside the unit. Both types of unit, from both the boys' and the girls' institution, are represented in the interviews by both youths and staff.
The interviews were conducted during the autumn of 2017 and may be described as semistructured. They were concentrated around three different themes: conflicts, the composition of the groups of staff and youths (inter alia gender, age, birth country) and the tool for conflict management. Conflicts are central to all the themes. The interviews were relatively open, and the interview participants were given substantial opportunities to guide the focus of the interview. They were encouraged to talk about concrete situations that they had experienced. Although all of the themes were discussed in the interviews, the interviews varied in terms of both their length (20-80 min) and their content.
The sampling strategy means that the study's results cannot be regarded as being representative for all special approved homes in Sweden. At the same time, it is possible to talk about theoretical generalisation, whereby the material is generalised to different theoretical ideas and concepts (Goodman et al., 2005: 10). In this sense, the results also hold for conflicts at other institutions and also for youths living in secure institutions in other countries. This is the case because total institutions of the type in this article have several common features, which are not dependent on either the age of those living in the institutions or the countries in which they exist.
Finally, a few words about interview data in the study of conflicts at an institution. Most of the literature referred to in this article involves ethnographies, although interviews might be part of the data employed in these. We will return to these different approaches in the concluding discussion. Here, we discuss the pros and cons of interview data. One of the advantages is that the process of data collection is significantly more efficient, especially in studies with extensive samples. However, a disadvantage is that the researcher loses the opportunity to witness the interactions themselves. What we capture in the interviews is therefore narratives of conflicts rather than the actual conflicts: or, in this case, narratives about interactions that lead to conflicts or that describe how conflicts develop and are managed. One potential issue is that this can lead to socially desirable narratives and to some extent it does.
While the narratives in interviews cannot be perceived as exact accounts of an interaction, this notion does not rule out relevant analyses and results. For example, Sandberg (2010: 452) argues that '(n)arratives are important not because they are true records of what happened, but because they influence behaviour in the future'. A similar approach is described by Czarniawska (1998: 20), who states that '(o)rganizational narratives are both inscriptions of past performances and scripts and staging instructions for future performances'. As such, the interviews are important for understanding both what has happened and how people might act in the future. In this sense, the identification of the conflict script becomes important knowledge regarding future conflicts at the studied institutions. Interviews therefore contribute with important knowledge regarding conflicts at institutions. However, these need to be analysed as narratives of these conflicts, which is our approach to them.

Analysis
The initial coding was broad and had the principal focus of sorting the material on the basis of different types of situation as well as reducing the amount of data to be analysed. Given the study's interactionist point of departure, the coding focused on narratives about concrete situations of human interaction. Conflicts between youths and staff, conflicts between youths, instances of being isolated and of the staff use of physical restraints constituted preliminary themes. Therefore, there was an analytical focus on 'what' was stated in the interviews. Additionally, we have been influenced by Gubrium and Holstein (1997: 164), taking an interest in 'how' it was stated. In a discussion of interview studies in an interactional framework, Järvinen argues that researchers should not lose themselves in 'how' stories are told but rather strive for including this dimension as a part of the analysis of 'what' is stated. In this sense, an interactionist analysis of interviews does not omit the actual interview situation -while, at the same time, 'what' is stated can still be given the highest priority (c.f. Wästerfors, 2007: 54).
The results of the analysis hold true for both locked and open units. No systematic differences between them were found that would motivate an analysis that distinguished between the two. We have, however, been observant of potential differences between the types of units, and this is mentioned in the analysis. Unless otherwise stated, the quotations presented in the text constitute representative exemplifications of the material as a whole.

