Are bullying and reproduction of educational inequality the same thing? Towards a multifaceted understanding of school violence

Violence is considered a major concern in society, specifically regarding school. In this article, the concept of school violence is explored. Different categorisations and manifestations of violence are presented, and the dividing lines between them are discussed. The questions highlighted in the article include the following: How is school violence defined, and what are the taxonomies of school violence? How and in what ways is school violence theorised and comprehended? The ways in which violence are characterised and comprehended vary between decades, contexts, disciplines and scholars. In general, violence is usually defined and can manifest itself through physical, relational, verbal, cyber and sexual expressions, and through systemic, structural, symbolic and objective forms. The former manifestations relate to more explicit and direct types of violence, which are also more often related to and included in the concept of school violence, while the latter manifestations focus on less visible and subtle forms. Definitions of violence are numerous yet highly overlapping, making school violence a complex and far from unambiguous concept that is problematic to define. Yet, the concept would benefit from including structural and symbolic forms of violence to avoid treating school violence as an individual issue at the expense of societal power relations underpinning school violence.


Introduction
Violence is usually defined and comprehended as actions or words that are intended to abuse, injure or cause pain to people or that are likely to damage something. Such a definition might also include forcing people to act or preventing them from acting in certain ways (Cambridge Dictionary, 2019;Oxford Dictionary, 2019). In a general sense, violence involves exerting power over others and can be interpreted as 'an exploitation of that primary tie, that primary way in which we are, as bodies, outside ourselves and for one another' (Butler, 2004: p. 27). Violence thus also refers to domination (Foucault, 1997) or the prevention of the potential to act otherwise; hence, power is the ability to act, and violence suppresses it (Arendt, 1970;Maze, 2018).
Violence is considered a major concern in society, particularly regarding school. School violence is a complex issue that encompasses questions about social and educational relations. Various kinds of violence have always existed in schools; for example, in the Swedish context, using corporal punishment to discipline students was institutionalised until 1958 (and remains in many countries), although its use started declining in the 19 th century Englund, 2016). School violence has been conceptualised, contextualised and expressed differently in various periods (Ellmin, 2008). Historically, school violence is almost exclusively situated regarding teachers' abuse of the students in their care, or teachers allowing violence to occur between students, for example, bullying as an approved form of violence for socialising new and younger students into the student community. The general picture today, however, is one in which students' violence against each other and against teachers is seen as the greatest problem Saltmarsh et al., 2012). The concept of violence has gradually been reconstructed; for example, the concept of 'illegal violence' has expanded and now includes more actions than it did a century ago (Wästerfors, 2016).
Much of the existing research on school violence is devoted to physical and verbal violence and conflicts between students and considerably to bullying . The Nordic countries particularly have been at the forefront of research on bullying. However, in recent decades, research on teacher-related violence, coarser forms of violence such as school shootings and school violence other than physical violence (e.g. cyber violence) has increased considerably (Saltmarsh et al., 2012).
Despite the long historical presence of violence in schools, an increased interest has been directed towards the issue only in fairly recent decades. During the 1990s, scholarly and public interest in school violence increased considerably, and the label 'school violence' started to be widely used to describe the occurrence of aggressive acts on school campuses (Furlong and Morrison, 2000). The issue of violence in schools gradually gained interest from researchers, practitioners and legislators. Numerous books, articles, studies and research projects on this issue have been conducted over the years, and seminars, conferences and journals focusing solely on school violence now exist (e.g. Journal of School Violence). This increased attention and interest has contributed to establishing a distinct research field that problematises school violence. This research field emphasises both empirical and theoretical issues and questions and focuses on victimisation and the extent, prevention and consequences of school violence for the students and staff of educational systems. This article is part of the emergence and reinforcement of the research field of school violence.
There is no overall consensus among scholars on how to define and classify violence or on what the concept might include. Although many concepts appear and reappear in scholarly discourse, they are not always clearly distinguished and often overlap. Also, diverse scholars operationalise concepts in various ways, particularly reflecting disciplinary fields, making it difficult to comprehend and summarise interpretations and results from studies in the field. Furthermore, definitions of violence depend on dissimilar contexts, traditions, periods and geographical areas. Different manifestations of violence can also have similar consequences, making it difficult to classify forms of violence or at least its consequences. For example, both physical and psychological expressions of violence may result in long-lasting somatic symptoms for the victim, thereby making psychological configurations of violence in some sense physical.
