Police Officers ’ Perspectives on State ( Police ) Violence : A Sociomoral and Psychological-Driven Study on Disengagement

This paper comprises a theoretical and empirical incursion into the phenomenon of state violence, namely police violence. Although extensively explored in different perspectives within and outside academia, police violence is complex and has not been sufficiently problematized yet. In our understanding, this phenomenon requires a dialectical and dynamic discussion wherein both the development of state powers (macro-perspective) and how these powers affect the subjectification of those who act on their behalf (micro-perspective) are articulated into a critical analysis. In more concrete terms, the present study contributes to such analysis by shedding light on the main processes of moral disengagement (cf. Bandura, 1990, 1999, 2004) disclosed in the internal perspectives of six Portuguese police officers about common daily work-related situations. To legitimize the resort to police violence, police officers rely heavily on different mechanisms of moral disengagement. For instance, sanitizing language (anchored in a police technical jargon) is typically used as a linguistic mechanism to disguise violent actions; advantageous comparisons with other law enforcement agencies or with the recipient’s conduct are typically employed; non-lethal violence is usually minimized and portrayed as innocuous; and the recipient of violence is usually dehumanised and seen as responsible for the acts of violence. The findings are discussed based on the moral disengagement theory (Bandura, 1990, 1999, 2004); on the denial approach (Cohen, 2001, 2003); and on the impact of organizational, legal and socio-cultural dimensions of police organization (e.g., Fassin, 2011; Huggins, Haritos-Fatouros, & Zimbardo, 2002).

studies is that police violence or use of force can be explained through different variables (e.g., situational characteristics, personal traits) and, if these variables are understood and controlled, the damage resulting from police violence can be mitigated.Yet, these works often fail to examine, at a macro-level, how police violence is constituted within state coercive powers themselves and, at a micro-level, how these powers and the instituted 'work of violence' affects several subjective dimensions (e.g., socio-moral, psychological) of police officers as 'powerholders' or state-representatives.
On a practical level, police violence continues to be well documented in terms of the seriousness of the harm caused by the violence and how frequently citizens fall victim to it (cf.Fassin, 2011;Friedrichs, 2009;Waddington, 1999).Portugal is no exception to this level of documentation.A first example is how the report issued by the European Committee for the Prevention of Torture and Inhuman Degrading Treatment or Punishment (CPT, 2013) has shown that Portuguese law enforcement agencies engage in violent behaviours (e.g., slapping, punches and kicks) not only before, during and after detaining suspects, but also inside police facilities and during interrogations.
A second example is the emergence of police violence as an important public issue in Portugal due to the intense media coverage of police repression and violence, mainly during the public demonstrations carried out between 2011 and 2013.These demonstrations arose as an expression of the social dissatisfaction felt towards the economic austerity measures and labour precariousness of the time, and towards the oppressive response to quell the demonstrations (cf.Duarte & Baumgarten, 2015;Soeiro, 2014).Altogether, these factors have caused actions taken by several Portuguese institutions, including the police force, to be questioned.Thirdly, it is worth mentioning the public scrutiny of violence embedded in the training procedures of the Portuguese armed forces, an instance of which resulted in the deaths of two military trainees during the last year of 2017.
To our knowledge, at a moment when police violence is under intense scrutiny, both in Portugal and in other countries, a more holistic and critical problematization of police violence is still required.This endeavour must begin with a critical dialogue between macro-and micro-level interpretations of police violence, and it needs to empirically address the perspectives of those who represent the state, including police officers (i.e., internal perspectives).However, in Portugal, this effort to critically understand the violence exercised on the behalf of the state has so far favoured citizens' perspectives (cf.Barbosa, Machado, Matos, & Barbeiro, 2012).In turn, the intent of this article is to analyse police officers' internal perspectives of police violence by shedding light on the main processes of moral disengagement (cf.Bandura, 1990Bandura, , 1999Bandura, , 2004) ) embedded in six Portuguese police officers' interviews concerning common daily work-related situations.While favouring a qualitative and exploratory research design, the specific goals can be summarized as follows: 1. To stress a critical understanding of police violence, at both a macro-level of analysis (i.e., how normative notions of police violence are constituted and developed within the coercive nature of state powers) and at a micro-level of analysis (i.e., how the nature of police violence is expressed subjectively by police officers, with a particular focus on subjective socio-moral and psychological processes); 2. To map the processes of moral disengagement disclosed by police officers at different loci of moral reconstruction (i.e., behaviour, outcome, agency and recipient); 3. To discuss and to integrate moral disengagement processes in light of related macro-perspectives favouring organizational, legal and socio-cultural readings of police violence.

