City University of New York (CUNY)

CUNY Academic Works

Dissertations, Theses, and Capstone Projects CUNY Graduate Center

2011

A Systematic Review of the Literature on the Effects of School

Bullying from the Framework of Jurgen Habermas’s Theory of
Communicative Action
Gary Kogan
The Graduate Center, City University of New York

How does access to this work benefit you? Let us know!

More information about this work at: https://academicworks.cuny.edu/gc_etds/3653
Discover additional works at: https://academicworks.cuny.edu

This work is made publicly available by the City University of New York (CUNY).
Contact: [email protected]
A SYSTEMATIC REVIEW OF THE LITERATURE ON THE EFFECTS OF SCHOOL BULLYING
FROM THE FRAMEWORK OF JURGEN HABERMAS’S THEORY OF COMMUNICATIVE
ACTION

by

GARY KOGAN

A dissertation submitted to the Graduate Faculty of Social Welfare in partial fulfillment of

the requirements for the degree of Doctor of Philosophy, The City University of New York

2011
 ii

© 2011

GARY MARK KOGAN

All Rights Reserved

 iii

This manuscript has been read and accepted for the

Graduate Faculty of Social Welfare in satisfaction of the
dissertation requirement for the degree of Doctor of Philosophy.

Mimi Abramovitz PhD

Date Chair of Examining Committee

Michael Fabricant
Date Executive Officer

Darryl Wheeler

Elizabeth Danto

Supervisory Committee

THE CITY UNIVERSITY OF NEW YORK

 iv

Abstract

SYSTEMATIC REVIEW OF THE LITERATURE ON THE EFFECTS OF SCHOOL BULLYING

FROM THE FRAMEWORK OF JURGEN HABERMAS’S THEORY OF COMMUNICATIVE
ACTION

by

Gary Kogan

Adviser: Professor Miriam Abramovitz

This project is a systematic review of the US quantitative, empirical studies on the effects of
school bullying for the purpose of determining the degree to which Jurgen Habermas’s social
theory, the theory of communicative action, can be used to understand the constellation of
measured effects. School bullying is defined as a systematic abuse of power: the empirical
literature on school bullying, therefore, provides a large data set on the abuse of power. The
review finds strong consistency between the theory and the results of selected studies suggesting
that Habermas’s theory of communicative action can explain and predict the mechanisms through
which the bullying experience can affect the targeted child.
 v

Acknowledgements

This dissertation would not have been possible without my wife, Rachel Epstein, whose support
has been unending and whose editing skills carried this project through. Thanks to my son,
Huaquan, for being a good sport about the time I spent away from him on weekends, for being
my key informant on playground play, and for teaching me how to throw a spiral. Great thanks to
Professor Mimi Abramovitz who has guided this project since its inception and whose social
justice commitment and ability to see the simple solution at the heart of a problem is an
inspiration. Special thanks to Committee members Professor Elizabeth Danto and Professor
Darrell Wheeler for enriching and deepening this project. Thanks also to Professor Irwin Epstein
for showing me how existing data can be used in creative ways and to Professor Frances Fox-
Piven for showing how powerful ideas can change the world. A special thanks to the librarians of
the CUNY and Westchester library systems for tracking down countless out-of-print books and
obscure materials. This project is dedicated to everyone who has ever stood up to a bully.
 vi

TABLE OF CONTENTS

INTRODUCTION…………………………………………………………..…………………1

PART I STATEMENT OF THE RESEARCH PROBLEM………………………………….13

PART II REVIEW OF THE LITERATURE………………………………………………….35

PART III RESEARCH DESIGN……………………………………………..………..……..99

PART IV ELABORATION OF THE THEORY OF COMMUNICATIVE ACTION….......119

PART V SYSTEMATIC REVIEW……………………………………………………..…....183

PART VI RESULTS………………………..…………………….…………………..…… 196

PART VII SYNTHESIS OF FINDINGS………………………..……………….…..…..…229

APPENDIX I…………………………………………………………………….……..……258

REFERENCES…………………………………………………………………….………...267
 vii

INDEX OF FIGURES

Figure 1: Role Transitions…………………………………………………43

Figure 2: Habermas’s Typology of Social Action………………………...96

Figure 3: Habermas’s Descriptive Typology………………………...…..136

.
Figure 4: Instrumental Sources of Social Action………………...……….140

Figure 5: Social Action Outcomes………………………………….…….146

Figure 6: Contemporaneous Effects of Being Bullied……………...…….150

Figure 7: Breach Effects…………………………………………….…….153

Figure 8: Active Effects of Bullying………………………………….…...156

Figure 9: Concept Map of Effects………………………………….….…..208

 viii

INDEX OF TABLES

Table 1: Definition of Terms…………………………………………..16

Table 2: Explicit and Embedded Theory………………………………21

Table 3: Suggested Mechanisms of Effects……………………………33

Table 4: Levels of Visibility………………………………………….157

Table 5: Studies Selected for Review………………………………...196

Table 6: Typology of Variables………………………………………204

Table 7: Variables Measured…………………………………………205

Table 8: Results for the Spatial Dimension…………………………..206

Table 9: Results from the Physical Dimension…………………… …208

Table 10: PTSD Results……………………………………………. ..210

Table 11: Play and Activity Results…………………………………. 212

Table 12: Social Results………………………………………………213

Table 13: Cognitive/Temporal Results………………………………..217

Table 14: Affective/Temporal Results………………………………...219

Table 15: Academic Results…………………………………………..222

Table 16: Behavioral Results…………………………………………224

Table 17: Suicide Results……………………………………………..226

Table 18: Empirical Evidence for Theoretically-Predicted Results.…232

 1

INTRODUCTION

This study will build theory on the problem of school bullying1, defined for the

purpose of the study as the enduring impingement of the autonomy of one or more

children by one or more peers with greater power such that it is not feasible for the

victims to either stop or change the interaction2. The social theory of German

philosopher and critical theorist Jurgen Habermas--the theory of communicative action--

will be used as the conceptual framework for understanding the nature of bullying and its

effects. This project will attempt to determine whether and to what degree his theory can

explain all or part of the bullying phenomenon.