Common existing and recurring concepts that have been used in research to conceptualise school violence include aggressiveness, bullying, direct/indirect violence, physical violence, verbal violence, harassment, relational violence, cyber violence and sexual violence . This plethora of concepts reflects the difficulty in comprehending violence distinctively and in understanding the configurations and expressions that violence might include. It also reflects school violence as a somewhat descriptive and individualising concept, grasping more or less visible, easily identifiable and concrete forms of violence between individuals in school.
Accordingly, this article aims to explore, investigate and discuss the content and building blocks of the concept of violence, especially regarding school. Different categorisations, interpretations and manifestations of (school) violence are presented, and the dividing lines between them are discussed and problematised. The questions highlighted in this article include (1) How is school violence defined, and what are the taxonomies of school violence? and (2) How and in what ways is school violence theorised and comprehended? Through reading contemporary and significant Anglo-Saxon and European literature, particularly psychology and sociology, which use the concept to analyse different social phenomena, I aim to explore and problematise violence utilisation in school. I will highlight different empirical examples of violence and explore different aspects of it. I am especially interested in developing and emphasising the multifaceted, somewhat contradictory and ambivalent aspects of the concept. This article makes a contribution by bringing together, reviewing and problematising various complex aspects of violence. Additionally, it situates them within the school as an institution and explores the concept of school violence. By including more varied perspectives on the phenomenon of school violence, not least symbolic and other subtle forms of violence, the article further problematises and nuances what is often an individualised discourse on school violence. This article is divided into three sections. First, different taxonomies of school violence are presented and discussed. The focus here is on the thematisation and methods of violence identified, particularly in psychological research. Second, some generally defined discourses focusing on physical violence are highlighted and explored. Here, the main question is: What is physical violence actually a case of? Third, sociological perspectives on different levels of violence are presented, from so-called individual to structural and symbolic configurations and their relation to school. Finally, the conclusive statements made in this article are summarised and discussed, including some critical points on how to approach and comprehend the relationship between school and violence. This article is not in any way comprehensive; it emphasises, describes and discusses contemporary, significant and recurrent conceptualisations and interpretations of violence. Issues related to violence, such as prevalence, consequences, prevention and motives, are limitedly considered.

Is violence always physically painful?
Violence is a wide-ranging and ambiguous concept that includes many aspects and expressions. It may be seen in anything from subtle neglect, a quick push, a derogatory comment, threats, selfinjury, suicide, sexual assault or the act of being silent when witnessing ill treatment to a massive and long-lasting war that includes killing, terrorist attacks and genocide. Violence can be aggressive, vicious and extremely frightening, or, as in popular culture and sports, entertaining, thrilling and funny. It can be private or public, individual or collective, obvious or subtle and comprehended as 'good' or 'bad'. It can be a separate incident, such as a slap in the face or a recurrent pattern, such as bullying, harassment or domestic abuse. It can be legal or illegal and legitimate or illegitimate. In some contexts, violence may be perceived as enjoyable.
Violence is many things, but it is not everything, and forms of violence, such as physical violence, may not be as common as we imagine. In opposition to everyday depictions in popular culture and explanations by certain authorities of violence being easy and natural under specific circumstances, such as family dysfunction or poverty, American sociologist Randall Collins (2008) argued that actual violent confrontations are essentially quite rare and are avoided by humans, despite societal conflicts, inequalities and power relations. Collins claimed that violence is the exception, not the rule, regardless of the motivations and conditions. In section two of this article, 'Violence as a situation and relation', Collins's arguments are elaborated further.
Generally, the most typical and everyday way to comprehend violence seems to be to situate it in a physical frame. Research on school violence has traditionally defined violence as physical violence against a person or property, thus excluding many other aspects of violence in school, such as bullying, threats or more subtle configurations . Physical violence has been conceptualised in various ways by different scholars and in different studies, decades and disciplines. It is often defined as a person intentionally physically attacking someone else, which may include hitting, kicking or other forms of assault (Felix et al., 2009). Some scholars include violence against material artefacts (material violence) (Mynard and Joseph, 2000) or threats in the concept of physical violence , while others classify threats as so-called verbal violence. Threats can be expressed in ways other than verbal and are thus not always a form of verbal violence. For example, a threat could be written or expressed non-verbally through intimidating stares or gestures.