Macro-Level of Analysis: Contesting State Powers and Definitions of Violence
As a specialized state body, a police agency is assigned by the state ii , the primary responsibility to ensure safety and order in modern societies with recourse to the use of force (Bittner, 1990;Morgan, 2000;Newburn & Reiner, 2007).States rely on police and military agencies as their representatives to endorse the monopoly of the legitimate use of force (Green & Ward, 2004).However, legitimacy of police is more often than not called into question, and claims of illegal force or violence tend to emerge.But these claims face the contradictions of taking violence within the constitution of a state's armed forces because the dichotomy between acceptable/unacceptable use of force is an ever-floating field of struggle for legitimacy.The limits between acceptable and unacceptable use of force (i.e., or between the use of force and violence) are controversial debates between internal (e.g., police officers, military forces, other state representatives) and external audiences (e.g., communities, social movements, civil society in general, academics).
Two reasons should be emphasized, among others, for such a problematic distinction between acceptable and unacceptable use of force, or between use of force and violence.To begin with, internal perspectives on police violence will at some point collide with external ones due to different positions, power dynamics, and claims on the effects and solutions needed to address police violence (Bradford, Jackson, & Hough, 2014;Lawrence, 2000;Taylor, Wyant, & Lockwood, 2015).For example, whilst a police officer may claim to have legitimately fired a shot in the direction of an evaded inmate after trying to stop him or her by other means (e.g., asking to stop), the civil society may claim that shooting a run-away from behind is illegitimate because it is a violation of human rights and, if other means did not work, police officers should have let the evaded inmate get away.
Louise Westmarland (2001) provided interesting insights based on several cases -seen first-hand during her PhD research -in which a "thin red line" may be drawn between legitimacy and illegitimacy of police use of force depending on the eye of the beholder/social audience.The author puts forward, for example, a case in which "some officers were dragging a 'druggie' out of the back of a police van, along the ground" (Westmarland, 2001, p. 525), and "he [the druggie] was then picked up by four of the officers, one at each 'corner', and his head was used as a battering ram to attract the attention of those inside operating the automatic doors" (ibid).The police officers were not troubled about using force as this kind of punishment at any point.The author, however, points out: "From most people's perspective, this would be excessive, as the emaciated heroin addict did not pose a physical threat to the much larger stronger officers who in any case outnumbered him.His 'crime' had been to be 'lippy' to the officers who were arresting him for suspected possession of illegal substance and a refusal to submit to a body search" (Westmarland, 2001, p. 525).
Mainly within internal perspectives, the distinctions between accepted and unaccepted use of force rest more frequently on the evaluation of the legal and disciplinary corpus than on the concrete harm associated with police conduct (cf.Prenzler, Porter, & Alpert, 2013).For example, even if a given police intervention has led to physical harm or to a clear violation of human and/or fundamental rights, police officers will rely heavily on the lawful limits of their professional codes of conduct endorsed by the state's legitimate and coercive monopoly of the use of force (cf.Cohen, 2001;Matthews & Kauzlarich, 2007;Waddington, 1999).Even the excessive use of force (i.e., that which does not account for proportionality and minimal use of force iii ) can still be seen as expected or necessary amongst police officers (Phillips, 2010).Law and other disciplinary norms will permanently be perceived as the ultimate standard to define, within state coercive powers, what an act of acceptable or unacceptable resort to vi-olence is.In this sense, internal or official perspectives tend to advocate violence as an atypical phenomenon within police agencies (Cohen, 2001) because violence itself is strictly understood as the use of force punished by the disciplinary or ethical codes that govern police intervention.
External perspectives emphasize, however, how police violence must be examined as a common feature of police work, which is framed and concealed within the context of the state's internal policy (Barbosa & Machado, 2010).
For example, within the criminological realm, a harm-oriented perspective (cf.Green & Ward, 2000, 2004;Green, Ward, & McConnachie, 2007;Ward & Green, 2000) has attracted much attention in the discussion of the notion of state violence beyond a mere legal-oriented approach iv .This perspective makes it possible to discuss all the harmful actions perpetrated on behalf of the state if such actions are acknowledged and censured as violence by a given audience even if external to state powers (Green & Ward, 2000).
The harm-based approach is crucial for establishing more critical inroads on police violence.Violence also becomes a major concept in that it is more far-reaching than 'use of force' or 'excessive force', both of which are profoundly associated with the unyielding and hegemonic acceptance of legal standards and internal codes of conduct (Soares, Barbosa, & Matos, 2017).Here, the term 'violence' appears as an all-encompassing expression since "it defies easy categorization; it can be everything and nothing; legitimate or illegitimate; visible or invisible; necessary or useless; senseless and gratuitous (…) rather than sui generis, violence is in the eye of the beholder" (Scheper-Hughes & Bourgois, 2003, p. 2, original emphasis).Indeed, in this paper, police violence is used as a non-rigid and flexible conceptualisation able to critically engage both with internal and external perspectives, therefore calling into question the hegemony and acritical legitimacy of internal perspectives (Soares, Barbosa, Gonçalves, Magano, & Matos, 2014;Soares, Barbosa, & Matos, 2017).This does not mean absolutely rejecting the legal underpinnings and internal codes of conduct that are always mediating and shaping the definition of police force, violence and legitimacy.Instead, it emphasizes and challenges the intrinsic contradictions that surround the exercise of violence on behalf of the coercive powers of the state.

Micro-Level of Analysis: Towards a Socio-Moral and Psychological Perspective on Police Violence
It is only when the internal legitimacy to define police violence is questioned that we realise how the constitution of the police subject is much more troubled with moral ambiguity and contradictions than it seems to be on the surface.As Didier Fassin (2011) notes, "[to] understand the viewpoint of law enforcement agents, and recognize that where most observers see flagrant abuse of ethical precepts, the officers themselves are convinced they are acting in accordance with the moral code of their profession" (p.198).

