Situating Zero-Tolerance Policies

Research into the effects of suspensions, expulsions, and zero-tolerance policies on a safe school climate is largely focused on American cases, since zero-tolerance policies in the US predate their Canadian counterparts. The studies demonstrate the ineffectiveness of these disciplinary mechanisms and the disproportionate way they are applied to racial minorities, students with disabilities, students who are defined as having behavioural problems, and lower class students (Dunbar and Villarruel Reference Dunbar and Villarruel2004; Morrison and D’Incau Reference Morrison and D’Incau1997; Noguera Reference Noguera1995; Skiba et al. Reference Skiba, Michael, Nardo and Peterson2002; Skiba and Peterson Reference Skiba and Peterson1999, Reference Skiba and Peterson2000; Verdugo Reference Verdugo2002). The Canadian literature is sparse, but reiterates the inequities that zero-tolerance policies produce and exacerbate (see Bhattacharjee Reference Bhattacharjee2003).

For example, Giroux (Reference Giroux2003a) argues that zero tolerance is a war on youth and symbolizes that young people are ‘disposable,’ while Hyman and Perone (Reference Hyman and Perone1998) posit that this disciplinary practice represents a new type of victimization of students. An extension of this critique is that zero tolerance becomes a way for young people—particularly racialized young people—to ‘prepare for prison’ (Hirschfield Reference Hirschfield2008) and strengthens the link between the justice system and the school (Kupchik and Monahan Reference Kupchik and Monahan2006). These characterizations of students being repeatedly victimized by zero-tolerance policies miss the nuances of how the safe school student is produced.

The concept of schools as producers of docile and disciplined bodies has a long theoretical lineage (Foucault Reference Foucault1977; Jones and Williamson Reference Jones and Williamson1979). Raby (Reference Raby2005, 84) concludes that a “docile, productive citizenry is thus envisioned, with those Others who fail to self-govern or to display prescribed self-respect disciplined through more sovereign applications of tools such as the zero-tolerance policy.…[T]here is little in these rules to suggest an active citizenship based on involvement in decision-making, challenge to the status quo or authority, independent thought, equality or genuine democracy.” Arguably the TDSB zero-tolerance policy is more than a uniform sovereign display of might, instead acting as a contradictory and fluid range of student subject positions. The application of the SSA by principals suggests that this policy can be understood as governmental and embodying a neo-liberal governmentality (see Garland Reference Garland1996; O’Malley Reference O’Malley1999; Rose Reference Rose, Barry, Osborne and Rose1996; Simon Reference Simon1994). Contrary to Raby’s assertion that ‘independent thought’ is absent, principals and codes of conduct are actively engaged in ‘making up’ (Hacking Reference Hacking1990) students as having the capacity to choose. The school is no longer framed as solely responsible for producing good citizens (Garland Reference Garland1996). In this unloading of responsibility from the school, students are ‘made up’ as being able to choose to misbehave. While zero-tolerance discourse has been subsequently removed from the current SSA by a different political government, the data in this article are from a particular historical period when SSA was experienced and interpreted by the TDSB as a zero-tolerance policy by principals and those training principals. While the policy has changed, analysis of the era in which the original SSA was in place provides context for a more fruitful mapping out of later legislative and procedural trajectories.

The SSA was enacted in Ontario during a period of widespread educational changes enacted by a new government. The Conservative Party won a majority government on June 8, 1995, using a ‘Common Sense’ platform and immediately made broad changes to education policy and practice. The pattern and rapidity of change in education in Ontario (see Dehli Reference Dehli1996; Majhonovich) was consistent with other governance transitions in Australia and New Zealand (see Dale Reference Dale1994); in Europe and Finland (see Rinne Reference Rinne2000); in the UK (see Woods Reference Woods2002); in the US (see Apple Reference Apple2001a, Reference Apple2001b); and in other Canadian provinces such as Alberta (Taylor Reference Taylor2001). Two important structural changes were ushered in by the Fewer School Boards Act and the Education Quality Improvement Act (see also Gidney Reference Gidney1999).

The purpose of Bill 104 (The Fewer School Boards Act) passed in 1997 and Bill 160, The Education Quality Improvement Act (1997), was to change the way education was administered in Ontario as it reduced the number of school boards, and school boards could no longer raise their own revenue. School trustees had their numbers reduced and also their salary slashed from $48,000 to $5,000 stipends. The Education Quality Improvement Act resulted in province-wide job action. Education in the late 1990s in Ontario was a volatile environment, and teachers were often depicted as lazy, self-serving, and in need of rigid forms of surveillance. In neo-liberal times, governments formulate the rules and then step back and let the contest play itself out (Pratt Reference Pratt2000). However, and rather paradoxically, the Ontario government also conferred enormous power upon itself in the sphere of education. Principals were also removed from the teacher’s union and became more of a managerial class. This time period was characterized by extensive changes in the administration of education resulting from the desire to, as the then Education Minister suggested, “create a useful crisis” in education (see MacLellan Reference MacLellan2009). School safety itself would similarly become a crisis point.