In contrast to physical violence, the concept of verbal violence relates to hurting someone with words (Dzuka and Dalbert, 2007). It may include degrading, teasing or insulting someone. As mentioned earlier, threats are occasionally included in this category, likewise rumours and racist and sexist insults (Olweus, 2010). Verbal violence hence overlaps with other categories, such as physical violence, relational/social violence and harassment . Different configurations of verbal violence, such as threats, are sometimes equated with psychological violence. However, all types of violence may have psychological implications, which might make this concept somewhat diffuse and superfluous.
Violence does not always involve hitting or insulting someone, for example. It can also target someone's personal relations, sense of belonging or social status. This type of violence is often labelled as 'relational or social violence' (Little et al., 2003;Espelage and Swearer, 2003;Povedano et al., 2015) and typically involves various kinds of exclusion, name-calling, rumour spreading, ignoring others, etc. Girls are more often victims of relational violence than boys . The same is true for so-called sexual violence. Although this concept can refer to various kinds of violent acts, definitions of sexual violence often assert that the perpetrator should have superior power, the victim should experience the sexual behaviour as unwanted and the behaviour should have negative consequences for the victim (McKinney, 1990). However, as Knight et al. (2011) and Göransson et al. (2011) argued, some behaviours in school that have been defined as sexual violence are normalised to an extent that may prevent the victim from describing them as undesirable. Examples of these behaviours include using different forms of sexist jargon, namecalling and invective.
If sexual violence is expressed specifically regarding a person's group affinity (e.g. gender or sexuality) and occurs over a period as a recurrent pattern, it is usually called 'sexual harassment' . Sexual harassment reproduces normative sexual and gender ideals and asymmetric power relations and thus maintains heterosexual masculine dominance over femininity (Robinson, 2005;Gillander Gadin, 2012;Messerschmidt, 2012;Conroy, 2013). Understanding the ways in which heteronormative masculinity is produced within aggression and hostility and plays out in school relationships is important for how school violence occurs (Saltmarsh et al., 2012). Sexual harassment may hence be described as a gendered and sexualised subset of the larger construct of violence.
The definition of harassment generally parallels that of bullying, although bullying is not necessarily expressed regarding a person's affiliation with a group. Often-stated criteria of harassment include recurrent and deliberately aggressive activities to hurt someone and the existence of an unequal power relation between the victim and the perpetrator. These criteria link the two expressions of violence, referred to as 'harassment and bullying' (Olweus, 1994;Craig and Pepler, 2007). However, scholars debate the issue of how to define terms, such as recurrent activities or unequal power relations, which makes these concepts difficult to clearly define .
Configurations of violence, such as harassment, bullying or sexual violence, can be expressed in various ways, such as verbally, relationally/socially, physically and non-verbally. These expressions of violence can also manifest themselves electronically through social media, for example. These configurations and expressions are seldom isolated phenomena, underlining the intersectional nature of experiences of different forms of violence. Where there is one form of violence there are often others, making different expressions of violence largely overlapping. Hence, one expression of violence can be a precursor or a correlate of another one. For example, research shows that among students being victimised on social media, most report also being bullied in several other ways (Cross et al., 2009;Pearce et al., 2011).
Many young people now spend a substantial part of their time online interacting with other people and developing social relations. This development has generated new forms of violence. As society changes, especially regarding technological development, the conditions for how violence is expressed also changes. 'Electronic aggression' may comprise many forms, such as organised criminals inciting violence, terrorists projecting force or recruiting into violent extremism, hate group activists encouraging interracial violence or sexual predators gaining access to potential victims (David-Ferdon and Hertz, 2007;Peterson and Densley, 2017).
Other forms of aggression and so-called 'virtual criminality' occur exclusively online (Grabosky, 2001). Some scholars have described how social media has become a vector for youth violence during the last decade and has changed the landscape of aggressive behaviour, resulting in so-called cyber bullying, harassment or violence (Patton et al., 2014;Diamond and Bachmann, 2015). This form of violence is communicated through modern information and communication technology and comprises, for example, threatening, sexual, insulting or offensive videos, text messages or photos delivered via social media. It may also include abusing someone's social media account to send messages from the victim's account (e.g. so-called 'facerape'), thus endangering the victim or damaging his/her reputation or relationships. What specifically characterises cyber violence and differentiates it from other types of violence is that offensive material posted online can harm the victim for a long time.