Bullying is a ubiquitous phenomenon within the developed countries in which it

has been studied, occurring at similar rates across social and economic strata (Nansel et

al., 2001). Estimates of the prevalence of bullying vary from 7% to 35% based on

different definitions of the phenomenon used in different studies, with variation in

prevalence most likely due to local school and community effects (Olweus, 1993; Sharp,

Thompson, Arora, & Cho, 2000).

Bullying among children is commonly characterized as ―school bullying‖ because

school is the venue where children, by legal mandate, congregate, thus school becomes

the focus of social scientific research for several academic disciplines. Children are not

free to withdraw from school of their own volition if they are being bullied, and most

parents have few schools they can choose from to send their children to. For many

children school is the only venue in which they interact with non-family members or

1
The term ―bullying‖ will be used in this paper in lieu of ―peer victimization.‖ The two terms are used
synonymously in the literature, but ―bullying‖ is the more commonly used term.
2
This definition was distilled from the various definitions and aspects of the bullying phenomenon
elaborated in the Definitions section below.
 2

people outside their religious or ethnic communities, making school the primary non-

family venue for social reproduction, the process of instilling social values, language, and

meaning (Habermas, 1987).

School bullying is a complex phenomenon that is comprised of several types of

behaviors that occur among schoolchildren for a variety of motivations (Arora, 1996).

Bullying takes place at different levels of social interaction, from one-on-one aggression

to, more frequently, group levels where a regular audience to the bullying behavior

gathers (Bukowski & Adams, 2005; Salmivalli, Lagerspetz, Björkqvist, Österman, &

Kaukiainen, 1996). Two factors distinguish bullying from other forms of peer aggression:

an imbalance of power between the bully and victim and the endurance of the bullying

over time (Arora, 1994; Craig & Pepler, 2003; Rigby, 2002b; Rivers, Duncan, & Besag,

2007; P. K. Smith & Sharp, 1994). The aggressive acts of bullying all share the quality of

enforcing a limitation on the victim’s autonomy in physical, psychological, and social

domains. The techniques of bullying are evolving with the availability of new technology

(Li, 2006; Mishna, 2007).

Three decades of research have shown that bullying is associated with an array of

debilitating consequences (Hawker & Boulton, 2000; Mynard, Joseph, & Alexander,

2000; Nishina, Juvonen, & Witkow, 2005; Olweus, 1980; Rigby & Slee, 2001; Erling

Roland, 2002; Sharp et al., 2000; Shellard & Turner, 2004) for both victims and their

bullies. Bullying contributes to problems with school climate, depressing academic scores

for the individual child and the school (Juvonen, Nishina, & Graham, 2000; Schwartz,

Hopmeyer Gorman, Nakamoto, & Toblin, 2005; S. W. Twemlow, 2004; Wentzel &

Asher., 1995). Bullying experiences have been found to be associated with depression
 3

(Haynie et al., 2001; Hunter, Boyle, & Warden, 2007; Morrow, Hubbard, Rubin, &

McAuliffe, 2008; D. J. Pepler & Craig, 1995; Erling Roland, 2002), anxiety (Baldry,

2004; Craig, 1998; Slee, 1994) ; aggression (Berthold & Hoover, 2000); and suicidal

ideation and attempts (Brunstein Klomek, Marrocco, Kleinman, Schonfeld, & Gould,

2008; Carney, 2000; Kaltiala-Heino, Rimpela, Marttunen, Rimpela, & Rantanen, 1999;

Omigbodun, Dogra, Esan, & Adedokun, 2008; Rigby, 1998; Erling Roland, 2002); and

avoidance of school for the victim (B.J. Kochenderfer & Ladd, 1996; Omigbodun et al.,

2008). Limited evidence indicates that bullying experiences, particularly on the part of

the aggressor, contribute to future violence (2007; Garbarino & Bedard, 2002; Vossekuil,

Fein, Reddy, Borum, & Modzeleski, 2002).

This study will build a theory about the mechanism of effect in bullying both to

guide social work practice in the specific problem of bullying and to contribute

knowledge to the field about dealing with other forms of oppression. By examining the

processes by which experiences of victimization affect children over time, this study will

address Finkelhor’s (1995, 2007a) suggestion that a theory needs to be developed to

explain how victimization affects development. Finkelhor uses the term ―developmental

victimology,‖ but he himself has not pursued this theoretical project. The research into

school bullying shows that many children who experience frequent abuse and humiliation

at the hands of someone more powerful than themselves will also experience changes in

their self-perception, sense of safety, mood or social life. These experiences can be

understood as forms of oppression which social work as a profession, and critical social

work in particular, is mandated to address.

 4

The desire to end oppression is found within critical theory and the related critical

social work. Walter Benjamin, an early critical theorist, felt that the social world should

be constituted in a way that allows people to construct their selves and relationships free

of domination (Meehan, 1994). Bullying can be understood to be one form of oppression

that can adversely affect children’s development. Intervening with bullying fits the

mandate of critical social work (Fook, 1993, 2002; Ife, 1997) which is an important

social work model in this writer’s native country of Canada.