This article ^{Footnote 3} draws on interviews with eleven TDSB public school principals who worked with seventh and eighth grade students during the 2002–2003 school year. The principals represent a fairly diverse sample of schools in terms of size, make-up of student population, and geographic location. The interview questions were designed to understand whether or not the SSA had changed principals’ daily practices, and to explore what principals perceived as the purpose of the SSA. The interviews focused on whether or not the SSA affected principals’ practices of suspending and expelling students, their perceptions of the advantages and/or disadvantages of the legislation, and difficulties with the implementation and operationalization of the SSA.

I also interviewed three Child and Youthcare Workers (CYW) who work with children identified as having behavioural problems. CYWs are not based at every school. Two are from the central district and work at a limited expulsion school, while one is based in the east end and works out of five grade seven-to-eight schools.

The three policy contacts within the Toronto Board of Education whom I interviewed are either workers in the Safe Schools Office (an office which oversees the administration of the SSA), or senior level social workers within the Toronto District School Board. I also visited the Support Program for Expelled Students (the name of the limited expulsion school) on two separate occasions and sat in on the daily practices.

Throughout the paper I refer to these interviews using codes. P stands for principals (n=11), SS stands for Safe Schools Administrators and policy contacts (n=3), CYW stands for Child and Youthcare Workers (n=3) and CC for codes of conduct (n=7). ^{Footnote 4} From the eleven school principals interviewed, seven codes of conduct were obtained.

Using Common Sense to Accomplish Safety

The creation of the SSA was part of the Common Sense revolution in Ontario. It reflected the belief that public schools were no longer safe and that students were the source of this instability. Comments such as “our schools are by no means, however, the safe havens they once were” (Walker and Eaton-Walker Reference Walker and Eaton-Walker2000, 46) preface any discussion on creating safe schools. These observations are not open to debate—we are told it is imperative to accept this as fact before we can try to solve the ‘problem’ (Walker and Eaton-Walker Reference Walker and Eaton-Walker2000, 2). The early 1990s promulgation of safe school literature packaged in a textbook how-to format (see Duke Reference Duke2002; Hill and Hill Reference Hill and Hill1994) further reinforced the common-sense view that schools are unsafe places for youth.

In Ontario, calling what was an amendment to the Education Act the ‘Safe Schools Act’ helped legitimize the belief that, prior to the legislation being enacted, schools had become unsafe. In fact, the long title of the original SSA was “An Act to increase respect and responsibility, to set standards for safe learning and safe teaching in schools and to amend the Teaching Profession Act.” Safety concerns ballooned after school shootings in Columbine and Taber. In Canada, those pushing for the creation of the SSA often pointed to the United States as an example. As Newman, the first politician to propose the SSA, argued:

Newman used common sense to assert that acts in schools were becoming more violent; he took it to be self-evident. Furthermore, Newman implied that without addressing school safety now, Canada would soon have similar “explosions” in school violence.

The TDSB purported to deal with the problem of unsafe schools through the use of zero tolerance, stating that “commitment to zero tolerance is central to the Safe Schools Policy” (TDSB 2001, B.1:3). The notion that safety would best be accomplished through zero-tolerance strategies that would “deliver a range of appropriate consequences for inappropriate behaviour” (TDSB 2001, B.1:3) legitimized the perception of schools as unsafe zones and students as those who needed to be disciplined. One principal described their school’s need “to remain free and public, but no less safe than private schools” (P-6). One of the implicit arguments for the SSA was the belief that private schools were better equipped to assure student safety than their public counterparts. The recurring story from administrators, the government, and principals was that the SSA could help accomplish safety.

The SSA reflected “the government’s belief that parents wanted school staff and administration to have more control and the perception that [youth] violence is increasing, which is true” (SS-1). While the first statement was prefaced by it ‘being a belief of the government,’ the interviewee’s belief in it being a ‘true’ statement is contrary to much research (see, for example, Doob, Marinos, and Varma Reference Doob, Marinos and Varma1995; Doob and Cessaroni Reference Doob and Cesaroni2004). Since increased violence is taken for granted as true (a ‘common-sense’ fact), the SSA was a very specific reaction to combat violence and produce safety in schools. As one principal put it, the SSA was a way for the school board to get serious “especially after Columbine” (P-6). Implicit in the use of Columbine (a high school in Colorado where two heavily armed students killed twelve students and one teacher) as an example of school violence is the assumption that the SSA will prevent that extreme form of violence. Common sense thus permeated the discourse on who was responsible for the lack of safety in schools—they are not the safe havens they once were because students themselves are increasingly choosing to misbehave, sometimes in grisly ways, in the face of systemic policies. The policy reiterations at the provincial and the individual board and school levels created this image of the student as the agent of unsafe schools.