Professionalization of violence
v within state-representative agencies such as in the case of police officers, entails ethical, social-moral and psychological implications which are endemic to the constitution of coercive powers (Fassin, 2011;Huggins et al., 2002).Moral disengagement theory can be helpful to understand the contradictions and dilemmas surrounding violence professionalization.Moral disengagement theory recognises that human beings tend to internalize standards and codes of conduct based on moral tenets and values (Bandura, 1990(Bandura, , 1999(Bandura, , 2004)).This process leads to a self-regulation process in which individuals do not behave in such a way as to violate the previously socialized moral values.For Albert Bandura, the exercise of moral agency operates in a dualistic way characterized by an inhibitory and a proactive component.In its inhibitory mode, self-sanctions are disabled, allowing the morally disrespectful and harmful conduct to occur.
In the scope of moral disengagement theory, a different set of processes must be defined and framed within its applicability to analyse police violence.With respect to the set of behavioural processes (i.e., the moral reconstruction of the harmful action in itself through moral justification, sanitized language and advantageous comparisons), moral justifications can be understood as the values and moral precepts used to turn a morally-perceived harmful action into a legitimized one (Bandura, 1990(Bandura, , 1999(Bandura, , 2004;;Hirschberger & Pyszczynski, 2011).This means that police violence is bound by a moral economy of the police work (Fassin, 2011).'Supreme values' such as security, social order, self-defence or freedoms are often used to rationalize the utilitarian vision of those who act on behalf of the state's powers (Belur, 2010a;Cohen, 2003;Kauzlarich, Matthews, & Miller, 2001;Zimbardo, 2004).Moral justifications are processes of implicatory denial (Cohen, 2001(Cohen, , 2003;;White, 2010).Implicatory denial refers to the use of political, moral and psychological rationalizations (i.e., "this happened, but it is justified") when a violent action is understood as such but its harmful consequences are denied (Cohen, 2001).
Secondly within this set of mechanisms, sanitizing language (or euphemistic language) refers to the use of palliative expressions and terminologies, which ascribe an acceptable narrative to detrimental conduct (e.g., "murdered innocent civilians" is conceptualized as "collateral damage") (Bandura, 1990(Bandura, , 1999(Bandura, , 2004).In the case of police officers, there seems to be a completely technical and professional terminology that guarantees the preservation of a positive image as agents of the state, even after committing harmful actions (Barreto, Borja, Serrano, & López-López 2009;Belur, 2010a).Police officers do not kill, instead they eliminate targets or terrorists; they do not stand simply for public order; they stop 'savages' or 'parasites ' (cf. Fassin, 2011;Huggins et al., 2002).Indeed, following literal denial (i.e., "this does not happen"), sanitizing language is relevant in understanding processes of interpretative denial, which implies reframing the violent act as a non-violent one (i.e., "this happened, but it is not violence") (Cohen, 2001(Cohen, , 2003)).It operates on a linguistic level of moral reconstruction wherein professional technical jargon appears as fundamental in a strong word game of meanings (Belur, 2010a).
Lastly in this set of mechanisms, advantageous comparisons take place when subjects compare their harmful actions with other harmful actions perpetrated by other subjects/groups/organizations, therefore conceptualizing the former as harmless and fair.For instance, subjects who have been victims of terrorism may portray their own violence as insignificant, or even benevolent (Bandura, 1990(Bandura, , 1999(Bandura, , 2004)).By way of advantageous comparison, police work is often based upon the contingent and proportional foundation on which every action is seen as an expected response to an even more harmful and unexpected behaviour from third parties (Soares et al., 2017).
With regard to the set of outcome mechanisms (i.e., the moral reconstruction of the effects of the harmful conduct), all processes implying avoidance, minimization and disregard to the harmful effects of violent actions can be found here (Bandura, 1999(Bandura, , 2004)).After committing a harmful action, perpetrators may claim that the victim has suffered no pain (Cohen, 2001).In this case, individuals do not dwell on the consequences of their misconduct.This psychological distance creates a mental barrier through the denial of a victim's suffering (Cohen, 2001;MacNair, 2003).Additionally, damage minimization arises when perpetrators distort the degree of injury associated with the inhuman conduct.On a regular basis, police officers tend to understand that their violent actions are better than inaction, and even if the latter tends to result in greater harms to society, these tend to be minimized or disregarded (Belur, 2010a).
With regard to the set of agency mechanisms (i.e., the moral reconstruction of responsibility towards the harmful action through displacement or diffusion), displacement of responsibility can be defined as the suppression or minimization of moral agency by placing blame, for example, on higher authorities (Bandura, 1990(Bandura, , 1999(Bandura, , 2004)).
While operating on behalf of hierarchical organizations and on violence professionalization, subjects can commit harmful acts during the course of their professional activities because they have developed a cognitive sense of "I am just doing my job" (Beu & Buckley, 2004).Responsibility and unethical consequences are thereby shaped within certain organizational conditions.From adopting internal coercive and disciplinary rules and wearing a uniform to the building of strong bonds of camaraderie that strengthens group cohesion, the police officer is subjected to a masculine social structure in which the subject accommodates authority and dominance (Huggins et al., 2002).Police officers then develop a positive self-image in which they are self-portrayed as non-violent individuals who only take on an assigned and professional role (Belur, 2010a).Moreover, agency displacement is fostered by the volunteer character and high legitimacy attributed to the authority of democratic states (Freire & Neto, 1988) and to the legal standards governing their actions (Bassiouni, 2010).
In turn, diffusion of responsibility requires a collective and segmented action that provides a solid sense of invulnerability, while moral consequences are ignored and self-censorship is disabled (MacNair, 2003).When professional roles are segmented and routinely performed, each task segment can be viewed as harmless, even if the global results of the task are unethical and harmful.In law enforcement agencies, diffusion of responsibility is prompted by the insularity of their work.Insularity can be defined as the organization of police work by intervention units (e.g., special operations unit, drug enforcement unit) which is a taylorization of the autonomous structures of law enforcement agencies (Huggins et al., 2002), leading to group cohesion and deindividuation.In turn, group cohesion and deindividuation in police work can foster the dilution of moral agency and, as a result, self-consciousness is minimalized with regard to personal moral standards (Gibbons, 1990).
With regard to the set of recipient mechanism (i.e., moral reconstruction of victim's characteristics or responsibility), police officers may also blame the victims or the circumstances surrounding them in order to explain and justify their harmful actions.Attribution of blame occurs when the violence recipient is assumed to be the agent responsible for the harmful conduct, one whose acts are seen as provocative and intolerable.Violence thus becomes an expected and inevitable reaction, a self-defence conduct (Bandura, 1990(Bandura, , 1999(Bandura, , 2004)).The responsibility for the harm is placed on the target.Consequently, victims and violence are both denied because the perpetrator of the violence becomes the 'real victim' (Cohen, 2003) and his/her acts become the expected punishments (Staub, 2011;White, 2010).
Finally, dehumanization is one of the most discussed psychological processes of moral disengagement.Almost all violent actions entail a victim's reconstruction, a process of dehumanization (Haslam, Bastian, Laham, & Loughnan, 2011).Once a person is perceived as a 'sub-human', violence becomes accepted because victims are seen as both insensitive and deserving recipients of punitive methods (Bandura, 1990).Dehumanization often entails a demonization process in which targeted individuals are labelled as belonging to negative social or political groups (i.e., folk devils) and framed as the prototype of 'evil' in a given society (Barreto et al., 2009;Cohen, 2002).
By including individuals in threating, deviant or criminal groups, usually police officers legitimize violent actions assuming that their only goal is to re-establish social order and to punish the 'wrongdoers' (Borja-Orozco, Barreto, Sabucedo, & López-López, 2008).
To sum up, police officers will be constantly duty-bound to legitimatize their own acts of violence in particular ways.This means that internal perspectives on police violence tend to sustain and reinforce the state coercive powers.The theory of moral disengagement is a useful framework to question the "self-legitimizing" internal perspectives and the socio-moral ambiguities of those subjects who work on behalf of state violence.In order to map these mechanisms of moral disengagement, six Portuguese police officers have been interviewed about common daily work-related situations.The following section explains the full research design.

Method Participants
This study comprised of all police agencies in Portugal working in internal security and crime prevention (i.e., two participants from the Public Security Police: 'PSP1' and 'PSP2'; two participants from the National Republican Guard: 'GNR1' and 'GNR2'), crime investigation and repression (i.e., one participant from the Judiciary Police: 'PJ1') and immigration control (i.e., one participant from the Borders and Foreign Service: 'SEF1').Accordingly, the participants constitute a purposive and heterogeneous sample thereby optimizing the possibilities to detail divergent views (Ritchie, Lewis, & El am, 2003).
The participants included six Portuguese white men with solid work experience in police forces, in all cases with more than fifteen years of service.All participants had experience in various facets of police activities throughout their careers.Furthermore, four participants have higher education; one concluded secondary education and the other elementary education.Concerning the age range, two participants were between 30 and 39 years of age; three were from 40 to 49 years of age; and one individual was from the 50 to 59 years old age range.Police rank, professional career or professional associations to concrete police cases are not displayed in this paper as all the participants were told that this information would not be registered under any circumstances.

Instrument
A semi-structured interview guide titled "Perspectives on Police Violence" (cf.Appendix) was constructed based on a pre-existing guide -"Perspectives on State Violence" (cf.Barbosa & Machado, 2010).Some of the original scenarios were retained; some others were removed because they were not adapted to the police officers' professional experiences; and, finally, new scenarios were introduced.
"Perspectives on Police Violence" comprises seven vignettes: a) social demonstration; b) police chase; c) use of aggression/torture; d) police search; e) prison; f) individuals barricaded in buildings; and g) containment of rioting at sports events.Each of these scenarios incorporates multiple micro-scenarios which evaluate some aspects that may influence moral agency inhibition.For instance, in the social demonstration scenario, micro-scenarios enable us to understand how identification with the 'other' and the demonstrators' behaviours (e.g., civil disobedience, aggressions to police officers or to third parties) may influence police officers' moral disengagements.Participants were asked to give their opinions on the use of police violence for each scenario.All possible dimensions that may introduce variability in the participants' perspectives of police violence were covered (e.g., the figure of the recipient, types of crimes, types of conduct in loco).