This writer began his social work career working in programs that practiced

structural social work (Moreau, 1990; Mullaly, 1997), a model of social work that

integrates the analysis of power into the ecological model of social work; however, I

found little guidance about how to use the structural model in casework practice. Further

reading and coursework into postmodern writers such as Michel Foucault and Jacques

Derrida suggested to me directions and ways to make the power relations within social

arrangements visible for clients and the social work profession; however, nothing

appeared to provide an overriding theory that could be applied to social work practice. In

fact, most contemporary postmodern theorists argued against any project of theory

development, preferring a looser understanding of how knowledge develops on a local

level and according to powerful forces.

One difficulty that arises from postmodern relativistic thinking is its challenge to

the absolutes of ethical codes, such as those of the social work profession. In particular,

the structural social work idea of increasing the power of marginalized communities,

often called ―empowerment,‖ raises concerns about the social work profession’s

inadvertently encouraging individuals and communities to take power in ways that

 5

diminish the power of others, as in the unwarranted use of violence (Fook, 1993, 2002;

Ife, 1997).

An article, Concept, Act and Interest in Social Work Practice, by Mary-Ellen

Kondrat (1995) proposed that the German philosopher Jurgen Habermas had solved this

problem of relativism through what Kondrat termed ―mutual emancipation.‖ Kondrat

described a model of social functioning that requires social actors to recognize and

integrate the knowledge and interests of others in making ethical determinations.

Responding to an early work of Habermas, Knowledge and Human Interests (Habermas,

1971), Kondrat laid out a theoretical model by which human beings act through a set of

rational interests. Kondrat’s article led me into reading Habermas’s work in depth. In his

―Reflections on Communicative Pathology,‖ originally published in 1974 (Habermas,

2001b), Habermas analyzes the ways family dynamics, communication in particular,

affect the personality of the child. This article and Habermas’s subsequent theory of

communicative action comprise the foundation of the present study.

Despite Habermas’s primarily socio-political focus, individual, group, and family

dynamics are at the core of his project, in that he suggests a just society is only possible

in circumstances where individuals are able to consider the position of ―the other‖ and to

maintain communication in the face of conflict – skills learned and promoted in families

and communities. For Habermas, the ability to communicate in the face of conflict is the

apotheosis of social skills.

Habermas bases his theory of communicative action on the idea that genuine

communication involves a mutual exchange, a process of sharing that is distinguished

from speech and action designed to produce a specific action. In Habermas’s theory, this
 6

writer saw a theory that could explain how children develop a set of social competencies

to create the conditions for socially just communication. Habermas, despite his critics’

assertions to the contrary (Allen, 2007), also appeared to offer a way to understand how

power gets translated into personal and communal pathologies. Habermas, in his two

volume elaboration of the theory of communicative action, provides a schema of social

action that, to this writer, provided a comprehensive guideline for intervening with

bullying and other forms of interpersonal oppression which had been part of my social

work experiences.

This project will test the applicability of Habermas’s theory to explain the

mechanisms of effects of bullying on children’s functioning and development.

The Problem in Agency and Social Work Contexts

Research into school bullying has been conducted primarily by social

psychologists and educators; however, scientific inquiry into the topic is currently in

decline, as evidenced by a precipitous drop in scholarly publications on the subject to a

low of 136 peer-reviewed studies in 2005 (Stassen Berger, 2007). Despite a lack of

consistent scientific data, educators and social workers are constrained to act, sometimes

in the face of legislation, to end bullying (Limber & Small, 2003), with few scientific

tools to guide practice (Stassen Berger, 2007). Bullying is virtually invisible within the

sociology discipline and has only recently become a focus of research in the social work

profession (Mishna, 2003).

School bullying comes to the attention of social workers in various practice

settings. School social workers come into direct contact with bullying within the host

setting of schools and non-school social workers encounter the phenomenon in child and
 7

family treatment settings. Yet Mishna (2003, 2007) acknowledges that bullying has

received little attention within the social work profession until very recently. In recent

times, fewer social workers have been employed by school districts, and those that are

employed may be providing services to more than one school (NASW, 2011). This

paucity of resources limits the ability of many school social workers to spend time and

resources on bullying interventions. School violence, on the other hand, has been a

significant issue for school social work, with overt acts of violence that invoke

disciplinary and juvenile justice intervention absorbing a great deal of school social

workers’ limited resources (Allen-Meares, Washington, & Welsh, 2000).

Bullying prevention has not been part of school social work’s mandate because

the problem has not been universally accepted as consistent with the traditional mandate

of school social work to intervene only when an issue directly affects academic

functioning (Miller, Martin, & Schamess, 2003). Evidence is emerging, however, that

bullying does in fact depress academic functioning both for individual children as well as

classroom groups (S. W. Twemlow, 2004; Wentzel & Asher., 1995; Woods & Wolke,

2004); therefore, intervening in bullying should fall within the traditional school social

work mandate (Astor, 1998; Mishna, 2003).

A strong case, based on the academic and personal effects of bullying, can be

made for an expanded mandate for clinical professions to provide direct services to

children and their families within schools affected by bullying, including prevention and

primary treatment (Adelman & Taylor, 1998; Frey et al., 2005; SAMSHA, 2003). Such a

mandate, however, will require funding for social workers and faculty time for teachers
 8

to implement anti-bullying and social skills building interventions into every day

curriculum.

Mishna (2003) asserts that social workers are in the best position to provide

counseling services to children, and consultation and training for school personnel to deal

specifically with school bullying. The social work profession is already established

within the host setting of the schools, and it has a set of conceptual skills with which to

analyze multiple social contexts as well as a set of practical skills with which to intervene

to improve the school environment.

Bullying, as a set of oppressive experiences, can be seen as an issue of social

justice. Intervening against bullying is consistent with the mandate of Critical Social

Work. Critical social work (Fook, 1993; Healy; Ife, 1997), a dominant paradigm in

Canada, shares with critical theory an emancipatory mandate that requires taking direct

action against oppressive forces and ameliorating the effects of that oppression.