However, the only response to school violence instituted in the SSA was providing specific guidelines as to when principals must suspend or expel a student—called mandatory expulsions or suspensions. As an administrator of the Safe Schools Office in the TDSB noted, “there is not enough [in the SSA about] prevention” (SS-1). Although one of the stated purposes of the SSA was to “encourage the use of non-violent means to resolve conflict” (SSA 301.(1)), there was nothing explicit in the SSA on how this could be achieved and what programs could be implemented; the government left that responsibility to individual school boards. School exclusion was the original thrust of the SSA, and only one principal (P-2) discussed something other than suspensions/expulsions when asked general questions about the SSA.

By outlining suspension and expulsion criteria, the SSA implied that schools were unsafe and needed to be managed through exclusionary circuits (see Rose Reference Rose2000). But framing the SSA as another example of how young people are disposable (Giroux Reference Giroux2003b) misses how risk and choice are emergent in a school setting. As Johnston (Reference Johnston2000, 67) argues, “Zero-tolerance policing is also a discursive device for describing a wide range of actuarially-driven police practices orientated towards the management of risk…” (emphasis added). Similarly, while zero tolerance is a way for an institution to show that something is being done about the perceived problem of violence, most researchers overlook the specifics of how safety can be produced. The assumption that schools were increasingly unsafe, coupled with an exclusionary legislation, opened up gaps for students to take up multiple subject positions beyond “enemy” or “disposable.”

Making the Rules Clear: Producing Responsibilities and Positioning Students as Choosers

According to a Safe Schools Administrator, the SSA created clear rules and consequences, formalized the process of suspension and expulsion, sent the clear message that safety is important, and protected the victim more than in the past (SS-1). It becomes ingrained as common sense that rules and consequences need to be made clearer to students so that they know what is appropriate behaviour and how to act upon that knowledge.

This marks a shift from when education became mandatory, by the twentieth century, and schools operated in loco parentis. According to Simon (Reference Simon1994, 18), in loco parentis in legal discourse was premised on two principles: “First, children and adolescents were special legal subjects whose lack of adult judgment mitigated against the extension of full liberties of adult citizenship and the imposition of the full burdens of adult responsibilities.… Second, that, as special legal subjects, children and adolescents required governance” (emphasis added). Governance implicitly comes from an external source, which makes the relationship between self-discipline and in loco parentis a very tenuous one (see Platt Reference Platt1977; Raby Reference Raby2005). The Ontario Regulation 298 s. 23(1) and the “Requirements for Pupils” illustrate the tension that emerges when competing rationalities of governance co-exist ^{Footnote 5} :

Students are told to exercise self-discipline while the doctrine of in loco parentis is in effect. This marks a shift to recognizing the subjectivities of students as rational choosers who can exercise self-discipline while also accepting discipline from teachers as parent figures. Policy reiterations such as these ensure that ‘choice’ is adopted in the lexicon as ‘common sense.’ The student is no longer to be subject to discipline but to also be the master of their own discipline. This shift also means that when behaviour breaks down, focus turns to the student, and not the school’s inability to guide the student.

The school is trying to produce a self-governing subject (see Rose Reference Rose1999; Turnbull and Hannah-Moffat Reference Turnbull and Hannah-Moffat2009). Schools were always ‘in the business’ of disciplining students, but the shift of responsibility from the school to the student is an explicit command. The theme of responsibilization (see Garland Reference Garland1996; O’Malley Reference O’Malley and O’Malley1992) was replicated in each individual school’s code of conduct where students are told to ‘be safe’:

• – ‘play safe games’ (CC-12);

• – ‘practice safe behaviour all the time’ (CC-9);

• – ‘act safely in all situations’ (CC-11);

• – one can ‘contribute to [the] safety of others by being aware of [their] actions’ (CC-6); and

• – students should ‘conduct themselves safely in school and work’ (CC-5).

Only one code (CC-8) did not legislate students to act safely. Safety is not just the business of the school—youths are told to be active participants in producing safety.

Codes of conduct predominantly listed rules and expectations for students, but one code had a section on what the student should expect from teachers. Teachers at this school were told to be sensitive so that “mutual trust and respect is developed between students and teachers” (CC-5). This expectation reinforces the teacher’s important role in creating a “warm and caring environment” (CC-5), which harkens back to the school operating in place of the parent. However, another item explained that students should expect teachers to “encourage the growth of confidence, responsibility and self-discipline” (CC-5). Empowerment in this particular context reinforces the power relation between teachers and students while seeming to redress the power differential (see Hannah-Moffat Reference Hannah-Moffat2000). ^{Footnote 6} The teacher’s responsibility is to create a space for the pupil to become responsible and self-disciplining. The onus to maintain discipline is not solely on the teacher. The teacher, historically envisioned as a substitute parent (Platt Reference Platt1977), no longer has the same parental responsibilities over the child. Whereas parents are often held blameworthy for actions committed by their children, the teacher’s responsibility for the student’s behaviour is minimized through this promotion of self-discipline.