Procedures Sampling
Initially, it was established that at least one officer from each Portuguese police organization should be interviewed in order to obtain a heterogeneous sample in terms of different law enforcement agencies.Purposive sampling grounded in qualitative methodology allowed us to represent a set of experiences that were deemed as relevant.
Along this line, therefore, instead of taking a large number of participants, this study has selected a small number of participants, and has provided an enlarged set of experiences regarding the police officers' fieldwork (e.g., police officers coming from different agencies and with a heterogeneous history of fieldwork).For example, the Public Security Police (PSP) and the National Republican Guard (GNR) are particularly oriented to internal security matters and crime proactive and repressive control.Indeed, social demonstrations or containment of rioting at sports events are scenarios most closely related to the work that these police forces engage in.In contrast, the Judiciary Police (PJ) deals with crime investigation and repression in a non-proactive way (e.g., homicides, issues of state security, economic crimes, terrorism, and misfeasance).PJ officers are not able, for example, to give an empirical perspective about social demonstration scenarios.However, they can express relevant points of view on the use of violence in the course of repressive actions to the aforementioned crimes such as homicides or issues of state security.Likewise, Borders and Foreign Service agents can provide different perspectives since their work context has specific features, such as working with immigrant populations and boarding controls.By putting together these different police forces, we were able to meet specialized and divergent views towards several police officers' activities.
Assuring a heterogeneous sample underscores the main goal of the non-probability sampling process wherein: "(…) units are deliberately selected to reflect particular features of groups within the sampled population.
The sample is not intended to be statistically representative (...) but instead, the characteristics of the population are used as the basis of selection" (Ritchie, Lewis, & El am, 2003, p. 78).
In terms of sampling strategies, participants were identified through special informants (i.e., gatekeepers).These informants provided the researchers with the contact information of police officers with a significant field experience (more than ten years) within the pre-defined police agencies.Our goal was to access participants with specialized experience who are thus capable of developing empirical, professional and, at the same time, critical positions on police violence.

Data Collection
The interview is the selected method of data collection given that it allows access to multiple interpretations and personal experiences about the studied phenomena (Mason, 2002;Ritchie, 2003).Participants were contacted via email in order to clarify the scope of the study and data collection procedures.During these communications, and after obtaining informed consent from the six participants, the interviews were scheduled at the most suitable time and place for the participant.The participants signed informed consent and all interviews were tape-recorded to avoid loss of information Guaranteeing anonymity for all participants in the moment of writing up was also of fundamental importance: no sensitive information such as name, police rank and professional career was registered at any time during the research.In addition, the ethical principles and recommendations on scientific research presented in the Deontological Code of Portuguese Order of Psychologists (e.g., non-maleficence, informed and voluntary participation, post-research information) were followed.
At the beginning of the interviews, the research team undertook an initial entry approach that helped to establish a relationship built on trust.First, the duration of the interview was negotiated with the interviewer.Second, the participants were informed that they could stop the interview at any time.For ethical reasons, it was agreed that any professional case disclosed during the interview would not be transcribed or publicized under any circumstance.
Lastly, the concept of violence was also discussed.As explored in the theoretical framework, violence may have separate understandings for distinct social audiences, and it has specific connotations within police forces.Police forces are regulated and affected by internal definitions of police force and/or violence.Consequently, at the beginning of each interview, we again discussed the study's general objectives and tried to deconstruct the importance of having a critical understanding of violence.
During the interviews, the research team tried to establish a rapport based on trust by fostering a non-judgmental atmosphere, one of active listening and constant reassurances of anonymity in the moment of writing up.The interview was also understood as an active process of co-construction, that is, a process of inter-reflexivity (cf.Pain, Kindon, & Kesby, 2007).The interview allowed the space to clarify, to deconstruct and to continuously inter-relate the emergent understandings of police violence for both the interviewee and the researcher.
Participants were also invited to reformulate the scenarios introduced by the researcher in order to improve the instrument's validity.Following police officers' clarifications, changes in the interview guide were systematically carried out (e.g., police concepts such as police raid and police search were introduced due to participants' clarifications on the distinction between public and private).This active and participative process allowed the research team to accommodate some technical language within the construction of each scenario.Moreover, some participants explained that a few of the scenarios were not compatible with their professional experiences.For example, all the participants mentioned having no experience with imprisonment scenarios.Only those settings that reflected the professional experience of each participant were discussed (cf.Table 1).
During the interviews, it was also discovered that the interview guide did not cover important scenarios for some participants.These scenarios were also discussed during the interview.For instance, the SEF inspector brought up the scenario of the transport of a prisoner.The participant described the scenario and explained 'why' and 'how' violence may emerge within this type of scenario.This dynamic and flexible process of data collection was not taken as a methodological disadvantage.By acknowledging the exploratory logic of this study, the imposition of an inflexible structure might have jeopardized the advantages of conducting interviews (Darlington & Scott, 2002).All of the interviews lasted about one and a half hours.
Content analysis was performed in order to more closely examine the data.This procedure can be defined as "a set of techniques for communications analysis in order to obtain, from systematic and objective procedures of messages content, indicators (quantitative or not) that enable the inference of knowledge regarding the conditions surrounding production and reception of these messages" (Bardin, 1977, p. 42).In this work, we defined the following indicators: the theme as the unit of analysis and the paragraph as the unit of context.The counting rule was the number of participants.
After defining our indicators, the techniques and procedures of data analysis followed Strauss and Corbin's (1990) model of data interpretation for content analysis.In this process, the first step is coding.In the present study, the encoded material was inserted into a semi-inductive system of categories.This means that no pre-defined grid of analyses was developed; however, "no qualitative method rests on pure induction -the questions we ask of the empirical world frame what we know of it" (Charmaz, 2005, p. 509).Therefore, theoretical sensitivity to data -characteristic of grounded analysis (cf., Strauss & Corbin, 1990) -was favoured, although assuming the permeability of the researchers' theoretical frameworks.
To improve the data analysis process, the three authors of this paper set up a peer-researcher team.At the beginning, the system of categories was mainly descriptive, close to the participants' expressions and descriptions (open coding; nvivo coding).Following the open coding principle, the data units were inserted into as many categories as possible.At this point, memos that marked impressions and relationships were made.These helped in the later consolidation of the categories system, which underwent re-classifications and re-arrangements according to the continuous codification of the material.This process is typical of a second phase of encoding in which data become more theoretical as a result of axial coding (Saldaña, 2009).At this point, the coding reflected moral disengagement (Bandura, 1990(Bandura, , 1999(Bandura, , 2004) ) and denial processes (Cohen, 2001(Cohen, , 2003)), which led to more conceptual categories.Finally, during selective coding, the generation and organization of categories has been conceptually more differentiated (Charmaz, 2005;Strauss & Corbin, 1990).A set of different strategies were used in order to verify the adequacy of the grounded model.Internal validation has relied on the constant comparison of the data across the several cases/interviews.External validation was achieved through the triangulation of multiple readings within the research team.In the end, respondent validation was also taken into account.Data analysis has been returned and accepted by the participants.Data storage and analysis were carried out using NVivo 9 software.