Scandinavian anti-bullying initiatives are based on an explicit project of social justice in

that reducing bullying is seen as a way to reduce marginalization among adult citizens

(Roland, 2000).

Alternative Views of Bullying and Victimization

For some children, the experience of bullying may not be harmful. Most research

into bullying is predicated on the assumption that bullying is a malignant phenomenon;

however, there is some evidence that some children who make the transition away from

the victim role, called ―escaped victims,‖ find that the bullying experience actually

strengthened their character (Dixon, Smith, & Jenks, 2004; P. K. Smith, Talamelli,

Cowie, Naylor, & Chauhan, 2004). Bullying experiences during childhood may also lead
 9

to the social reproduction of adult roles of submissiveness and domination (Dixon et al.,

2004; A. Sutton, Smith, & Swettenham, 1999), which may be considered as positive or

negative depending on one’s standpoint on roles of social control. One study (S. Brown

& Taylor, 2008) shows that adults who were victimized as children tend to be employed

in subservient roles more frequently than those who bullied or those who were neither

bullies nor victims.

Percy-Smith (2001) points out that systematic abuse of power is endemic to

capitalist societies which require people to take roles with varying degrees of coercion:

In contemporary social constructions, bullying is increasingly seen as a problem.

Yet a paradox exists in that, for example, the systematic abuse of power has
throughout history characterized many processes of social interaction, for instance
in the workplace by bosses, teachers in schools or by parents in the home, and
continues to do so in some cases (Percy-Smith & Matthews, 2001, p. 51).

Kogan and Chandan (2004) in their qualitative research into school violence

found a small number of teachers and parents who expressed opinions in favor of

bullying. Most of this group felt that bullying was an important part of childhood and

helped the children become tougher and more prepared for an adult world. One father

threatened to hit his child himself if he did not learn to stand up to his tormentor.

No research studies have yet been published that examine the relationship

between childhood bullying and adult violence and aggression3; however, German

educators during the Nazi era promoted bullying, without using that specific term, as a

way of preparing children to take on adult roles of oppression such as concentration and

death camp workers. Kamenetsky (1996) in her analysis of Nazi children’s literature

shows how bullying was promoted and taught as a way to reproduce adult roles of

domination and even cruelty.

3
There has been a longitudinal study that has followed Norwegian children into young adulthood.
 10

Bullying among adults in the workplace is another emerging topic of research

(Cowie, Naylor, Rivers, Smith, & Pereira, 2002; Hoel, Rayner, & Cooper, 1999; P. K.

Smith, 1997). However, it is not yet clear that workplace and school bullying are related.

Longitudinal studies begun in Scandinavia in the 1980s and 1990s are just beginning to

measure adult roles to determine whether children who were bullies or victims continue

their roles into adulthood: the longitudinal research cohort is just entering the labor

market so some answers to that question may be imminent.

Bullying and Social Competence

Social competent figures in several studies in this systematic review and is also a

fundamental concept in Habermas’s theory of communicative action. Social competence

is a diffuse concept that comprises a wide range of behaviors. Segrin (2000), in an

attempt to capture the common feature of all social skills, defines social competence as

the ability to interact in a way that is appropriate and effective. Within the empirical

literature, social competence – or social skills--is measured in several ways including the

child’s number of reciprocated friendships, subjective measures of the quality of the

child’s interaction, the ability to negotiate, sense of humor, perceived attachment, among

others.

Definitions and Methodology in Bullying Research

Despite the fact that bullying has been the subject of research for 30 years,

bullying researchers have not yet coordinated definitions or methodology. To date, there

have been no attempts to develop conventions that would coordinate definitions and
 11

methods to allow studies to build logically one on the other. This lack of coordinated

activity leads to a very confusing picture of bullying and its effects.

With governments and states implementing policy and laws against bullying

(Limber & Small, 2003), intervention programs are proliferating (Limper, 2000;

McGrath, 2007; Scaglione & Scaglione, 2006); however, there is scant evidence for the

effectiveness of these programs (Rigby, 2002a, 2003; J. D. Smith, Schneider, Smith, &

Ananiadou, 2004; P. K. Smith, Pepler, & Rigby, 2004; Stevens, De Bourdeaudhuij, &

Van Oost, 2000). Three meta-analytic studies concluded that there is no evidence to show

that any bullying intervention program has been effective and there is even some

evidence that certain interventions may actually increase bullying (D.J. Pepler, Craig,

Ziegler, & Charach, 1994; Vreeman & Carroll, 2007). By contrast, there is evidence that

violence, as disaggregated from bullying, can be both prevented and reduced (Mytton,

DiGuiseppi, Gough, Taylor, & Logan, 2006). This leads to the conclusion that bullying is

an intransigent may be a more damaging subset of aggression.

Dubin (1978) suggests that interventions in the applied sciences are unlikely to

succeed when there is no adequate theoretical explanation for the underlying

phenomenon. This appears to be the case with bullying. Many bullying intervention

programs are created with little or no reference to any supporting scientific literature

(Stassen Berger, 2007).

Abuse of power is the defining element of bullying interaction, yet none of the

theoretical models currently used in the bullying literature explicitly consider the

dynamics and effects of power. Power abuse is the single variable that distinguishes

bullying from other forms of aggression. Imbalance of power needs to be a significant

 12

feature of a theory that explains the bullying phenomenon and forms the basis of effective

intervention to improve outcomes for children involved in bullying. This project is an

attempt to mine the existing empirical literature to begin the process of building such a

theory.
 13

PART I STATEMENT OF THE RESEARCH PROBLEM

This study will examine whether and to what degree Jurgen Habermas’s theory of

communicative action (Habermas, 1984, 1987) can be used to understand the

mechanisms of effect of school bullying. The project will build a theory using data drawn

from a systematic review of literature on the effects and correlates of bullying.