The responsible student in the TDSB has many characteristics that reflect this negotiation with the older welfarist logics as well as the technologies of choice. Responsible pupils are dressed appropriately, orderly, and well-groomed (see also Raby Reference Raby2005). They are also gendered and racialized. For example, rules on gossiping promoted a gendered subject by the differential presentation of the male bully, who waits in the schoolyard to pick a fight, versus the female bully, who spreads rumours and bullies emotionally. Bullying behaviour of young women is presented as something that is less visible, and therefore more insidious. Headgear, bandanas, hats, and do-rags are not permitted in the majority of schools and are associated with gang affiliations. The do-rag restriction symbolizes the common-sense intersection of gang, race, and gender because the do-rag is worn almost exclusively by young black men. The codes thus seek to reduce risky behaviour with normative and common-sense assumptions about what is gang paraphernalia. Not everyone who wears a do-rag is in a gang. But the decision to wear a do-rag is presented as irresponsible, since teachers and principals can misinterpret it as gang behaviour.

The responsible student is also someone who knows (and complies with) the rules (CC-11, CC-12). The majority of principals interviewed noted that codes of conduct were distributed widely to students, as if this had manifestly increased their school’s safety. Some codes were placed at the beginning of the child’s school agenda, while some principals explicitly told parents to go over the codes of conduct with their child. Students even needed prove their knowledge of the codes of behaviour through their signature. Other codes stated they would be prominently displayed and “effectively communicated to all and understood by students, staff, parents or guardians, and the community” (CC-8). Softer ways of telling the pupils to know the rules also existed: “[students should] familiarize yourself with the following school rules and be sure you understand why they are a necessary part of a safe school environment” (CC-7). The explicit command to “know the rules” coupled with the detailing of consequences (CC-7) constitutes children as rational actors and frames the principal-pupil interaction differently than the substitute parent with a “special legal subject” as under in loco parentis doctrines. Safety is thus connected to knowing the rules; if the rules are known, and if they are legislated to be known, the school will be safer because students will act upon this knowledge and make safer choices.

While it is not obvious that actually knowing the regulations helps youths avoid all forms of disobedience, the need to make the rules known reflects a common-sense belief that people, and increasingly adolescents and children, will ‘follow the rules’ if given full information. ^{Footnote 7} Putting school disobedience in the realm of ‘common knowledge’ through explicit direction to the students and parents, as Valverde (Reference Valverde2003) has argued, makes them responsible for this knowledge, whether they have read the rules or not. Furthermore, since the rules are in the realm of ‘common knowledge,’ principals believed this created a “level playing field” (P-4). The playing field represents this belief in common knowledge as a field of action. Misbehaviour is out of the expertise of the school and in the hands of students and parents (see Valverde Reference Valverde2003). The common knowledge of safety thus operates as a method of empowerment, a technique of governing through freedom (see Cruikshank Reference Cruikshank, Barry, Osborne and Rose1996; Hannah-Moffat Reference Hannah-Moffat2000). With the emphasis on knowing the regulations, any behaviour that breaks the rules is set up as a wilful disregard for the behavioural codes—a student will be disciplined when he has chosen to disregard the rules and must then be held accountable for this choice.

It is through making safety common knowledge that choosing to act safely is also put in the realm of common sense. As one principal commented, “very few choose to be bad, and if you know it [the code of behaviour], it is preemptive. It helps them stay out of trouble if they are serious about school” (P-6). This rationality creates a different subject: an ‘other’ that needs to be targeted. While some transgressions will be accepted as normal, the problem student is one who cannot be pre-empted through knowing the codes of behaviour because they will choose to be bad. This implies that no form of discipline or correction on the part of the school will make any difference. As Garland (Reference Garland2001, 187) notes in regard to crime control practice, “the mundane opportunist offender now stands much closer to centre-stage…and the needy, pathological offender much less prominent.” If the student cannot be held accountable, or is resistant to being held accountable, the student is positioned as an opportunist who has successfully eluded the purposes of school discipline.

Safety has moved beyond something that the school is supposed to ensure—pupils must now be active participants in school security. Students are thus constructed as capable of making decisions to be responsible individuals. For responsibilization techniques to work, the student must be made up as a subject and conceded a degree of freedom, which is a different configuration from an in loco parentis orientation. The student cannot be viewed as needing to be rescued if they are misbehaving; rather, students are supposed to be self-governing and governable through their freedom (Rose Reference Rose, Barry, Osborne and Rose1996, Reference Rose1999). The freedom accorded to students in schools is linked to the discourse of choice in schools. Students are not helpless to their peers, evil teachers, or the environment—their potential is wrapped up in their own ability to choose to be good. Conversely, students can also choose to do bad things, and therefore choice becomes a site to secure the school and manage bad students. Since every student is told to know the code and every student is responsible for contributing to a safe environment, any contravention of the regulations is positioned as resulting from the student making an informed decision.