Findings
Conflicts between youths are significantly less common than the almost daily conflicts between the controlling staff and the controlled youths -and on this issue the youths and the staff were very much in agreement, although they identified the causes of these conflicts as being located in the 'other' group. The described conflicts are relatively often about trivial rule breaking and other deviations from the social order that the staff are seeking to maintain. Social control at the special approved homes is uniform, recurrent and consistent on the one hand, and unpredictable, variable and constantly under negotiation on the other. In the context of the ongoing work of social control, the youths both allow themselves to be subdued and offer resistance. Both alternatives express themselves in varying adaptation strategies. These phenomena will be analysed on the basis of three themes. The thematisation is cumulative in the sense that the first theme, consistent micropolitics, contributes to an understanding of the second theme, inconsistent micropolitics. Together these two themes present a picture of the social control that the youths are met with, while the final theme directs a focus at the youths' adaptation strategies. 3

Consistent micropolitics
On one level, social control at the special approved homes is characterised by extreme, almost mechanical precision. This is the case in situations where the staff describe that they have ''had enough'' and when what we have labelled the conflict script is set in motion. The conflict script is an informal process, comprised of a series of predefined steps that continue as long as a youth continues to resist. The first step is that when the staff have had enough, the youth must go to his or her room. If they do not do so of their own accord, the following steps come into play: (1) being forcibly transported to their room, (2) physical restraint followed by being transported to their room or to isolation or (3) first being transported to their room, and subsequently to the isolation cell. 4 The exact point where staff have 'had enough' and choose to implement the conflict script varies a great deal. However, once it has been set in motion the dynamics of the interaction between the youth and the staff change radically, and the alternatives for action are predefined for both parties -a process described as escalation: the more the youth resists, the greater the level of coercion from the staff. This means, however, that the youth is able to avoid being subjected to additional coercion by choosing to give in and no longer offer resistance, which is a significant factor in the power dynamic between the staff and the youths. The different steps in this process were described by Viggo in the following way (the emphasis is original): Viggo SG: Otherwise you usually say, you can go to your room and calm down. Then we can talk about it, and if that doesn't work and so on, then you have to like, and if they keep going, then you have to help them of course. And if that doesn't work, then it can result in an isolation. It's like, you know we try to treat them with as much low arousal as possible, but sometimes it doesn't work. Or when they're really riled up, then it doesn't work.
A number of elements for an understanding of the conflict script can be seen in Viggo's narrative: the escalation when the youths do not do as they have been told, the responsibility for this being ascribed to the youths, while the involvement (and power) of the staff is concealed by the description of their using 'as much low arousal as possible'. The staff are also described as being consistent and uniform in their actions.
In the staff's narratives, maintaining the social order in the 'environment', as they refer to it, is of central importance. This is a prerequisite for the youths changing in line with what the staff have identified as desirable behaviour. In the context of this work, it is important to define and act against deviance in relation to the social order such as various forms of rule breaking. For example, youths who are loud, violent or cause a disturbance in other ways are separated from the rest of the group (cf. Goffman, 1961: 87). It is to this end that the staff implement the conflict script. Ebba provides an exemplification of the process in the following quote Ebba SG: They [the youths] go round and are unpleasant, hit the walls, spread quite a lot of negativity on the unit, say bad things and that kind of stuff. And then it might be that, then we make the judgement that, no, this youth cannot be out on the unit just now. Because it is affecting the others too much. And so you say ''now you should go to your room, and we can have a talk there'' and if they don't want to go in, then we have to, as we might say, help them in and maybe take hold, I mean not a restraint hold, but nonetheless like this, ''come on now'', and so you might take hold of their arm and walk. And that is a situation that arises really often, that it then becomes a physical conflict, because then they struggle and get angry.