Abuse of power is accepted by contemporary bullying researchers as the core

element and defining feature of bullying (Farrington, 1993; P. K. Smith & Sharp, 1994),

yet the processes by which this abuse of power affects children have not been explored

(Finkelhor, 1995; Morrison; Rigby, 2002b, 2003) . While Habermas’s theory of

communicative action does not use the discrete term ―power,‖ it does explore how action

is coordinated through differing means, including direct and indirect coercion and

strategic manipulation. It is this writer’s contention that Habermas’s theory provides a

conceptual framework with which to understand the complexities power and influence

through the lens of action coordination.

The theory of communicative action is the result of Habermas’s project of

―universal pragmatics‖ (Habermas, 2000) which specifies a set of logically irreducible

social competencies and social conditions that make knowledge and social action

possible (Habermas, 2001c; Heath, 2001). Universal pragmatics attempts to understand

and explain the social world with direct reference to social processes without the need for

metaphor (McCarthy, 1984).Biological metaphors, such as ecological theory that

compares the human environment with biological ecology, break down when applied to

human communities; for example, the boundaries in human communities cannot be as

 14

easily demarcated as the ecosystems of separate plant and animal species (McCarthy,

1984).

Habermas’s project also has a goal of building a just democratic constitutional

state. It might appear that the issue of school bullying, with its local, individual,

community level focus is remote from Habermas’s expansive goal; however, Habermas

shows that the basic competencies for building a just state are acquired in childhood.

Scandinavian anti-bullying projects are often predicated on claims that reducing bullying

will decrease the problem of marginalization among adult citizens (Roland, 2000).

Habermas himself applies his ideas to family dynamics analyzing communicative

pathology from a family systems perspective (Habermas, 2001b). In Habermas’s vision

of ideal speech conditions, he views the skills of meaning-making and consensus building

as essential components of ethical action in a just state.

Two sets of distinctions are central to Habermas’s theory. He separates social

action into ―instrumental‖ and ―communicative‖ action; and he separates the ―lifeworld,‖

the part of the social world in which meaning and conventions are formed through

consensus and discussion, from the ―systems‖ such as economic and administrative

institutions that are invented to solve problems of distribution in complex societies. How

these two sets of distinctions will form the basis for understanding the effects of bullying

in this project and will be more fully elaborated on below.

According to Habermas, communicative action is the source of myriad acts of

meaning making and solidarity; by nature it is a highly autonomous sphere of action.

Instrumental action, on the other hand, is an effective tool to meet specific ends and is, by

nature, low in autonomy. To Habermas, systems and instrumental actions are neutral
 15

concepts signifying the necessary steering mechanisms required to organize the lifeworld;

however, instrumental action misapplied can cause an array of social ills (Habermas,

1987). These abstruse concepts will be more fully explored in the theory chapter of the

actual systematic review.

Habermas’s typology of instrumental action corresponds exactly to the typology

of bullying actions that have emerged from a large body of studies that explicate and

delineate the bullying phenomenon. Free social play among children, in contrast to

instrumental action, shares with communicative action the defining feature of an absence

of instrumental motivations (Lever, 1978). These concepts will be briefly described in the

following sub-section.

The Major Concepts

Bullying is distinct from the larger set of behaviors that constitute ―peer

aggression.‖ The latter term is used to connote the full range of aggressive and coercive

actions among children that may or may not include bullying (Dupper & Meyer-Adams,

2002). In the theoretical and empirical research, the terms ―bullying‖ and ―peer

victimization‖ are used interchangeably. ―Peer harassment‖ is a sub-type of bullying and

denotes the verbal forms of bullying. For this paper, the term ―bullying‖ will be used to

denote a range of phenomena characterized by the abuse of power, including forms of

peer harassment.
 16

Table 1: Definitions of Terms

Term Definition

Aggression hostile or violent behavior

ambient violence hostile or aggressive behavior that is calibrated to

remain below the threshold of adult intervention

Bully a child who performs coercive or malicious acts

against (or infringes upon the autonomy of) a child
with lesser power

bully involved refers to children involved actively in bullying,

including victims and bullies

bully-victim a child who is bullied and also bullies other children

communicative action speech acts intended to build meaning and

solidarity (in its pure form, it is devoid of ulterior
motives)

indirect victimization bullying acts enacted by other people in ways that

hide the identity of the bully

instrumental action Speech acts intended to elicit specific actions in the

social world

Interaction a behavioral, affective, and cognitive event that

occurs between individuals

peer acceptance/rejection the degree to which a child is accepted or not into

social relationships and peer activities

peer aggression a set of aggressive and violent behaviors that

includes aggression among and between power
equals

power (similar to instrumental action) social action oriented toward achieving a goal

power bases the attributes that comprise the disparities in power

relational aggression attempts to strategically damage the social

relationships of another person

Relationship a succession of interactions between individuals

known to each other –mutually recognizing (Hinde,
1976)

rough and tumble play children’s play behavior that mimics aggressive and
bullying activities
(Table 1 continues on next page)
 17

social competence the ability to interact in a way that is appropriate

and effective (Segrin, 2000)

social withdrawal a child’s avoidance of social contact

speech acts vocal and physical actions intended to convey

meaning, build meaning, and instigate action in the
social world. It is inclusive of communicative and
instrumental action.

Teasing provocation directed at some act or attribute of the

recipient (Keltner, Capps, Kring, Young, & Heerey,
2001, p. 235)

victim or target child a child who is the recipient of bullying acts despite
a desire or attempts to stop these acts

Bullying is defined as aggressive or malicious acts performed by a person with

greater power against a person with lesser power (Farrington, 1993), or, as more

generally as the systematic abuse of power (P. K. Smith & Sharp, 1994). The aggressive

acts of bullying repeat or endure over time despite the victim’s attempt or desire to stop

the abuse.