Teachers and principals are still an integral part of providing a good education but cannot be the sole guarantors of a good education, as students are instructed in the codes of behaviour to be responsible for “learning” (CC-9) and “developing talents and gifts” (CC-7). The pupil who resists school is not simply in need of more discipline in the hopes of being saved and made by teachers into better people as the child savers thought (Platt Reference Platt1977). Instead, the problematic pupil is chastised for not being an active participant in her own education. The focus for readjusting behaviour weighs on the choices the student makes. It is only through students choosing to learn that the child savers’ dream of producing better citizens can occur. For example, one code (CC-12) states that a positive contribution to the school community “can only be based on self-discipline as well as mutual respect for each other.” The student has the capacity to choose to make this positive contribution. School authorities can only be relied upon to do so much.

As governing through choice emerges in discourse, it re-codes past governing strategies such as the needy pathological offender Garland discusses. For instance, if a student continues to be disruptive, then the student is at fault for “not accepting feedback of correction” (P-6). The notion of correction, a tactic used during disciplinary regimes (see Foucault Reference Foucault1977), gets re-coded as a choice. Correction is available to those who choose to accept the feedback and act upon it. This is evidenced in one principal’s belief that suspensions are “ineffective because students aren’t held accountable—they can just shrug it off” (P-2). By governing students through choice, the disciplinary role of the school is subsumed. No longer are they students in need of correction, but students who need to be held accountable for their behavioural decisions.

One principal remarked that “most know when they are doing something wrong and chose not to control their behaviour” (P-12). This comment is interesting in light of the mitigating circumstances that had to be considered before excluding students. The regulations (106/01 for mandatory suspensions and 37/01 for mandatory expulsions) stated that the suspension/expulsion is not mandatory if:

‘Losing control,’ which is supposed to be considered a mitigating factor in exclusionary decisions, was couched by P-12 as a rational choice. The ironic result is that although a student may not be in control of her/his actions at the time of the infraction, he has chosen to be in that state. Thus, governing through choice is one way for the system to negotiate the liberal educational pull to include, and the zero-tolerance push to exclude, students. It repositioned how safety was envisioned and accomplished, and diminished the responsibility of the institution.

Governing Through Choice: Reassembling Risk and Safety in the School System

The systemic implications of producing student subjectivities through technologies of choice evidence a shift in responsibility away from the school. I have termed these developments as schools governing through choice instead of Simon’s (Reference Simon2007) conceptualization of governing through crime, because the former better accounts for the dual shift of institutions managing blame, and the latter assumes there is no economy of blame. Power (Reference Power2004, Reference Power2007) has argued that institutions are increasingly interested in reputation risk management. In the case of education, it means that school authorities will continue to create rules for appropriate behaviour, but that it is incumbent on students to make the choice to become well-behaved citizens. The school as a reputation risk manager recognizes that students who misbehave still deserve an education, but the onus shifts to students who must choose to comply with school rules. This enables the education system to avoid blame when an education is not provided to difficult students.

Partnerships, emblematic of the broader neo-liberal shifts in governance (see Rose Reference Rose, Barry, Osborne and Rose1996), are central to the production of safety. ^{Footnote 8} Rhetoric about partnerships was evident in the second policy component of safe schools in the TDSB, ^{Footnote 9} where successfully implementing the safe schools policy hinged on partnerships with the school community (TDSB 2001, B1:3). Even the ability to ‘solve crimes’ on school property required partnerships (TDSB 2001, A.4:8). This assumes that everyone who is a ‘stakeholder’ is responsible for the system. The school should no longer be conceived of as the sole provider of safety within its walls.

The school system cannot make good on the promises of the ‘child saving’ era in much the same way the criminal justice system cannot solve the problem of crime (Garland Reference Garland1996). The principals echoed the institutional reality that responsibility for churning out well-adjusted citizens has moved beyond the conception of the school as a disciplinary machine popularized by Foucault (Reference Foucault1977). Thus, choice became a way to manage the reputational risk of the school system because it suggested the machinery still worked—but only if given a particular and programmable subject.

Discussion of accountability and meaningful consequences for the student does not necessarily translate into an increase in the use of suspensions and expulsions for every case, as the governing-through-crime thesis would suggest. As one principal remarked, “there are times when you want the child to be completely accountable for their behaviour and suspension is not the way to do it” (P-2). Suspensions may be the easy way out in these instances, but holding students responsible is the guiding principle behind the disciplinary decision.