Ebba's description is reminiscent of Viggo's but lacks the final stage of the escalation process: isolation. The alternatives available to the youths are greatly restricted, and since leaving the situation is equivalent to obeying the staff, it provides no chance of 'saving face' (Goffman, 1961). When a youth is separated from the rest of the group, it is also made clear to everyone present who constitutes the 'problem' (cf. Emerson and Messinger, 1977: 121). Irrespective of the degree of physical force employed, the staff are 'helping' (cf. Viggo) the youth in question to their room, or to the isolation cell, which may be seen as an expression of the way punishments are 'phrased in a language that reflects the legitimated objectives of the [total] institution' (Goffman, 1961: 85). From the staff's perspective, the first step of sending the youth to their room is about giving the youth a 'chance' to do the right thing, and if the situation escalates into violence, it is the youth who bears the responsibility. Given the power (im)balance that is in play, there is no doubt about how the conflict will be defined and described in cases where the youth offers resistance (cf. Emerson and Messinger, 1977). The youths' narratives stand in stark contrast to the staff's description of themselves as using a low arousal approach and that it is the youths who 'govern' the conflicts. Rather, the youths argue that it is the staff's actions that escalate conflicts. Theo even wondered whether: 'It was intended that they should work that way' and he argued that 'here it's the rule rather than the exception that the staff rile youths up fairly hard'. One reason for the rigidity of the conflict script is that the authority of the staff is dependent on their following the conflict script as far as necessary. Backing down is not an option. In this sense, the way a situation develops is determined by the youth's actions, but the positions are locked on the basis of conditions determined by the staff. If the conflict script is not followed as far as necessary, it is not only the social order that is threatened but also the self-image of the staff as fair and consistent (cf. Goffman, 1961: 7) -something that it is important for the staff to maintain, particularly when the audience is comprised of other youths (cf. Wästerfors, 2009a: 82). The staff, however, often motivate the use of these measures by reference to security issues. The following statement from Sonja provides an example of this, and also illustrates the staff's need to appear consistent and to never back down.
Sonja SB: We actually have the mandate to act, but we cannot be passive because it is dangerous. Because it's not good and there are a lot of gang boys and they sense it immediately, that ''wow, now I've got a real opportunity here.'' When it comes to the boys, it is often constructed as a question of the safety of the staff and the other youths, whereas the girls need to be protected from themselves (cf. Laanemets and Kristiansen, 2008: 90f). Unlike the boys, the girls are not viewed as constituting a physical threat. The youth's gender, together with age, also appears to be of significance for the staff's readiness to set the conflict script in motion. Ebba (SG) had previously worked at a unit for boys of postcompulsory school age (usually 16-19 years) and she discusses these differences in the quote presented below. This type of comparison is rare in the interview material, and there is no corresponding narrative in the interview data from other staff members. The quote nonetheless captures the differences in the way control is exercised in relation to boys and girls that emerge in the narratives.
Ebba SG: I felt that the staff there [the institution for boys] adopted a low arousal approach a little more than we do here. I don't know if it's because of the target group, or if it's due to management or so on, but. There [the staff] waited longer, you didn't step in as quickly and I have also questioned that here, and in a comparison between a big 19-year-old guy and a small 14-year-old girl [ . . . ] It's easier of course simply in physical terms to step in physically and take hold of a little girl. Because there's less physical resistance. I don't think that you're as, you hesitate more before stepping in and taking hold of a big 19-year-old boy.
Previous research on the significance of gender at special approved homes has emphasised the gendered structures that characterise these institutions (cf. Pettersson, 2014;Laanemets and Kristiansen, 2008). Boys are given more room to manoeuvre, and behaviour that is violent or problematic in other ways is viewed as an undesirable extension of an otherwise acceptable masculinity. Girls, on the other hand, are viewed as deviant when they act in breach of normative femininity. Therefore, they are not allowed to act in a way that strains norms or limits to the same extent as the boys (cf. Bosworth and Carrabine, 2001;Ericsson and Jon, 2006). As a result, the threshold for the type of behaviour that can and should be corrected is considerably lower for girls, and rather than reducing this to a question of the girls offering 'less physical resistance', it may be viewed as constituting part of the work to re-socialise these girls into respectable femininity (cf. Vogel, 2012: 32). If a youth goes to their room 'of their own accord', this is regarded as a sign of progress in the treatment process, since in doing so, the youth is acknowledging the staff's definition of what constitutes 'correct behaviour' (cf. Enell, 2017). In cases where the situation escalates, the youth's resistance instead confirms the need to segregate and remove the youth from the 'environment' in order to preserve order and security. While the staff try to avoid segregation, the conflict script makes this difficult since it contributes to producing a situation where the positions of both parties become immutable. This, in turn, increases the risk that the situation will escalate. Irrespective of this dilemma for the staff, the social order is given the highest priority and the conflict script is applied when the staff perceive that this order is threatened -in this regard, the micropolitics of the staff are consistent. On the basis of the immutable positions of the conflict script, it becomes possible to understand why youths and staff who are asked to describe violent situations and the imposition of isolation often begin by talking about bickering over sugar cubes in the coffee, time for lights out and other things that would appear quite trivial to an outsider.