Bullying is distinguished from rough and tumble play (P. K. Smith & Boulton,

1990) and playful teasing, both of which have features that resemble bullying in content

but do not cause harm and are accompanied by clear signals of support, such as warm

touch or facial expression (Keltner et al., 2001). Teasing is understood as intentional

communication that calls attention to an attribute of the recipient (Keltner et al., 2001).

Both rough and tumble play and teasing can move into direct aggression should the target

child misunderstand the intent of the play or should the play or teasing become too rough

and hurtful. Rough play, then, can turn into aggression or may be understood as

aggressive by one of the children.

Bullying involves the deliberate selection of a target with fewer resources of self-

defense. Bullying is different from ―aggression,‖ hostile or violent behavior which can
 18

arise reciprocally out of a disagreement or mutual antipathy among equals. Power

disparities, or ―power bases,‖ occur when the bully is superior in ―physical size, strength,

number,‖ or ―social status.‖ This writer has also observed, in the context of school work

practice, physically small children using power bases not covered in the bullying

literature. These smaller children may bully stronger or higher-status children through

audacity, their willingness to break social rules, or ferocity, their willingness to take

physical risks by being physically or verbally cruel. In one instance, a very small child in

a younger grade terrorized children three grades above him and over one foot taller in

stature by his ferocity (Kogan & Chandan, unpublished data).

Bullying children may also use their own powers of observation to find ways to

gain a power advantage. Craig & Pepler (2007) point out that knowledge of a victim

child’s particular vulnerability is a power base when it is exploited deliberately to cause

distress.

The description of instrumental power provided by Jurgen Habermas, although he

does not use that term, is action taken within the social world to achieve a goal. The

Habermasian concept of power will be elaborated on more fully in the conceptual

framework section as will the implicitly rational nature of Habermas’s social theory.
 19

Current Theories in Bullying Research

Several prominent researchers (Finkelhor, 1995, 2007a; Finkelhor, Ormrod,

Turner, & Hamby, 2005a; Ladd & Troop Gordon, 2005; Morrison, 2006; Rigby, 2003)

attest that there has been very little theory development into the nature of bullying, or into

the particular mechanisms of effect that operate on the children involved in various

bullying roles. The theories that have been most commonly used to explain the

phenomenon are social learning theory, social information processing theory, and

ecological theory. A fourth and incipient theory, developmental victimology (Finkelhor,

1995, 2007a; Finkelhor et al., 2005a) offers a tentative initial model for explaining the

mechanisms of effect on children of all power abuses, including bullying by peers and

abuse at the hands of adults. A small number of new or more obscure theories have also

been associated with bullying research; these and the theory of communicative action are

briefly reviewed here.

Social learning theory proposes that children are aggressive for the rewards that

accrue to them or are aggressive due to aggressive behavior learned in their families

(Bandura, 1979). Bullies may act out to assure their own safety, attain and maintain status

in their peer group (Hazler, Hoover, & Oliver, 1992); this theory tends not to be applied

to victims and the ill effects of bullying, although it can be assumed that some children

may be victimized due to a lack of social skills.

Social information processing theory (Crain, Finch, & Foster, 2005; Crick &

Dodge, 1994; Dodge, 2004; Dodge & Frame, 1982; Gifford-Smith & Rabiner, 2004) is

similar to social learning theory and suggests an internal, cognitive misrepresentation of

the social world. Children may bully because they misunderstand the cues available in the
 20

social environment; this inaccurate cognitive representation of the social world causes

them to attribute negative motivation to neutral or benign social actions which in turn

leads to aggressive and violent acts (Dodge & Frame, 1982; Peets, Hodges, Kikas, &

Salmivalli, 2007; Salmivalli & Helteenvuori, 2007).

Ecological theory is espoused by a number of American bullying researchers

(Blackburn, Dulmus, Theriot, & Sowers, 2004; D.L. Espelage, Bosworth, & Simon,

2001; Swearer & Espelage, 2004). To study bullying in its social context, Rodkin and

Hodges (2003) use the term ―peer ecology‖ for the various situations in which children

interact. A child’s particular friendship group may intersect with several contexts as

children form relationships with others encountered in different school and community

contexts, such as in the school classroom or in religious groups. The classroom itself is a

peer group which may include nested peer groups in the form of cliques or informal

friendship networks. While ecological theory helps understand the social contexts of

bullying it does not offer an explanation of bullying per se.

Table 2 lists the major and minor theories that have been applied to bullying.
 21

Table 2: Explicit and Embedded Theory

Theory Application to the Topic of Bullying

Cognitive Dissonance Bullying occurs due to distorted social cognition

Developmental Victimology Bullying affects both th present and future

functioning

Ecological Perspective Bullying occurs in a complex peer social

environment

Social Development Bullying adversely effects development of social

skills

Social Learning Bullying occurs due to lack of accurate social

knowledge and damages social knowledge

Social Information Processing Bullying damages the way we understand social

processes

Traumatology Bullying causes symptoms of traumatic stress

Theory of Communicative Action Bullying prevents social support and solidarity

Noguera (1995), using Foucault’s techniques of analyzing institutions (Foucault,

1977a) to explain the educational system, explores the structure of power inherent in the

organization of mass education. He argues that the present United States school system

emerged from an organizational model similar to that of prisons and mental asylums:

institutions that enforce rules, regimentation, punctuality, and order. His central thesis is

that vestiges of the social control element of these institutions remain extant in the

structure and culture of most schools today despite the efforts of well-meaning principals

and teachers who want to interact humanely with children. Noguera suggests that the

system of surveillance and control inherent in most US schools are in themselves

oppressive and contribute to the climate of violence which it is intended to quell.