After this review of potential methods of violence in school, the next section of this article describes violence regarding situation and relation. These two perspectives seem to be connected in their approach, as both claim to grasp what violence is really a case of, how it is generally triggered and how it can be comprehended. In their explorations of physical and interpersonal configurations of violence, both emphasise rather explicit, contextual, social and micro-oriented types of violence.

Violence as a situation and relation
Amongst the most influential scholars problematising the concept of violence is the American sociologist Randall Collins (2008, also see Wästerfors 2016, who argued that the concept of symbolic violence (later discussed in the article) does not help 'at all to explain real violence but muddies the analytical task' (Collins 2008: p. 24). Unlike symbolic violence, Collins claimed that physical violence has a true core referent that can be studied using micro-situational observations. At the start of his comprehensive micro-sociological study, where he explored the concept of physical violence, Collins explained the importance of analysing violent situations: First, put the interaction in the centre of the analysis, not the individual, the social background, the culture, or even the motivation: that is to say, look for the characteristics of violent situations. That means looking for data that gets us as close as possible into the dynamics of situations. Second, compare across different kinds of violence. We need to break down the usual categorieshomicides in one research specialty, war in another, child abuse in another, police violence yet elsewhereand look for the situations that occur within them. Not that all situations are the same; we want to compare the range of variation in situations, which affects the kind and amount of violence that emerges (Collins, 2008: p. 1).
Although physical violenceat least interpersonal and not governmentally sanctioned violencerelated to certain categories and, for example, is more common among young men in school than among young women or people in any other category, not all young men are violent (Wästerfors, 2016). Alternatively, not all configurations of masculinity are (re)produced through violence (Messerschmidt, 1993(Messerschmidt, , 2004, although certain (hetero) masculinities (and occasionally femininities) are performatively produced and, if spoiled, reclaimed within the terms of physical aggression (Messerschmidt, 2012;Saltmarsh et al., 2012). Even if a young man exercises violence, doing so does not mean that he is violent most times. Conversely, it is not only young men who can be violent; women, children and older people are also sometimes violent. The fact that violence is statistically associated with certain categories is therefore insufficient to explain it and its prevalence (Collins, 2008;Wästerfors, 2016).
Furthermore, physical violence is expressed in myriad ways, making it difficult to use certain categories as general explanations for all physically violent courses of action. According to Collins, it is insufficient to explain violence via social stratification and different categorisations. Instead, the characteristics of violent situations should be the object of analysis to identify patterns in the interactions that may problematise simplistic and essential categories concerning those involved. The cause of violence is neither found 'inside' the perpetrators (i.e. in their sorrow, machismo or anger) nor in structures 'outside' them (i.e. their poverty, lack of education or exclusion). Rather, Collins (2008) argued that individuals are not violent; it is situations that are violent. The actual causes of violence are to be found in specific situations within the micro-dynamics of communicative interaction. Individuals using violence are thus violent during specific circumstances and situations that enable violent acts. Collins (2008) claimed that violence is the exception rather than the rule, regardless of motivations and conditions, and despite societal inequalities and injustice. He confronted theories that explain violence as something that is easy and straightforward to perform and situated violence as a concept opposing the human physiological 'nature'. Violence comes neither easily nor automatically, in Collins' view. Antagonists are afraid and reluctant, and their confrontational anxieties establish a powerful emotional barrier of fear and tension against violence. When violence actually occursoften in 'incompetent' and 'clumsy' ways, as most people fear violence and are not 'good' at itpathways around these emotional barriers must be found. Violence is thus equivalent to the different pathways or patterns that circumvent human barriers of fear and reluctance. One example of a pattern that triggers violence is a confrontation with a 'weak' victim who is in some kind of subordinate position, such as in domestic abuse or bullying in school. Another pattern is violence that is staged in front of an audience. The cheers of an audience can 'help' the aggressor circumvent the emotional barrier and provide him/her with emotional dominance about the victim.