Inconsistent micropolitics
In this section, we move away from the rigidity of the conflict script and the immutable positions that it gives rise to and instead shift the focus towards more common situations. These often lead to mere situational measures on the part of the staff -or they are treated with indulgence or are completely ignored (cf. Kivett and Warren, 2002: 6;Price, 2005: 47). In this respect, the social control at special approved homes may be characterised as unpredictable, variable and under constant negotiation in the context of human interactions. It is micropolitically inconsistent. The staff often describe themselves as a fairly close-knit group. They are 'all agreed', 'back each other up' and emphasise the importance of upholding rules and being consistent (cf. Goffman, 1961). Conflicts arise if the staff fail to 'maintain a consistent line'. If the staff fail to do so, this is regarded as a result of conditions related to the composition of the group of youths, for example, that the 'wrong target group' has been placed in the institution (cf. Wästerfors, 2009a: 85f). However, sometimes individual staff members may be identified as not acting in line with the collective rules or as being too rigid.
Rules are central to the work of maintaining the social order (Goffman, 1961: 87). Belief in the rules is more or less essential, and all forms of disorder may be met with demands for more and clearer rules (cf. Franzén and Holmqvist, 2014). Viggo (SG), for example, said that the youths 'need clear rules and structures that are easy to understand'. For their part, the youths more or less unanimously described a rule system characterised not by uniformity and predictability but rather by caprice and arbitrariness (cf. Price, 2005: 46). The unpredictable application of the rule system gives rise to conflicts between youths and staff, while at the same time also functioning as a means for the youths to divide the staff into those who are 'good', who can turn a blind eye, and those who are 'hard', who stand for a more rigid application of the rule system (cf. Goffman, 1961) (the emphasis is original).
Astrid: They [good staff members] aren't as hard about following some of the less important rules, I mean some that maybe shouldn't even be rules in the first place. If someone trusts you when you have a good relationship, so to speak. I'm not really allowed to be in the kitchen. But on occasion I might be allowed to go in and wash my hands when they're baking or something. I mean small things like that, which don't matter, because, you know, they trust me.
In this case, flexibility regarding the rules was interpreted as a sign of a 'good relationship'. It confirms a closeness to and trust for the youth, which stands in contrast to staff who are formal, rule-governed and inflexible. The quotation also illustrates the limitations that the youths must relate to as a result of the extensive regulation of their lives in the special approved homes. At the staff's discretion, youths might be allowed to wash their hands in the kitchen. Staff members may even consciously (or unconsciously) ignore rule breaking in order to avoid a situation escalating, an 'easy out' to borrow the term used by Kivett and Warren (2002: 27). However, doing so requires that either other staff members are not present or they hold a collective line. Otherwise, as was explained by Manuela, this can lead to conflicts.
Manuela: I and a few other youths were sitting and eating and one of the staff was sitting there too. So we talked about a few things, and then I told them what had happened at my social worker meeting. The second talked about the same thing, and the third talked about her dad. The staff member sat there, he didn't say anything. He didn't make any comments at all, you know. I mean, ''Don't talk about that'' or anything. Then his phone rang. At the exact moment that his phone rang, another staff member came in, and this was Kajsa. She just, ''You can't talk about that kind of thing''. So I said, ''I mean, there was a staff member sitting here, we didn't say anything wrong, he didn't say anything. Why should you come in and make a load of comments?'' I mean, she's too strict. So she made it into something really big. 5 According to Manuela, the situation escalated and resulted in one of the youths, who had shown 'attitude' towards Kajsa, in addition to being sent to her room (in accordance with the conflict script), having a forthcoming activity (outside the institution) rescinded. This punishment may be understood against the background that privileges linked to the outside world tend to be used as a strategic means of exerting pressure to produce change in the inmates of total institutions (Pettersson, 2017;Goffman, 1961;Wästerfors, 2009a). When Manuela and the other youths described 'hard' staff members, they often spoke about how the actions of these staff members escalate situations, making trivial deviations from the rules into 'something really big'. At the same time, 'really big' things tend to involve setting the conflict script in motion, which locks the positions of both the staff and the youth. At this point, the staff ascribe responsibility for the way in which the situation develops to the youth -the greater the degree of resistance from the youth, the greater the use of coercion by the staff. What may start as a trivial breach of the rules thus easily becomes transformed into a further expression of the youths' pre-existing problems. If the staff instead show indulgence, this can serve to confirm the existence of an emotional intimacy between the youth and the staff (cf. Astrid).