There have been no major attempts to explain the reasons and mechanisms

through which children are affected by bullying. The theories and ideas discussed thus far

in this section explain the etiology of the bullying phenomenon and allow us to put
 22

bullying into the context of the social environment, yet none of the theories suggest a

mechanism by which bullying affects children.

Finkelhor (1995), noting that the lack of theory explaining the effects of power

abuses, including bullying, suggests that there may be elements that affect children

common to all forms of abuse. Finkelhor (1995) constructed a model which he calls

―developmental victimology‖ that proposes two axes of influence: effects that take place

in the present environment and those that affect the future. He explains that a child who is

abused may experience stress contemporaneously, and that this present stress may

impede the child’s ability to negotiate developmental stages which then affects her future

adjustment. While Finkelhor thus outlined the idea of developmental victimology, he has

not pursued it in his subsequent research.

Theories concerned with the mechanisms of traumatic stress (Vanerkolk, 1994)

posit that trauma arises from the strong emotions that pertain to experiences of

powerlessness resulting in a set of physical and psychological symptoms that can endure.

One of the hallmarks of traumatic stress is its potential to evoke an instantaneous and

monumental shift in the traumatized person’s understanding of the world. In that way,

trauma theory fits with other cognitive theories but it also shows how the body is affected

by such symptoms as sleep disturbance and anxiety.

In a single study, Mynard and Joseph (2000) demonstrate that specific

mechanisms are at play when different forms of power imbalance occur in bullying. The

authors show that relational bullying, where the bully’s identity is unknown (strategic

indirect bullying), is associated with higher levels of post-traumatic stress, while physical

bullying is associated with lower self-worth. Mynard and Joseph’s work appears to
 23

indicate that there are discrete mechanisms of effect at play in bullying experiences

depending on the nature of the bullying. In the years since Mynard and Joseph published

their study, however, no studies have either replicated or built on this discovery. This

lack of follow up illustrates Stassen-Berger’s (2007) assertion that bullying scholarship is

in decline.

This writer’s exhaustive archival search in the fields of social work, education,

urban studies, psychology, sociology, and social psychology has yielded no theoretical

literature dealing with the question of the power differential specific to bullying4. As has

been said, the theories to date used to explain bullying largely concern cognitive

processes (Crick & Dodge, 1994) and social learning theory (Bandura, 1977, 1979; Craig,

1998; O’Connell, Pepler, & Craig, 1999), but do not attempt to explain the salience of the

power differential components of bullying.

Catalano and colleagues (2004) describe a promising theory of social

development which is more expansive than most of the theories discussed above. Like the

theory of communicative action it suggests that bonding to a socialization unit, such as a

school, as predictive of future outcomes. The product of a 20-year longitudinal study, this

theory finds that positive bonding to a school is the most powerful resiliency factor in

determining positive outcomes for students later in life. It proposes a four-part model of

socialization that includes the development of social competencies, the perceived and

actual opportunities for interactions with others, and the rewards that accrue from

interactions. Social Development Theory has, to date, been applied only to the prevention

and treatment of delinquency but it may be able to explain some of the effects of

4
The field of geography has also neglected the study of bullying, but a small number of geographers have
begun setting a research agenda for the topic (Andrews & Chen, 2006)
 24

bullying, and, in fact, uses similar concepts to the Habermasian theory being used in this

project. Both bonding and social competencies are key Habermasian ideas which will be

applied to the problems associated with bullying in this project. Habermas’s idea, as will

be explored in the next section is able to explain both the etiology of bullying as well as

its effects.
 25

Theory: The theoretical frame that will be used to explain the findings

As stated above, German philosopher Jurgen Habermas is situated in the school of

thought called critical theory. Critical theory is strongly identified with the Institute for

Social Research which was founded in Frankfurt, Germany, in the 1920s, and is also

called the Frankfurt School, as are the group of thinkers directly or indirectly associated

with the Institute (R. H. Brown & Goodman, 2001). The term ―critical‖ derives from Karl

Marx’s critique of the political economy in his Das Kapital (Marx & Engel, 1955).

Within critical theory, the word ―critical‖ refers to the study of the mechanisms or laws

through which modern society was formed, with an emphasis on ameliorating the

deleterious effects of capitalism on the social world (Marx & Engel, 1955).

The central objective that critical theory advances is human emancipation through

a reordering of the social sciences into a unified project that makes ―sense of concretely

experienced problems and aspirations and (reveals) possibilities of future development‖

(Habermas, 1984, pp. 537-538). This goal of emancipation arose likely as the result of the

oppression and injustices that occurred through Nazi genocide and oppression in their

native Germany.

Jurgen Habermas considers himself a product of both Nazi-era and postwar

Germany. He openly discusses his membership in the Hitler Youth and the fact that he

briefly served as a child soldier in the German army posted on the Western defenses at

the end of the World War II (Habermas, 1992). Habermas dates his political awakening

to the immediate postwar period when he listened to the Nuremberg trials and saw

televised images of the Nazi concentration camps. His project can be summarized as an

attempt to determine the conditions upon which a democratic and constitutional state can
 26

be built that will protect the rights of minorities. The evolution of Habermas’s theory will

be outlined in the next section, including the method he used to develop his theory.

The theory of communicative action has never before been applied to the

problem of bullying or, in fact, social casework practice. Habermas himself applied it to

family dynamics in an early paper but generally applies the theory to the formation of just

and democratic constitutional states. This project will use his theory to derive concepts

appropriate to the understanding of the consequence of victimization and to build a

theory that explains how abuse of power can damage children both in their

contemporaneous functioning and in their development.