Findings and Discussion
Police officers whose duties inevitably involve the use of violence -legitimized or not by their professional code of conduct or by any legal canon -engage in several moral disengagement processes when acting on behalf of the state coercive powers.The selective activation of these moral disengagement processes emerges at the same time police officers' moral standards remain unchanged.In other words, and following Albert Bandura's framework, our results show how moral disengagement allows for the emergence of harmful conduct even if basic "anti-violence" moral standards (e.g., peaceful ideals, no-harm principles) remain unaffected (for further details, see in this section:

Maintenance of a fixed set of moral standards).
Different mechanisms of moral disengagement were displayed by the participants.The discussion of such mechanisms is organised in this paper in accordance to Bandura's distinction between different sets of mechanisms such as behaviour, outcome, agency and recipient mechanisms (Bandura, 1990(Bandura, , 1999(Bandura, , 2004): 1. Behaviour mechanisms are reflected in the use of sanitizing language framed within police officers' technical and professional jargon, for example.Advantageous comparisons with third-parties (e.g., law enforcement professionals, others' victimization); moral justifications and rationalizations anchored in the defence of superior values and ideals (e.g., life, safety, social order); job requirements (e.g., obtaining information), among other moral principles (e.g., proportionality), are some examples, to name a few, about this set of mechanisms (for further details, see in this section: Behaviour mechanisms); 2. Outcome mechanisms are associated with non-lethal harm inflicted on victims of state-perpetrated violence, and which tends to be minimized in terms of its effects (for further details, see in this section: Outcome mechanisms); 3. Agency mechanisms, such as displacement of responsibility to a superior authority or order, usually a court order, are usually pervasive in police officers' narratives.'Being a cop' and the duty to comply with the law also emerged as important positions (for further details, see in this section: Agency mechanisms); and 4. Recipient mechanisms have emerged in the form of the concept of dehumanization and of attribution of blame, as police violence tends to be seen as acts of retaliation in the face of resistance, disobedience or disrespectful conduct toward others (for further details, see in this section: Recipient mechanisms).

Maintenance of a Fixed Set of Moral Standards
The creation of 'professionals of violence' may be achieved without any shift in their moral standards.It is through this process that the perpetrator's positive self-concept remains unaffected (Bandura, 1999(Bandura, , 2004)).Along these lines, several participants (4/6) stressed how violence is blameworthy and constitutes a moral wrongdoing, rejecting violence as a moral principle which guides their actions: "There is no one, given their archetypes already shaped, that will say to you: violence is legitimate.Why?Because they are formatted to do good things" -PJ1 Thus, some of the participants have unveiled forms of literal denial (Cohen, 2001) in their stories in an attempt to clarify how police violence is an uncommon phenomenon, one that is almost non-existent: "Yet I also want to say... that [police violence] is not very common, right?Police officers are normal people" -PSP2 Literal denial was then revealed when participants stated that overall police violence -mainly considered in their perspectives of illegitimate use of force -is an atypical phenomenon only associated with certain police officers who are seen as non-compliant with police rules due to their stable internal characteristics.As argued by situational theorists on the perpetration of violence (e.g., Zimbardo, 2004Zimbardo, , 2007)), in order to justify harmful actions recognized as immoral, subjects may claim that these are not systematic but rather casuistic and particular actions carried out by others who are not properly acculturated to a fair and well-regulated organization.This individualized and essentialist theory based on the "rotten apples approach" (e.g., police officers do not use violence; they only act within the legal limits) is commented upon by other researchers on police violence (Belur, 2010a(Belur, , 2010b;;Lawrence, 2000) who have shown how police violence becomes normalized as it is only ascribed to rogue cops.
As above, these individualized perspectives are often related to a "blue wall of silence" (Lawrence, 2000;Workman-Stark, 2017) that is, a 'code of silence', secrecy and solidarity built upon a recipe knowledge (i.e., a tacit form of knowledge about what should and should not be done to cover up misconduct) within police culture (Chan, 1996).
These perspectives clearly undermine and obscure the contextual and culturally defined nature of police violence.
Likewise, this may explain why police officers are compelled then to keep the word 'violence' away from their vocabularies when they speak about their work and about the situations that they face in everyday professional life.
Violence could well cause officers to reassess the structure of their professional category and the cultural cohesion of their group, or even to crumble their self-concept as 'moral guardians' of positive values.Violence problematizes harm, and it requires the acknowledgement of the contradictions that give rise to what the police view as institutionalized notions of acceptable and unacceptable force.It is a challenging terrain characterized by cognitive dissonance.In fact, police officers' choice of words in their interviews illustrates how difficult it is to construct a moral positioning in which their moral assumptions and their violent actions are brought together: "Do you think that any police officer is concerned about the consequences because he thrashed a guy like that?No! Now, ask me: is it legitimate?No, it is not!Yet it is understandable.Is it acceptable?Acceptable is a complex verb.Ask me the question again with another verb!" -GNR1 Notions such as violence make police officers more aware of the moral contradictions presented in terms of principles and practices, which Stanley Cohen (2003) has coined as the 'paradox of knowing and not-knowing'.Put simply, since stable and universal moral tenets nourish a positive self-image (Staub, 2011), it then becomes necessary to refuse certain features, motivations or characteristics of the harmful actions.Moral disengagement processes emerge then as the categories used to make sense of the moral agency inhibition amongst police officers.