This illustrates the complexity of the ways conflicts are handled, talked about and thus also defined (Emerson and Messinger, 1977: 123). The measures used by the staff are central to what happens next, but Astrid's and Manuela's stories tell us that the social control employed at special approved homes cannot be reduced to unidirectional measures whose objective is to maintain an absolute system of rules (Wästerfors, 2009a: 115). Instead, the rule system is rather variable, unpredictable and dependent on who is causing the problem, what kind of relationship this individual has with the staff who are looking to resolve the issue and also what kind of history of troublesome behaviour the youth in question has (Emerson and Messinger, 1977). In other words, the social control at special approved homes is micropolitically inconsistent. The conflict script represents an exception in relation to this order, but even here, what is (micropolitically) consistent is restricted to how the staff apply the conflict script. What may precede the staff having 'had enough' varies a great deal, as can be seen from Manuela's narrative. Following the rigorous rule system of the total institution requires constant effort on the part of the inmate (Goffman, 1961: 42). Having to relate to an unpredictable rule system does not make this task any easier. While inconsistent micropolitics provide the youths (and the staff) with certain benefits, assessing which problems will be ignored, indulged or subject to disciplinary measures is no easy task (cf. Crewe, 2011a: 513;Sparks et al., 1996). As Astrid explains, 'you know, they don't talk to each other about what they have agreed, but rather they're so different, and that makes a lot of [youths] really angry'.
In the staff's narratives, it is rather about rational security work instead. This is characterised by daily risk assessments of the youths and a sensitivity to deviations in 'the environment'. In the youths' narratives, however, the staff's security work is described as a latent cause of irritation, which manifests itself as soon as the staff take security measures that the youths perceive as being unnecessary and provocative (cf. Goffman, 1961: 66).
Theo: If I've had a discussion with the staff or when other staff notice that two people, I mean have a discussion for more than like three minutes then the staff start to flock around you in order to be prepared to jump on you if you like get angry. And it's also that you ask the staff that are around you the question, ''What are you doing, why are you standing here and waiting for me to get angry?'' And then they say, ''No, we're not standing here waiting, we're just standing here because we feel like it.'' And then the staff tease you, ''Yeah, it's a free country, we can stand where we want.'' In previous research, responding with teasing and humour has, among other things, been described as a successful way for staff to maintain social order without having to resort to coercion (Kivett and Warren, 2002: 22f;Wästerfors, 2009a: 101). The staff in our data also describe a humorous jargon, but they were careful to point out that there are limits -which differ from one staff member to another: Viggo: We [the staff] are all different, how you might take a joke or the way we talk to each other or that kind of thing. And I think the youths usually learn this, and they learn it quite quickly. We think we are quite clear, like, ''No, now you've gone too far for me.'' This statement expresses a form of double standard, where the youths are expected to be sensitive, perceptive and understand different forms of jargon on the one hand. On the other hand, the youths are not uncommonly described in dysfunctional terms, which requires consistency and a collective line on the part of the staff. In other words, the social control exercised at these institutions is variable, unpredictable and negotiated in the context of human interaction (Kivett and Warren, 2002;Price, 2005;Wästerfors, 2009a). In order to deal with this micropolitical inconsistency and to avoid the conflict script process being initiated, the youths need to adapt. Moreover, the institutional social control is adapted to the resistance it meets, which makes the youths find new ways of expressing this resistance. In this sense, control and resistance are tightly interwoven and in a state of constant negotiation and change (Pettersson, 2017: 23;Ugelvik, 2014: 44). In the context of the final theme, we discuss the adaptation strategies described by the youths and the resistance they offer to the limitations imposed by the staff and the institutions.