The provincial legislation was explicitly about keeping schools safe through the use of exclusionary techniques similar to the tactics employed in shopping centres and other neo-feudal spaces (see Shearing Reference Shearing2001). Yet, the senior school cannot be easily compared to these spaces because there was an underlying desire to keep senior school students in school. One principal said: “Our object is to keep kids in school even when you read it [the SSA] and it sounds like getting rid [is the objective]” (P-12). Thus, risky students needed to be understood as containable rather than simply excludable.

Some students have behavioural problems that by definition make them riskier. But they are not easily excludable even under the SSA because, as one principal rhetorically asked: how can she suspend everyone when youths “are not on a level playing field” (P-9)? This principal’s concern was noteworthy given the criticism of the discriminatory impact of the SSA on minority children (see Bhattacharjee Reference Bhattacharjee2003; McMurtry and Curling Reference McMurtry and Curling2008; School Community Safety Advisory Panel 2008) and children with learning or developmental disabilities (Orwen Reference Orwen2003). Principals were therefore aware of the inequities a strict application of the SSA would magnify. ^{Footnote 10}

Another respondent said, “If you know the child has the potential for behaviour problems, the role is to prevent and be aware. We [may be] aware but cannot provide constant supervision. Repeated unsafe behaviour builds a case for different placement; if the behaviour is consistently risky [then they are] not in the right placement” (P-5). There were other factors in play beyond banishing students from the school system entirely. Even if the pupil is repeatedly presenting unsafe behaviour, the school still tried to find the right placement. One principal said that “these guidelines are there to help us but at the same time you cannot forget that the student is a human being and…sometimes the needs [of the student] don’t go with the guideline’s expectations” (P-7). Rather than a rigid interpretation of the SSA and TDSB policies, principals actively negotiated between the discourse of choice and the school as disciplinary machine by talking about the needs of students.

Governing through choice also reconstitutes all students as potential risks. Every student becomes a suspect, as they are all capable of making bad choices. Pupils are increasingly seen as potential risks to the institution (Kelly Reference Kelly2000, Reference Kelly2001). However, viewing students as potential risks is an effect of governing through choice, not the cause. ^{Footnote 11} The aim of education is to foster good choices so that students may become predictable. Risk is reassembled in governing students through choice. Risk, in this sense, stems from the space opened up by constructing students as free subjects. Risky students are those who do not choose properly.

Uncertainty is ever-present in schools, and students involved in serious matters comprise a particular future risk. As one Safe Schools administrator suggested: “if he did it today why wouldn’t he do it tomorrow” (SS-1)? A major effect of the SSA was an increased institutional need for surveillance, which is achieved by establishing a more extensive paper trail. Informal suspensions/expulsions were no longer permitted because, as one principal put it, “the board’s recommendation is to do the paper trail because if this kid turns out to be the kid in high school who’s bringing a weapon to school, everyone is going to want to know what he was doing last year and the year before” (P-1). The paper trail becomes a mechanism to produce knowledge of risk about students who have already seriously misbehaved, and to insulate the institution so if ‘he does it tomorrow’ he will already be documented. In establishing this narrative of badness the principal is also protected, because the student’s future misbehaviour can be traced to earlier problems instead of having an uncertain origin. ^{Footnote 12}

The difference between this conception and the idea of dangerousness as explored by Castel (Reference Castel, Burchell, Gordon and Miller1991) is that risk is not just embodied in a particular sub-population, but in all young people. Beck (Reference Beck, Adam, Beck and VanLoo2000, 214) argues that “the concept of risk reverses the relationship of past, present and future. The past loses its power to determine the present.” While the focus of risk is on the future, the past—particularly in corrections and education—is ‘colonized’ by the future. The past becomes a predictor of future behaviour to manage an uncertainty that emerges from schools governing through choice. Safety and security become overriding concerns as the category of deviancy has been supplanted by the category of at-risk (Tait Reference Tait1995; Bessant Reference Bessant2001). Whereas deviance implies a cause that can be corrected, at-risk implies an uncertainty that needs to be addressed, but the outcome will still be an uncertainty. The best to hope for is a predictable or calculable uncertainty.

In regard to the mitigating factors from the original SSA, there appeared to be an internal contradiction. ^{Footnote 13} One Safe Schools Administrator (SS-1) contended that section c) worked as an override, and was always in effect. Thus, if the student posed a risk to themselves or others, she could still be suspended/expelled even if she was not in control of her actions. The assumption is, despite sections a) and b), that students have both the responsibility and ability to control their actions. The argument seems to nullify the mitigating circumstances because a pupil could pose a risk precisely because he was not in control of his actions, or was unable to appreciate the long-term consequences. One Safe Schools Administrator said that principals “think these [mitigating] sections mean that they should do nothing, when really they only have to consider the mitigating circumstances and not necessarily do nothing” (SS-1). Thus, these mitigating circumstances existed to show that the institution observed some semblance of due process and concern for human rights. Most principals did not think these sections meant they should do nothing, but they were ambiguous in describing how these sections worked. As one principal said: if students do not control their actions or are “normally not in control, they are a danger” (P-2). Thus, even though the section is organized to ensure that special-needs kids—who are not in control of their actions—could not simply be excluded from school, the fact that these kids simultaneously pose a safety risk meant that these mitigating circumstances possibly existed only in a rhetorical capacity. The inability of autistic children to potentially control their own actions—to effectively choose to do the right thing (or choose to do the wrong thing)—is particularly problematic for a regime governing through choice, and thus they seem to become more excludable. Since the autistic child is incapable of being governed through his choice-making, the potential risks he poses become more centrally placed.