Instead of focusing on violence regarding the situation or the specific circumstances that make an individual violent, Edling (2016) highlighted violence as a 'relation'. Notably, this perspective differs from relational/social violence, which was previously described as encompassing exclusionary practices, name-calling, etc. Violence as a relation seems to be more general and refers to violence as a social course of action; violence appears in interpersonal relationships and interactions between people. Similar to Collins's view, this perspective emphasises the social dimension of violence; however, it accentuates the relation rather than the situation. A social relation always comprises at least two subjects, and viewing violence as a relation specifically highlights what is occurring between individuals. This perspective explores the significance of violence as a relationship and examines how violence, as among many relations, interacts with other relations (Fiske and Rai, 2015;Edling, 2016).
Imagine Individual A bullying Individual B at school because A wants to confirm his/her friendship with Individual C, who is an antagonist of B (Edling, 2016). This example provides a snapshot of a social and relational phenomenon in which violence is present among several relationships that interlink the subjects (Fiske and Rai, 2015;Edling, 2016). This relational model assumes that humans' social life is coordinated by four fundamental motivessolidarity, hierarchy, equality and proportionalitywhich condition our values and actions (Rai and Fiske, 2011). Using this model to analyse the example above, the action from Individual A is motivated by A's need to express solidarity with C. Violent courses of action may thus be considered 'morally right' if they are motivated by conceptions of violence that justify its use in balancing social relations. However, this justification is of course highly subjective and related to how concepts such as violence, justice, moral and the 'nature' of social relations are interpreted. It is certainly debatable whether violence can ever be morally right, regardless if it is comprehended as individual, situational or relational. Yet, this model can be used to theorise expressions of violence that are conducted to undermine social inequalities or other forms of injustice, such as the organised and violent resistance exercised by social, independent and liberation movements (Lilja and Vinthagen, 2009). Like Collins' model, this model shifts attention away from individuals and their characteristics; however, it also shifts attention away from the actual violent situation.
The next and final section changes focus and highlights different sociologically informed configurations of violence playing out on more structural and symbolic levels regarding schools.

Violence as a silent 'dark matter' in school
The old ironic saying that 'the law is impartial: both the rich and the poor are forbidden from sleeping under the bridge' is an implicit recognition of structural violence and the ability of society to both create and ignore it (Metta Centre for Nonviolence, 2019).
Violence is understood as more or less explicit and as actions that occur on dissimilar levels. School violence and the different configurations of violence exemplified above are often interpreted and manifested explicitly and situated within rather individualised and situational conflicts and relations at the expense of a structural and macro-oriented understanding of the social issues underpinning school violence. However, violence is not only comprehended and manifested on an individual level but can also be expressed as relating to groups, institutions or societies. Institutions, such as schools and other workplaces, may (re)produce oppression and hierarchical structures based on, for example, gender, race or sexuality, thus exercising violence on an institutional level. This is so-called systemic violence: Systemic violence operating in schools contributes to the construction of educational contexts as dangerous places. Systemic violence can be defined as any institutionalized policy, practise or procedure that negatively impacts or discriminates against disadvantaged individuals or groups (Saltmarsh et al., 2012: p. 4).
The manifestations of violence, whether explicit or implicit, often correspond to different levels of violence. The more structurally anchored violence is, the more implicit its manifestations seem to be. Consequently, violence on societal and cultural levels, which is connected to whole societies, nations or regimes, often manifests in subtle, indirect and normalised and hence not easily recognisable ways. This kind of violence includes societal norms and discourses, such as the normalisation of masculine hegemony or a contempt for 'weakness' (Henry, 2010).
To conceptualise societal and indirect aspects of violence, the Norwegian sociologist Johan Galtung coined and developed the concept of structural violence (Galtung, 1969;Gilligan, 1996;Winter and Leighton, 2001). This refers to a form of hidden and collective violence that is typically embedded in the structure of social and cultural institutions. Structural violence prevents people from accessing important rights, living out their potential as humans or satisfying their basic needs. It comprises prejudices and discrimination against people based on intersecting power relations and manifests in institutionalised racism, class discrimination, sexism, etc. It produces, among other things, social and political inequalities, inadequate education and unemployment. According to Galtung (1990), structural violence is legitimised and justified through cultural violenceaspects of a culture, such as religion or science, which makes structural violence socially acceptable. Under systems of structural violence, 'good' people may participate in actions that harm other humans without intending to do so, simply by performing their routine obligations, perhaps reproducing what Hanna Arendt called 'the banality of evil' (Arendt, 1963). Although structural violence is constructed differently than direct and physical violence, as it works slowly and erodes human values, these forms of violence are inextricably related to each other. Accordingly, aggressiveness and violence that occur in local institutions, such as schools, are inseparably intertwined with overall societal and structural expressions of violence.