Adaption strategies and (in)consistent micropolitics
The conflict script constitutes a central aspect of the social control at these institutions. It prescribes not only the staff's actions but also those of the youths. The youths are thus painfully aware of the essence of the conflict script: the greater the level of resistance, the greater the level of coercion. This is a reality the youths find difficult to relate to, and they expressed a sense of impotence and resignation in relation to the way the script is constructed: Filip: I mean, in places like this, you have no power. I think you should be able to discuss things more. I mean, without them using force and that kind of thing. Because that's like what I often, I like, I mean I question the rules and that's nothing strange. I'm in a new place, and I'm just ''Why should we do this?'' And then I've also had teachers and everyone tell me that ''You should question something if there is something you don't understand'', and then when I do question something, then they get all provoked and tell you to ''Go to your room'' and so on.
According to Filip, the critical thinking he is taught in school is not compatible with the demands for adaptation that characterise the total institution. Instead, a more important lesson is understanding the construction of the conflict script. The youths' internalisation of the essence of the conflict script tends to be taken as an indication of successful treatment work, and youths who follow the rules and succeed in avoiding being subject to isolation and physical restraint, by going to their room as soon as they are ordered to do so by the staff, are regarded as more 'well-functioning' (cf. Rowe, 2011: 584). However, in the youths' narratives, this appears to be a case of what previous research has labelled superficial adaptation (Pettersson, 2017: 88f). This is a way for the youths to protect themselves -to retain a self-image of who they are and who they want to be, while simultaneously avoiding negative sanctions (Goffman, 1961). One of the tasks of the total institution is to change people. Theo had just said that 'in a way you get broken down here': Tove: Broken down, how do you mean? Theo: Yes, mentally broken down. Tove: How does that express itself, what happens when you become mentally broken down? Theo: Yeah, you know, you get a completely different way of thinking and living. You successively forget the world as it was before you came here, and once you've come here it gets like, I mean one way, for example, is that the staff hold youths down when they get aggressive, and don't just let them do what they want. That's probably one way I can imagine. The isolation cell, you know a lot move first to [Unit name] before they come here. And that's also another way. They sit in one and the same room the whole time. If they say, or I mean, if they swear one single time there on [Unit name], they isolate them, so they have to sit in their room the entire day. So like in the end, it gets so that you do every little thing the staff say without any kind of protest. That's probably kind of the idea, I can imagine.
Theo's narrative is one among many in the material that illustrates how resistance and adaptation are an ongoing process at these institutions, and how the staffs' power causes youths who want a relatively quiet life, and to avoid being sent to their rooms and so on to have to learn to comply with the staff. The meaningful aspect of this story is not whether the youths actually become isolated after using a single curse word. Nor is it central whether Theo believes this himself or whether he is exaggerating. Instead, his story functions as a powerful way of illustrating to the interviewer how the staff break down the youth, particularly at the unit the majority of youths are first placed at. The staff are careful to emphasise the importance of having follow-up conversations with the youths following a conflict, particularly in cases when the conflict script has been applied, but also in connection with less serious incidents. For the staff, it appears to be about having an opportunity to justify their actions, while the youths are able to express their feelings, or as Wästerfors (2009a: 42, our translation) has put it, 'the conversation is about the student admitting to his error and promising to improve'. Thus, the youths are aware of what is expected of them in the context of this type of conversation, and they say what the staff want to hear in order to avoid negative consequences (cf. Pettersson, 2017). Superficial adaptation also constitutes a problem in relation to the other youths, particularly in conflict situations: The dilemma expressed by Filip in relation to his role as spectator may be understood in light of the cultural expectations of group loyalty that characterise the relationships among inmates of total institutions (Goffman, 1961;Maier and Ricciardelli, 2019: 244). When the conflict script is applied in relation to a youth, a form of cultural pressure is exerted on the other youths to show their disapproval, in order to maintain this group loyalty; a youth otherwise risks 'being frozen out by the youths'. Besides the ever-present risk of having the conflict script applied to oneself, the chances of one's protests being listened to and actually being able to affect the situation are almost zero according to the youths (cf. Emerson and Messinger, 1977: 121).