This internal contradiction of choice and riskiness of students was also where resistance to zero tolerance was mobilized. One principal (P-1) recognized that not every child would pass the mitigating circumstances of the SSA because they may not fully appreciate the consequences of their actions—they are, in P-1’s words, “just kids.” This belief in the power of the mitigating circumstances could explain why this principal had never formally suspended a student under the SSA. ^{Footnote 14} P-1 resisted the primacy of choice as a mode of governance because students may lack the capacity to be the entrepreneurial subject that governing through choice envisions. It is crucial to keep in mind that while governing through choice is enabled in schools, it is still fluid, flexible, and particular to the school board, the school, and the principal.

Bowditch (Reference Bowditch1993, 500) argued that in making disciplinary decisions, schools “sought to punish ‘types of students’ more than ‘types of behaviour’”; but in the context of senior schools in the TDSB, it would seem ‘types of choices’ were being held accountable and targeted. It was the school system that posits misbehaviour as a rational decision made by students, which in turn made misbehaviour a threat to school safety. Thus, making students more responsible in actualizing their own self-discipline is the space in which the school operated to minimize risk from students, as well as the threat of the school system being held accountable for failures in discipline.

In institutional exclusionary programs, it was evident that choice was being acted upon. These were programs that showed the education system still had a duty of care to their students yet, repeatedly the message within these programs was that the student had to activate their own good behaviour to ensure school safety.

Targeting the Bad Chooser

When governing through choice breaks down, choice emerges in strategies to re-position students so they can learn how to make good choices. However, it is in these disciplinary programs that choice had a starker emergence, existing uneasily with other governing techniques. Principals were actively involved in the reconstitution of choice in schools on the ground showing how the rhetoric of choice became institutionalized as common sense. For example, one principal created a program for problem students he called ‘Second Choice.’ The use of the word choice instead of chance implies that the student, when committing the bad behaviour, was confronted with alternative paths and chose wrongly the first time. It also implies some tolerance of missteps, as long as they are within normal limits. Of significance is the description of the program as segregative and intensive, where assessment occurs in “such a way that we get to know [the pupils], get to know their families” (P-4). The program was geared toward knowledge production of the student and family circumstance, which was a way to responsibilize and hold the family accountable. Through partnerships, the family is brought in to be an active component in getting the student back on track and solving the problematic behaviour, thus reinforcing the idea that school safety is everyone’s business. The program was described as intense, “not intense meaning busy, but intense meaning we get a really close-up look at what they do, how they interact, some of their feelings and concerns” (P-4). It thus takes the form of remedial help, through segregation, with the underlying intent to get to ‘know’ the resistant student. Safety was not only achieved through partnerships and shifting responsibility back onto the student and the student’s family, but also through knowing and containing the risky behaviour. The slippage between helping the student for his own sake and helping the student for the sake of managing the risk to the school is crucial to understanding how principals negotiate safety.

The shift in empowering young people with rights and responsibilities is a necessary condition for a program such as Second Choice to be developed. Nevertheless, this program represents a reaction of one principal facing a dearth of available resources. Understanding a program that fits into the framework of choice and responsibilities—yet was not a TDSB mandated program—is important. It questions the heuristic strategy of grouping all new technologies of government under the auspices of a ‘Big Brother’ (which is the imagery of many of the zero-tolerance scholars discussed earlier), and instead represents something more similar to the surveillant assemblage (Haggerty and Ericson Reference Haggerty and Ericson2000) or governmentality approach (see Rose Reference Rose, Barry, Osborne and Rose1996). Multiple strands of surveillance or governing tactics start to co-merge and thus have the appearance of being a coherent top-down tactic, when in fact they resemble each other as they work from the bottom-up. Second Choice is a program developed on the ground. Choice has become an organizing principle for education but is not a legislated demand by the state. If we imagine that students are being governed through choice and their freedom to choose, as opposed to being disciplined through the welfarist model (see Rose Reference Rose, Barry, Osborne and Rose1996, Reference Rose1999), this makes the student speakable in new ways. These developments are linked through the new understanding of the student. As these codes and comments from principals have shown, choice is a site of producing security and safety in schools.