Moral Disengagement Processes Behaviour Mechanisms
All of the participants (6/6) used sanitizing language which was directly associated with police technical and professional jargon.Expressions such as "void, immobilize, neutralize" (4/6), "muscular action" (3/6) and "tie down" (2/6) were highlighted to refer to police repressive and violent actions.Sanitizing language is deeply connected with interpretative denial.Authors such as Belur (2010a) have shown that through the use of euphemistic language and professional jargon, a guiltless and acceptable version of the harmful conduct is created at personal, legal and social levels.Our findings support this connection.On several occasions participants referred to harmful police actions as non-violent behaviours with a legitimate framework: "The use of a police baton, but I do not consider this a violent action.That is, I do not… I can't consider that as a violent act.Rather, it is an attitude… a firm voice (…)" -PSP1; "Now, let's define violence (laughs)… Hum, is it violence a 'pair of slaps'?You also need to set a scale for this kind of intensity" -SEF1 Advantageous comparisons (i.e., Bandura's mechanism to make sense of the comparisons made between the perpetrators and other persons/entities to legitimize their own violence) were popular amongst the participants (6/6).This moral disengagement process emerged through the contrasting of police officers' violent conduct with other harmful acts done by the target (5/6), also as a form of blaming the victim: "The paradox that exists is that they [the recipients of violence] require that we abide to the law, that we respect the rights -the rights they have -even though they do not respect the rights of others."-GNR1 Further in the interviews, some participants also contrasted their conduct with other police officers' conduct seen as more harmful (3/6).Following the 'rotten apples' theory, a police officer's actions are not portrayed as harmful since other officers' conduct is worse.An uncritical view concerning legal and authorized harm is augmented: "At least from what I have seen nothing too grievous.Obviously, these situations, when compared to situations that my PSP and GNR colleagues have to face, are nothing but child's play."-SEF1 Finally, two participants contrasted their harmful conduct with other victimization experiences conceptualized as more harmful (2/6): "When millions people's lives are at risk, does cutting a finger make any difference?"-GNR1 On the other hand, to solve cognitive dissonance between moral values and immoral conduct, people may rationalize and justify their harmful conduct, that is they may use moral justifications (Bandura, 1990(Bandura, , 1999(Bandura, , 2004;;Hirschberger & Pyszczynski, 2011).Several participants (5/6) used different moral principles and 'the supreme good' in order to endorse violence.Among the more cited ones, our data emphasizes life: "Look, the… the most… most protected and most sacred thing is life, isn't it?(…) it is like I am going to kill to save a life.(…) But the essential thing, the most valuable thing is life, isn't it?(…) In that case, the use of force is always legitimate, whatever it may be" -GNR1 Secondly, it brings the idea of security to the fore: "Because the one who is performing police searches has to… can apply the necessary violence to ensure people's safety.Then, the context of violence arises" -PJ1 And, finally, public order maintenance: "This is our job.To maintain the public order" -PSP2 The appeal to higher values on behalf of 'the common good' allows for the reinforcement of Cohen's paradox (i.e., the subject acknowledges the harm associated with his or her violent behaviour but without ever recognizing the suffering caused to others).The harmful nature of the violent action is extinguished.What in fact happens is a contradictory over-emphasis of certain moral values and of police officers' moral superiority with respect to other audiences (Staub, 2011).In the above participant's statement -"it is like I'm going to kill to save a life (…)" -violence becomes legitimized from the same moral assumptions that should invalidate and delegitimize it.In fact, non-violence can be depicted as immoral and violence as a 'necessary evil' required to effectively overturn a situation of extreme injustice in which, paradoxically, violence seeks a pacifist ideal (cf. MacNair, 2003).
Job requirements were also mentioned by the participants (4/6) as priorities which may legitimize the use of violence, such as obtaining information and collecting evidence in terms of drug related crimes: "Perhaps in a patrol case, it is easier if the police officer himself slaps the individual once or twice to make him tell where it is… where something is!" -PSP2 "Now, if he gets rid of the drug, my work goes down the drain.So, sometimes when trying to protect that evidence, we often get involved in physical confrontations where violence inevitably erupts."-PSP1 Mainly in drug-related crimes (in which the evidence of the crime is the possession of the drugs themselves), violence is clearly part of the work environment.For the participants of this study, long hours and days of police work can be destroyed in short minutes during an encounter with a target if the subject succeeds in getting rid of the drugs.What is at stake in these statements is career objectives and rewards.They believe that there is an amount of violence that can be legitimized to achieve a level of organizational performance.
In turn, the proportionality principle also provides an explanation for the perpetration of violence because, according to the participants (4/6), violence becomes justified when others use violence against them: "(…) now, there's an expression we use: 'we dance along with the music'.If they are violent with us, they shouldn't expect anything from us other than payback."-GNR1 Therefore, police action has to be proportional and adjusted to these actions.This principle will receive further attention in the sections below.Following the previous statements, but in a less expressive way, some participants (2/6) also justified police violence as being a necessary 'work environment' by stating two different principles, namely 'violence is a way of taking the law into our own hands': "Maybe there is the idea that the judge won't carry out justice because what you've done is not even… is not enough for you to go to jail.However, if you ask me, you go from here, you go to the judge, but you will go with a 'heated body'!" -PSP1 And, secondly, 'violence can only be stopped by violence': "(…) because an individual who is attacking someone, if he is also subject to an attack, he will usually stop the aggression he is committing!" -PSP1 As seen in other previous studies (e.g., Fassin, 2011), these principles are often called upon when the police officer clearly views the targets of violence as guilty parties, but also immune to harsh punishment under the law.

Outcome Mechanisms
Most of the participants (5/6) reconstructed the outcome of their harmful actions primarily when they referred to non-lethal damage.For the participants, 'minimal harm consequences' are seen as the product of police use of force; they can be perceived as irrelevant or as too small for a problem-posing approach.This kind of violence tends to be conceptualized as harmless or as a mitigating factor of harmful consequences that are deemed as unimportant: "Imagine that if you shoot him… and if it wounds his leg, you don't kill him.You just stopped him, wounded him and as he was in pain he stopped (pause).What happens to the police officer?Nothing, he will probably be praised, 'it was great', 'nice capture', and 'you're the man!"-GNR1 The infliction of harm is foreseeable and respected within the masculine police culture, as "the ideal man [is] autonomous, brave and strong" and surely does not show feminine features such as displaying emotions or caring about others' suffering (Ely & Meyerson, 2010, p. 4).Police officers are thus enthusiasts of risk-taking and they will not withdraw from acting violently towards others.So, the focus is not on the infliction of harm itself, but its calibration under concrete situations instead.This evaluation process is precisely a "men's work" which can be starkly and objectively developed by the experts on violence themselves.
Once violence is professionalized, converted into a work environment, there are some kinds of harm that cannot be totally measured by everyone because, otherwise, the job of policing would lose its meaning.From the viewpoint of the principles of proportionality and minimum use of force, and as the previous participant's account shows, 'minimal harm consequences' are the expected and desirable outcomes of a mindful use of violence as a resource.
A large number of the participants of this study disregarded the harm committed if such injurious action resulted from the careful analysis of the extent to which the violence had to be used.Most of the participants were truly convinced that they can develop such appraisals in indisputable ways.
Harmful consequences are under-estimated and disregarded.In doing so, participants held to the notion that their acts of violence would never instigate torture or inflict harm from mistreatment, and thus they maintained a positive self-concept in which there was an expected and beneficial level of violence proper to social control (Zimbardo, 2007).In much the same way, it is important to question the features of the 'acceptable versions of police violence' and what could unequivocally define its limits and the appropriateness of law to this end.Defining what constitutes an act of violence is a question that remains, and it will never be an unbiased process and hence proportionality and minimum force criteria will permanently be constrained within subjective appraisals.