Significant to the discussion on hidden and indirect forms of violence, especially regarding schools, is the subtle and unrecognised violence that falls within the concept of 'symbolic violence' (Bourdieu and Passeron, 1977;Bourdieu, 1989Bourdieu, , 1991Kupfer, 2015;Oxford Reference, 2019). In the Bordieuan sense, symbolic violence is a legitimised process referring to the imposition on dominated classes/groups by the dominant group/class of an ideology. Through this process, which is legitimatised by concealing the power relations that are its basis, the dominant group legitimises its own interests and cultural capital, giving them the presence of the natural way of being in the world (regarding correct class, ethnicity/race, moral, style, etc.). Symbolic violence thus refers to 'the violence which is exercised upon a social agent with his or her complicity' (Bourdieu and Wacquant, 1992: p. 167). Alternatively, people contribute to reproducing their own subordination by internalising naturalised structures that tend to subordinate them.
In some sense, the concept of 'symbolic violence' seems to theorise an attained hegemonya 'soft' power that maintains its authority without the need for violent coercion, which produces complicity (the effect of power beyond dichotomies, such as coercion and consent) on the part of the population (Laclau andMouffe, 1985/2001;Bourdieu and Wacquant, 1992;Bourdieu, 2000;Connolly and Healy, 2004). Symbolic violence is produced through language, and its effect is the naturalisation of discourses that legitimate power relations and the domination system (Bourdieu, 1991;Recuero, 2015). Violence, violent expressions or, according to Butler (2013), injurious speech can thus be a product of discourse and categorisations; that is, violence can stem not only from what is said or from the meaning of what is said, but also from a system of knowledge produced by what can and cannot be said and can be controlled through certain socially accepted rules (Foucault, 1972). When sexism or racism is expressed through cyber violence or sexual harassment, for example, stereotypical discourses on gender, sexuality and race are reproduced and naturalised, thus creating a 'knowledge' of how things are (Foucault, 1972;Recuero, 2015). Hence, any time sexism or racism is expressed in school, symbolic violence is reinforced. However, symbolic violence situated in school is not necessarily included in research, policies and discussions regarding the concept of school violence or how school violence is constructed, situated and comprehended (Herr and Anderson, 2003). Accordingly, school violence is constructed as a somewhat descriptive, limited and individualising concept, grasping only visible, easily identified and concrete forms of violence between individuals in school, while symbolic violence is merely an abstract construction not materialised in the everyday life of children attending school. Bourdieu and Passeron (1977), however, showed how symbolic violence emphasises the central role of education in maintaining power relations of domination and social inequality. Teaching occurs in a society embedded in power relations and is used by the powerful to secure their privileges. For example, symbolic violence occurs when children are labelled as, for example, failures in schools or when their cultural capital is unvalued and unrecognised because it might differ from the teacher's. Hence, curricula and teaching content are not neutralall knowledge is socially created and thus related to power hierarchiescreating 'valuable' and 'non-valuable' knowledge (depending on class and culture). Accordingly, students are produced within unequal power relations in school, which strengthens inequitable environments and, hence, violence in school.
While Bourdieu emphasises the process of naturalising power relations, the Slovenian philosopher Slavoj Zizek (2008) interpreted symbolic violence as a representation of a subtle and objective type of violence that sustains the status quo. Zizek was interested in the ways in which certain types of violence are represented and perceived in society. He argued that objective violence refers to the invisible violence inherent in the 'normal' state of things. This violence is embedded in the structures of society, including fundamental political and economic institutions, and generates inequality, poverty and outbursts of subjective violence. 'Objective violence is invisible since it sustains the very zero-level standard against which we perceive something as subjectively violent' (Zizek, 2008: p. 2). The invisibility of objective violence thus enables us to recognise subjective violence, which is a visible and physically constituted dimension of violence that can relate to a recognisable activity, such as terrorism, beating or rioting.