The provincially mandated “innovative” strict discipline project (Ontario Ministry of Education 2001) that handled students given a limited expulsion in the TDSB—termed the Support Program for Expelled Students (SPES)—further illustrates how students who makes bad choices are targeted. Only in British Columbia, Alberta, and Ontario were school boards legally required to provide alternate education for excluded children (Bell and Tremblay Reference Bell and Tremblay2001). The government recognized that something must be done with the student who is excluded. The ‘need’ to have programs for these youth partially stems from the perceived problem that kids kicked out of school cause more crime (Berger and Graham Reference Berger and Graham1998). As Janet Ecker, then Education Minister pointed out: “Students who are expelled or suspended from the regular school program must learn to deal with their disruptive behaviour problems, but they still need an education. Sending them out on the street only puts the problem elsewhere” (Ontario Ministry of Education 2000). Students have the responsibility to deal with their behaviour, but the school board still needs to deal with this problem.

The SPES was a formalized program for students given a limited expulsion. These students did not need to attend any programming and could wait for the expulsion time to pass outside of the school system (it was for a maximum of one year). However, it was strongly advised that students attend a limited-expulsion program such as the SPES, since their progression through the program was noted in school records and principals were aware if the student did not attend such a program or was consistently disruptive/truant (SS-2). Although there was more surveillance of the student in the limited expulsion program (and more resources), it was still voluntary to attend. Those who volunteered were positioned as taking responsibility for their actions and complimented for beginning to make the right choices. In this sense, it operated as a kind of forced volunteerism to weed out those students who remained non-committal to attending school. The student who attended this program also was more marketable to schools once the program was completed. The governance of students through choice was visible upon entering the SPES, where there was a statement on the front door to the classroom reading: “Not bad kids, just bad choices,” highlighting that students in this program have made the wrong choice. To anyone witnessing the program, the students behind the door were positioned as ‘others’ who needed to be acted upon and imbued with tools to make the right choices in the future.

The SPES was comprised of a 60% educational component and a 40% non-educational component (A.4: p. 9). The non-educational component was called ‘character education,’ evoking a re-moralization program focusing on development of personal and social skills. Expelled youths were targeted by the non-educational, moralizing component to develop new habits and new ways of thinking. However, these techniques of moral governance were pasted on governance through choice.

An open letter from a teacher to an anonymous student at the SPES (2002) read:

Two contradictory conceptions of youth co-exist in this letter. The student is recognized as not an adult because he is not 18 (in itself an arbitrary cutoff in terms of developmental psychology; see Grisso and Swartz Reference Grisso and Swartz2000), while at the same time it is assumed he is equipped with enough skills to be responsible, and to choose what type of adult he will become. Once again, the SPES teacher can only do so much for the student, unless the student is more co-operative. It is the student who assumes responsibility for being “serious about change.” The teacher continues to morally admonish the student, whose thinking and habits are not “in line with reality”; however, the student is discussed as actively making this decision.

These tensions were more evident in Chill Power, a program within the SPES. Chill Power explicitly encouraged pupils to move away from evil thinking and listen to their good conscience. The mascot of good conscience was a hip-looking, wise owl carrying some books, while evil thinking was represented by a single burning flame. These mascots were prominently displayed on posters in the classroom, reinforcing strategies of moral governing and choice simultaneously. Every student has a flame and an owl by their ear as a moral compass helping them to weigh the costs and benefits of their actions. Students who make bad choices, and thus fail to become governable through their freedom, have let their evil thinking guide them and that is why they now find themselves in the SPES classroom.

In Chill Power, students watched videos on various ‘risk-taking’ topics such as sex, drugs, and violence. Each video begins with the developer of Chill Power—who was also the SPES teacher—making anecdotal comments on the given topic about somebody who was doing well in life until he engaged in X, always with a horrendous result (such as death, disease, injury, police involvement, etc.). The video had role-playing sessions, and afterwards there were activities in workbooks for the students to complete.

While Chill Power can be read strictly as a re-moralization program for delinquents, it dovetailed nicely with fostering good neo-liberal subjects. This is not about moral governance but about producing a responsible subject who is capable of choice (see also Turnbull and Hannah-Moffat Reference Turnbull and Hannah-Moffat2009 for a similar meshing of governing strategies in female probationers). The SPES fits with Kelly’s (Reference Kelly2000, Reference Kelly2001) observation that initiatives targeting at-risk youth seek to produce the entrepreneurial student, or the pupil who is a good chooser through a wide assortment of strategies. If pupils are clean in person and habits (as the Requirements for Pupils words it), or made more moral, the actual space for choice is limited and they become more predictable. An entrepreneurial subject is celebrated by her capability to choose (for example, by choosing to behave appropriately or to attend the SPES voluntarily), thereby making the possibilities for action by the student seemingly endless. At the same time, the apparently limitless reign of the enterprising subject needed to be constrained in order for subjects to remain governable. In programs such as the SPES, students could be rendered more predictable in their ability to choose. Making students certainties enabled governance through responsible choice and freedom when they were returned to the general school population.