Agency Mechanisms
As for agency locus processes, once again all participants (6/6) deflected responsibility in terms of harmful conduct, even if none of the participants stressed issues of diffusion of responsibility.In accordance, in their perspectives, only displacement of authority was an important question.Police violence becomes acceptable when framed within the concept of obedience to hierarchical orders (5/6), to the duty of 'being a cop' (4/6) and to the law itself (4/6).At this point, the data reflect a fundamental issue in the exercise of police duties, namely obedience and uncritical conformity to hierarchical order (cf.Bandura, 1999Bandura, , 2004;;Beu & Buckley, 2004): "I have hit someone when I wore that beret.I hit them, right?But it was legitimized by the power of the state.I mean, 'you've got to beat those people' and I went there and beat them.It was much easier for me, I came back home and I slept" -GNR1 In rigid organizational structures such as police agencies, in which hierarchical superiority is highly inculcated, the legitimacy attributed to authority is unquestionable.Hierarchy and authority are part of the far-reaching police training and culture developed concerning the values of cohesion and 'no-fear of fighting' (Huggins et al., 2002;McCulloch, 2001).A 'culture of authority' is built upon male-oriented roles.To a certain extent, this culture is reflected in the many different symbols that the beret illustrates.Wearing a beret is symbolic of male authority.It can signify internal social control and it reinforces detachment from external audiences (Workman-Stark, 2017).
The beret places the officer high upon an ordered pyramid.It also allows for the exercise of authority outwards since it demands the recognition of such a figure of authority by broader social audiences.
Even when facing orders the participants do not agree with, the orders are carried out with no questions asked.
Obedience to authority arises as an agency displacement process.The concept of guilt is then attributed to a hierarchical superior who, in turn, shall be responsible for harmful consequences (Bandura, 1990(Bandura, , 1999(Bandura, , 2004;;MacNair, 2003).When we refer specifically to obedience to superior orders, the presence of a court order seems to be crucial to moral disengagement in respect to the harmful conduct (5/6): "Even if it is very difficult, right?It is very difficult in that situation (…) but there is no other option.With a court order in our hands, there are no excuses."-GNR2 The contradictions between legality and harm proves then to be pervasive.Even in a situation of recognized harm, court orders are seen as imperatives.Legal standards can offer the necessary justification for the recourse to police violence, namely to expected and predictable violent actions: "As I said, we have to enforce the law no matter who gets hurt!" -GNR1 In this sense, our data confirms the idea that violence is constantly under examination but this process is mainly ruled by departmental policies, lawful limits and by broader general expectations of what a 'good cop' means (cf.Cohen, Lawrence, 2000).However, we argue that legality is a very ambivalent notion to use to discuss the legitimacy of violence because there are many grey areas.The question is not only about the police officer's conduct and about his or her conformity to legal standards, but rather about their complex connection to the moral justifications of violence, discretionary power and lawful jurisdiction.As Schmitt (2004, p. 4) points out "whoever exercises power and government acts 'on the basis of law' or 'in the name of the law', he does nothing other than what a valid norm permits jurisdictionally".When a participant says that he acts "in the name of law", he is ascribing a 'legal character' to certain harmful conduct.Consequently, we claim that a moral subterfuge emerges in which legal principles and frameworks are not ends to achieve or duties to fulfil, but instead means to justify police violence, to obtain jurisdiction.Legality will undoubtedly bind and dictate what police officers can do in a certain situation; however, the decision is always taken in loco by police officers.
Although highly underestimated, police officers enjoy high discretionary power and the ability to bring norms, tools and resources to a concrete situation (Fassin, 2011).Our data show that, when facing a moral dilemmatic and inherent contradictory work situation (represented in the aforementioned seven scenarios), legal standards are tools used as subterfuges, which police officers most often call upon to justify certain violent acts and to reject others.

Recipient Mechanisms
All of the participants (6/6) rebuilt their perceptions of the victims/targets of violence by dehumanizing them (5/6).
According to the participants' perspectives, the characteristics of the victims, such as the level of danger, criminal intent or malevolence, are constructed from an essentialist point of view which allows the violent behaviour: "I have a colleague who says (I also agree): human rights are for human people.A guy who rapes a twelve-year-old child isn't a human being (pause).So, he cannot have human rights.He may be entitled to some other things.A specific set of rights should be created for that kind of person" -GNR1 This example clearly challenges the concept of humanity, and more concretely the human uniqueness (i.e., features that make humans unique in comparison to other animals) (Haslam, 2006) of sexual abusers is completely dismissed.Moreover, the lack of recognized humanness bestows moral superiority (Zimbardo, 2007) and fosters binary classifications of groups, such as us and them (Haslam, Bain, Douge, Lee, & Bastian, 2005).Following the above participant's quote, this dichotomy is presented between "police" versus "sexual abusers".Our data show how the otherness experience thus works as a sanitizing language tool to rationalize and dehumanize the target as dangerous, criminal or evil.The police culture of authority is thus also reflected in the construction of otherness.As some authors underline (Chan, 1996;Fassin, 2011), the problem with the logic of crime fighters or of 'fighting the bad guys' is that it relies on the stereotyping and marginalization of the persons who commit crimes: "We have developed a certain aversion to those people [persons with criminal records previously known by the police officers].We have created levels and levels of hatred and contempt, and also of not positive thoughts about them such as «when I lay my hands on you, I am not going to do you any good»" -GNR1 For police culture in general, assorting general public into categories of dangerousness or badness can be a way to minimize the very risks of their profession, a way of maintaining 'the edge' (Paoline, 2003).Fittingly then, blame attribution has also emerged as a moral disengagement process (5/6) often related to the notion that every police action is a consequence of the target's resistance, disobedience and disrespectful conduct and thus anchored in the proportionality principle: "If he is violent, we use violence (…)" -PJ1; "What we take for granted is that we would have a lot of difficulty (any police officer would), if we arrived there and immediately started beating or charging, right?But, if we have given them the chance, once, twice, three, four times to leave…" -GNR1 Blame attribution can be related to the idea of a non-collaborative or non-cooperative subject who often triggers the violent behaviour exhibited by the police officer (cf.Crawford & Burns, 2008;Garner, Maxwell, & Heraux, 2002;Klahm & Tillyer, 2010).One of the most representative features of police culture is perhaps the constitution of an authority that cannot be defied 'from the outside' under any circumstances.Regina Lawrence (2000) has previously developed this idea as the expressive value of police use of force (i.e., not as an instrument but rather as an end in itself): during encounters with citizens, violence can give a lesson about police authority while demonstrating who is in charge.In situations where authority is threatened, participants of this study clearly exhibited approval of violence whether it was seen as acceptable or not: "So, I give a legitimate and clear order, and the subject does not obey, so he ends up reacting against me aggressively, and on purpose (…) Almost all of these aggressions, let us say 99.9% of them, the police response will always go beyond what is strictly necessary.This is the most honest way I can put it…" -

PSP1
Resistance and dispute of powers is taken as the triggers to a 'game of forces' that, in police officers' perspectives, must adamantly be won by the 'good ones' (Westmarland, 2002, as cited in Green & Ward, 2000).Separation and isolation between the police and non-police audiences continue as authoritarian postures are reinforced (Paoline, 2003).