Since it is visible and seems to occur against the backdrop of nonviolence, subjective violence is perceived as abnormal and problematic; it draws popular attention and removes our focus from objective violence, which Zizek considered to be the main problem. However, subjective violence not only distracts us from paying attention to objective violence but is also generated from objective violence. Objective violence has no clear perpetrator, so we do not recognise it. People may consider objective violence to be a peaceful and non-ideological normality attacked by subjective violence, which is performed by 'evil individuals' and 'fanatical crowds'. When subjective violence erupts as a reaction to a brutal system (i.e. objective violence), we blame the perpetrators of the subjective violence rather than the system. This process hinders us from recognising how objective violence actually explains and causes subjective violence. Objective violence thus needs to be acknowledged as commonly accepted in society to be challenged (Zizek, 2008).
Systemic violence [objective violence comprises symbolic and systemic violence] is thus something like the notorious 'dark matter' of physics, the counterpart to an all-too-visible subjective violence. It may be invisible, but it has to be taken into account if one is to make sense of what otherwise seem to be 'irrational' explosions of subjective violence (Zizek, 2008: p. 2).
Subjective violence, such as sexual and gender violence in schools or young people rioting in the suburbs, expressing frustration and claiming recognition, can thus only make sense regarding subtle and 'unmarked' (Haavind, 1998) objective violence, which invisibly exercise its impact via social, cultural and economic inequalities in society. However, research has shown that the normalisation of subjective violence (according to Zizek's definition) in young people's lives and the 'everydayness' of school violence have made seemingly visible forms of violence more or less invisible or unremarkable in many young people's lives and among school staff and educational researchers (Saltmarsh et al., 2012).

Concluding discussion
In this article, various types and perspectives of (school) violence are presented and discussed. The concept of violence has been shown to include various forms and manifestations. The ways in which human aggressiveness and violence are characterised and comprehended vary between decades, contexts, disciplines and scholars. Yet, generally, violence is defined and can manifest through physical, relational, verbal, cyber and sexual expressions, and through systemic, structural, symbolic and objective forms. The former manifestations relate to more explicit, concrete and direct types of violence, which are also more often related to and included in the concept of school violence, while the latter focuses on less visible, indirect, structural and subtle forms. This division partially depends on whether the definition in use is psychologically or sociologically produced. Different scholars within disciplines have also put various forms of emphasis on various perspectives. This article contributes to the literature by bringing together these numerous aspects of violence in relation to the concept of school violence. Including a broader set of perspectives on violence into the comprehension of school violence as a phenomenon not only problematises simplistic views of school violence as merely explicit and individual, but also helps to better understand school violence and thereby increases the possibilities to counteract it.
Upon reviewing these different comprehensions of violence, obviously, the definitions of violence are numerous, not clearly differentiated from each other and often overlap. The complexity of violence seems to relate not only to different academic disciplines producing different definitions but also to the definitions being linked to and focusing on different, often highly varied, yet inextricable aspects of violence.
For example, the methods and intentions of violence are often used to define violence; a physical attack is defined as violence intended to cause pain or injure someone. Indirect and direct violence are defined through the relationship of the subjectswhether the victim knows who the perpetrator is or not. Relationship is also in focus in relational violence and when defining harassment, sexual violence and bullying, which presume an unequal power relation between those involved. Whether the violence is recurrent or not, which involves the extent and pattern, is also constitutive in defining different types of violence. Bullying and harassment, for example, presume a recurrent pattern of violence, as political violence such as long-lasting wars does. Whether violence is explicit or not, relating to visibility and normalisation, is crucial in clarifying differences between subjective and indirect types of violence, such as objective, structural and symbolic violence. Other criteria through which different categories of violence are defined and conceptualised in research are: consequences (e.g. sexual violence is partly defined by negative consequences), targets (e.g. harassment targets a person's group affinity), participants (e.g. spouses in domestic violence), situations (i.e. Collins' definition of violent situations), contexts/settings (e.g. cyber violence and school violence), levels (i.e. individual, collective, systemic and structural violence) and causes (i.e. violence as relation and situation).