However, the youths should not be understood as being completely disciplined. If they were, there would be no conflicts. The youths who devote themselves to superficial adaptation also describe situations involving resistance and conflict. Thus, the conflict script constitutes a powerful tool for the staff, and since the youths have internalised its essence, the staff may be viewed as to some extent having succeeded with the principal goal of the total institution: that of getting the inmates to 'self-direct themselves' in a way that is viewed as 'manageable' by the staff (Goffman, 1961: 87). In the following concluding discussion, we will discuss important consequences of this, and the relevance and limitations of the results for similar institutions, in different contexts.

Concluding discussion
Previous research has noted incarcerated youths' perceptions of powerlessness in relation to staff (see, e.g. Pettersson, 2017;Kivett and Warren, 2002). This is also one of a number of distinguishing characteristics of the total institution (Goffman, 1961). The conflict script and the way it forms social control provides an additional piece of the puzzle in our understanding of this powerlessness. The conflict script generates immutable positions -the staff cannot back down, since their authority is at stake, and the youth know that resistance will result in the use of coercion. The conflict script is thus central to an understanding of how trivial breaches of the rules, or other forms of disturbance, can escalate into situations that involve the use of force in the form of physical restraint and isolation. The conflict script also forms how inmates (superficially) adapt to the social control that is exercised. It is applied at both open and locked units and in relation to both girls and boys. At the same time, there are differences, for example, between girls and boys, in relation to when the staff start to apply it. Although the conflict script has relevance for both the open and the locked units, the youths' narratives imply that it is applied earlier and for less serious rule breaches at the acute units where the youths are often placed initially. The alternatives to negotiate for both the youth and the staff, who are part of the inconsistent micropolitics, seem to be narrower in this regard at the most locked units.
Despite the fact that the conflict script, once it is initiated, fixes the options for action available to the staff, it also involves clear advantages. It can help staff to maintain an image of themselves as consistent and uniform in their actions, despite life at these institutions being characterised by inconsistency and a wide range of different reactions to similar situations and rule breaking. Another advantage is that the logic of the conflict script ascribes responsibility for escalating situations to the youths instead of the staff, despite the staff's obvious power advantage. This can be compared to arguments presented in prison research, regarding the trend towards responsibilisation and the increased pressure placed on inmates to make the 'right choices', which in turn leads to a reduction in the responsibility of the prison staff to help and support inmates in their return to society (Crewe, 2011b;Neumann, 2011;Werth, 2013). In this sense, the conflict script becomes part of the narrative that the staff use to describe themselves and their colleagues, whereby they can describe themselves as acting in line with the low arousal approach despite the fact that they describe isolating youths for things that start as very minor incidents.
We conclude with a few words about the study's relevance and limitations. The previous research discussed consists mainly of ethnographies, and the majority of these studies arguably comes closer to 'the natural history of trouble' (Emerson and Messinger, 1977), because they have spent more time in the field with all the methodological advantages that it entails to be a part of the everyday life of the institutions. For instance, conflicts can fester for days, causing troubling atmospheres that can be difficult for staff and youths to recount verbally. To some extent, we still come close to the everyday conflicts through the verbal accounts of these interactions. Here, we can also utilise the fact that many of the stories are similar when it comes to social control; the conflict script, the capricious application of the rule system and the superficial adaptation, to name a few. Interviews also elucidate how the individuals directly involved describe their intentions. In this sense, the conflict script may be said to constitute part of a narrative of institutional practice that is in many ways shared by both youths and staff and that the youths appear to learn as they become more integrated into the life of the institution. Whether it actually constitutes practice is more difficult to say, but the fact that it is recurrently described by both youths and staff would indicate that this is the case.
Descriptions of youth institutions in other countries show that there are substantial similarities between these institutions and the Swedish institutions this article studies (cf. Cox, 2011;Kivett and Warren, 2002;Price, 2005). The spiral of punishment may involve somewhat varied steps, depending on staff powers, but its use and logic are probably highly relevant to youth institutions in other countries than Sweden. It is less clear to what extent the conflict script is also useful in relation to adults. Power relations are less unequal between adults, since the imbalance is not intensified by the age dimension. On the other hand, many of the characteristics that can be observed at youth institutions are also present in prisons (Rowe, 2011) and secure substance abuse institutions (Laanemets and Kristiansen, 2008). Thus, it is not unreasonable to assume that control practices similar to the conflict script may also be applied in institutions for adults.