Conclusion
This study highlights how police officers' perspectives concerning police violence are comprised of inherent contradictions of representing the state as 'power-holders' of coercive power.Moral contradictions are pervasive and represent the very foundation of violence professionalization.Our data show that whereas, on the one hand, the participants proved to be morally disengaged towards police violence, on the other hand they tended to view violence as moral wrongdoing.As previously argued, police officers are surrounded by several legal frameworks, moral convictions, and organizational and cultural factors, thus placing them in a dissonant and demanding moral position.
The separation between legitimacy and illegitimacy objectively ascribed (or not) to violence is highly problematic and difficult to solve.Moral disengagement processes can help us to understand the ways police officers deal with this moral discrepancy, mainly in situations where they feel professionally obliged to use force and/or to act violently.
Regarding the limitations of this study, since the data is exploratory, further in-depth studies must be developed in order to provide a more systematized analysis.Data generalization is not possible to achieve.Even so, some implications must be discussed.Firstly, our approach asks for a paradigmatic shift on how police violence is understood and studied because it requires a deep deconstruction and critique of the main conceptualizations (e.g., proportionality, minimum use of force, acceptable/unacceptable use of force) used to make sense of police work.
It asks for an analysis of police officers as power-holders of the state's legitimate monopoly of force.Police violence cannot only be understood by taking into account an enriched and systematic understanding of force triggers (e.g., contextual and personal characteristics enhancing the likelihood of a given police officer to use force) but rather by deconstructing police violence itself both on a macro and micro-level of analysis.
Secondly, current discussions on police violence have not accounted for how violence legitimacy is constructed within the perspectives of state-representatives themselves.Most of the studies in police violence do not problematize their own internal perspectives of violence.They are thus guided by the same privileged rhetoric that reinforces minimalistic consciousness and intense normative justifiability amongst police officers.The present study has thus highlighted the processes of moral disengagement and the conditions surrounding them that lead to the normalization and professionalization of violence in the oral perspectives of police officers.To a broad extent, this knowledge can promote a more critical understanding of police officers' everyday lives and therefore reduce the banality of violence within police organization and culture.In Portugal, this is so far the only work addressing this question from this point of view.
We recommend the extension and adaptation of our methodological design to other countries and/or the inclusion of other internal (e.g., military forces) and external perspectives (e.g., civil society).In future methodological scenarios, we also recommend the inclusion of female police officers in order to understand moral disengagement and gendered patterns.Since it was not possible to interview prison guards in this study, we also recommend the inclusion of this group because inmates' insubordination to prison rules can promote physical violence perpetrated by prison guards (e.g., baton beatings, punching, slapping) (CPT, 2013).To address this professional group, the interview guide used in this study -"Police Perspectives on Police Violence" (see Appendix) -can be a useful instrument given that it has a scenario, which is exclusively focused on custodial circumstances.
As police violence continues to be a controversial and all-pervading phenomenon, academic (external) critical analysis must strive for further efforts, which may question the naturalization, professionalization or habitus of violence within police forces.The psychological, ethical and moral-driven inquiry led in this paper follows these concerns.Although it is exploratory, it can establish the basis to work towards a systematized body of knowledge about violence professionalization and the respectively socio-moral, ethical and psychological implications related to disengagement and minimalistic conscious.

Notes
i) Distinctions between use of force and violence will be further developed in this paper.For reasons also pointed out later, the more inclusive notion of police violence will be favoured.
ii) For a conceptualization of the state, it is necessary to recognize how it works as a 'real illusion' in the sense that it must be acknowledged as a by-product (and as a producer) of concrete social relationships with real effects (Abrams, 1988).The illusory and naturalized existence of modern states required the development of a set of material conditions (e.g., institutions, agencies), which govern and work on behalf of the common good.Then, the state-idea becomes reified as it gains an autonomous symbolic identity (ibid).Despite the differences in 'how' powers are developed within more autocratic to more liberal states, the Weberian monopoly of the legitimate use of force (coercive power) (cf.Weber, 1964) is a shared feature amongst states that represent the ultimate and maximum capacity to make use of force (Green & Ward, 2004).
iii) Internal perspectives of police violence tend to rely heavily on the interconnected notions of proportionality and minimal use of force to determine the limits of acceptable and unacceptable use of force (Morgan, 2000).Supposedly, proportionality is examined by a 'force continuum' (McLaughlin, 1992) in which the police officer's response should comply with the target's resistance or non-resistance.The 'force continuum' is then an assessment in order to use the minimal force needed to deter a target.Accordingly, this examination is processual and dynamic in that "it should provide a means for escalating force when the subject shows noncompliance and a means for de-escalating force when the subject complies" (ibid, p. 65).
iv) It should be noted that even a legal-oriented perspective can be used to contest the harm committed on behalf of the state.The strategic use of the notion of state crime within criminology is a clear example of such an attempt (e.g., Barak, 1990Barak, , 1991Barak, , 2010;;Kauzlarich, 2007;Matthews & Kauzlarich, 2007).Once every state can be seen as a 'juridical person', they can be made responsible for violations of domestic and international laws (Chambliss, 1989).But this legally based notion encounters many difficulties to problematize state-caused harm because it relies heavily on legal terminologies and procedures.Within domestic and international legal realms, the notion of crime concerning authorities/states is supplanted by the notion of 'human rights violations'.As a result, even if referring to misdemeanors, 'state crime' is more a criminological concept instrumentalized with the aim of recognizing and questioning state powers, controls, abuses and so on, than a legally recognized concept.Thus, this notion mainly emphasizes a doctrinaire, theoretical and strategic use of the notion of crime, rather than the realization of a strong mechanism of contestation and punishment.In addition, legality is a double bind because a state's legitimacy depends on the power to legally reframe and legitimize the use of force exercised in its name (Chomsky, 1999).v) Violence professionalization follows the dominant logic of today's labour market but applied to police forces in terms of training, segmentation of work, dominant masculine sociability and hierarchy, verticality in the decision-making processes, religious respect to top-down rules, and performance management, among others.Key to these processes is the assimilation of violence just as a working tool and police officers' self-realisations as the qualified masters of the professional ethics of violence.According to Huggins et al. (2002), in the case of police agencies, violence professionalization is achieved through "specialized training in a particular body of knowledge, a rigid division of labour, hierarchy of decision making, self-imposed occupational standards, and impersonal and universalistic rules for appointment, promotion, demotion, and remuneration" (p.207).

Table 1
Scenarios Discussed per Participant