More confusingly, several types of violence can be sorted into the same categories according to many of the criteria mentioned abovecriteria that also highly overlap. Most types of violence are thus inextricably related to each other. Consequently, the different definitions of types of violence are far from being unambiguous concepts and are problematic to define. Nevertheless, perhaps they can be brought together into a temporarily consistent concept by focusing on how violence actually works, how it is ontologically constituted and how it is typically comprehended, as follows. Violence hurts people or damages things. Violence relates to mechanisms that force people to act or prevent them from acting in certain ways or that otherwise circumscribe their potential, relative freedom and choices. Moreover, violence is often associated with some sort of action, active force or power that is destructive and has negative consequences. In short, violence is considered to be 'bad' and dangerous. Perhaps the most dangerous form of violence is the one that is hidden from viewsymbolic violence, which exercises its control invisibly because it is a part of the structure, hence normalised. If we do not perceive it, we can neither acknowledge nor undermine it.
To add to this confusion, Zizek (2008) argued that not all types of violence are completely bad and not all types are active. Whether an act is conceptualised as 'bad' and violent is always contextual. According to Zizek, if societal structures generate violence, then we should organise society in new ways rather than actively validating things as they are. As reproduced by the media and by economic and political institutions and their representatives (i.e. politicians, bureaucrats, etc.), objective violence thrives when people enthusiastically engage in the prevailing order. He argued that the truly difficult thing is to step back. Those in power often prefer even a 'critical' participation, a dialogue, to silencejust to engage us in 'dialogue', to make sure our ominous passivity is broken (Zizek 2008: p. 217).
Zizek concluded that if violence is progressive and disrupts the social order (including objective violence), then we are not violent enough. However, exercising subjective violence as an act of resistance against the prevailing order is not the correct thing to do, as this only confirms the naturalness, invisibility and superiority of the objective violence. How, then, can we counteract objective violence? By doing nothing, Zizek repliedby refraining from actions that are based on the system's premises. By exercising passive and negative activism and resistance rather than validating the status quo, we can violently disrupt oppression systems. The ultimate violent act is thus to passively stop an oppressive system from functioning or, as Zizek eloquently articulated, 'sometimes doing nothing is the most violent thing to do ' (2008: p. 217). Drawing from this rather unorthodox definition of violence, situating it within school, school violence could confusedly neither be, for example, physical or relational nor symbolic or objective. Instead, it would be considered the act of refraining from actions that are based on the system's premises, that is, the school. Consequently, school violence would in the Zizekian sense be the act of exercising passive resistance against, and hence disrupting (and reorganising), the oppressive school.
Regardless, according to Bourdieu (1998), all forms of violence are paid forthe structural and symbolic violence is matched sooner or later in the forms of overt and visible everyday acts of violence occurring, for example, in school. Schooling perpetuates social inequalities and hierarchies that underpin much violence in society and school. To return to the question in this article's title, bullying and the reproduction of educational equality are not necessarily the same thing or the same configuration or expression of violence. However, they relate to each other such that symbolic violence in school may (re)produce explicit forms of violence, for example, bullying, and vice versa. As Saltmarsh et al. (2012) and Herr and Anderson (2003) argued, not acknowledging and considering the symbolic and systemic forms of violence risk embedding school violence in a narrow psychological paradigm. Treating school violence at the individual behavioural level occurs at the expense of a structural and contextual understanding of the social issues and power relations that underpin school violence. It also occurs at the expense of understanding violence as a situation or relation, as discussed earlier. School practices and staff, cultures and policies, including dominant discourses on authority and discipline, that is, the institutional setting of the school, may strengthen inequitable environments in schools, resulting in escalating conflicts and inherently violent social situations and relations between students and teachers.
Consequently, addressing subtle and symbolic forms of violence in school and making them an integrated part of the discussion about and definition of school violence is crucial to avoid a reductionistic analysis of school violence. It will not only be empowering for students struggling with unequal social conditions but also enabling them to problematise and deconstruct power relations and feel validated and supported (Herr and Anderson, 2003). It will also help to comprehend and reduce the physical or other manifest and visible forms of school violence linked to the symbolic one. What then is symbolic violence in school, if not school violence? Violence in school? Or violent schools? Considering power relations in schooling, hence interrupting or exploring symbolic violence, is one step towards a more egalitarian and less violent school. Leaving symbolic violence unanalysed could probably itself become a matter of symbolic violence.

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.