Summary of Habermas’s Theory of Communicative Action

Jurgen Habermas’s theory of communicative action offers a conceptual

framework that can deepen the understanding of how social forces affect the individual

(Borradori, 2003; Habermas, 1984, 1987; Honneth, 1992). Habermas proposes a set of

mechanisms though which action in the social world can be coordinated. In the following

section the system that he used to derive these concepts – universal pragmatics – will be

outlined.

Universal Pragmatics

The conceptual research technique at the core of Habermas’s theoretical project is

a process he calls ―universal pragmatics,‖ or the search for the specific set of logically

irreducible competencies and conditions that make knowledge and social action possible

and allow people to reach understanding in ways that are least distorted (Habermas, 1984,

1987, 2001a; Heath, 2001). Habermas sees consensual communication present in the
 27

social fabric in everyday acts of solidarity, friendship, and family life. As ordinary people

engage in daily conversation about life and the world, provide mutual support, and

negotiate common activities, they contribute to the development of conventions of shared

meaning. This often happens through a consensual process that functions reflexively

within the social world. According to Habermas, efforts to reach consensus are the ideal

form of communication and necessarily require the recognition of different points of view

(Habermas, 1984).

Habermas bases his analysis of social pathology on the ways that communication

diverges from the ideal of consensus (Habermas, 1987; Honneth, 1992). He proposes an

ideal set of interests at the core of social cognition and an ideal situation for consensual

communication (1987). For example, Habermas’s ideal speech situation is one in which

all speakers are recognized and have equal opportunity to express themselves. However,

Habermas uses the ideal condition as a heuristic, a device to assist in conceptualization; it

should not be deduced from this discussion that he believes the ideal condition is easily

or routinely attained in human interaction.

Speech/Acts

Habermas uses the term ―speech/acts‖ in place of ―communication‖ to denote the

utterances, written language, and actions intended to convey meaning and induce or

coordinate action, since, within his schema, not all speech/acts qualify as genuine

communication, i.e., consensual communication. Speech/acts are divided into two

categories: ―instrumental‖ action and ―communicative‖ action; the former includes

speech/acts intended to achieve a goal, while the latter includes the speech/acts intended

to achieve shared meaning or consensus – true communication, which is reciprocal.

 28

The idea of communicative action as speech/acts operating consensually

represents a paradigm shift similar to the introduction of the strengths perspective in

social work (Antonovsky, 1996; Saleebey, 1996). The strengths perspective challenges

social workers to privilege the strengths and resources of the individuals and

communities with which they work and marshal those resources to solve social problems.

Habermas suggests that the solutions to the ill effects of oppressive power involve

creating the conditions for consensus and solidarity, and applying them to the political

and social world (Habermas, 1987).

The Lifeworld and Consensual Process

The theory of communicative action proposes that the consensual intersubjective

processes at play in everyday life constitute, in themselves, the world of human beings:

he calls this the ―lifeworld.‖ Ritzer and Smart (2001), in interpreting the theory of

communicative action, define the ―lifeworld‖ as ―those interpretive patterns that are

culturally transmitted and linguistically organized, which…include the formation of

group identities and the development of individual personality‖( p.208). The lifeworld

includes the material, emotional and intellectual processes that comprise living organisms

and communities.

Habermas’s theory of communicative action is a social theory that elucidates the

fundamental internal behavioral controls and the communicative and interactional

competencies, or social skills, through which an individual negotiates the lifeworld. The

concept of lifeworld may be better understood in light of the French language translation

as ―monde vecu‖, or the ―world as it is lived‖ (Ferry, 1991), and in which the world of

meaning and language is formed in a process of intersubjective communication

 29

(Crossley, 1996). Ferry suggests that identity is formed in our daily interactions within

the lifeworld including in schools, and not exclusively in our intimate relationships.

Habermas makes a distinction between the lifeworld and the systems that are

created to organize activities in the lifeworld. Systems are created to fulfill a function

such as provide food or exchange. Habermas asserts that problems occur when the

constraints of a system supersede the needs the lifeworld. Nazism is an extreme example

of what can occur when a system uncouples from the lifeworld that created it. The

concept of system will be more fully elaborated below.

The theory of communicative action has the potential to provide a theoretical

framework for the strengths perspective in social work. Communicative action is thought

to contain myriad acts of support and repair. In this project, some of the mechanisms of

effect by which children have been harmed by bullying involve their alienation from

sources of support: this alienation will be more fully discussed in the systematic review

below.

There is some research that shows how mechanisms of support and solidarity

work in the real world, lifeworld. Granovetter (1973), though not citing Habermas or

critical theory, supports the notion of the importance of informal relations. He suggests

that there is significant value in weak ties in the forms of acquaintanceship and informal

networks of relationship had been underestimated in social theory which favored deeper

intimate relationships. Stack (1975), in an ethnographic study of a low-income

community found that families supported each other in small but significant ways that

concretely improved the lives of individuals in that community. Kogan and Chandan
 30

(2004) also found patterns of support that improved the lives of families in a lower-

income community and among children to support the victims of violence and bullying.

From Habermas’s theoretical perspective even Granovetter may have

underestimated the importance of these weak ties. Within the theory of communicative

action the informal networks, weak ties, are the interstices through which all human

relations are formed. It is through the medium of acquaintance and social contacts that

deeper bonds are able to be formed outside of the family.

Forming and maintaining weak ties, however, requires a considerable set of social

competencies. Several skills allow easy maintenance of weak ties including such qualities

as a sense of humor, respect for others’ time, the ability to exchange valuable or

entertaining information, sharing of resources, exchange of labor and ideas, reciprocity,

calibration of demand, compliment giving, negotiation, flexibility, and the authentic

presentation of self. The person with good social skill, regardless of such life conditions

as intelligence, socio-economic level, and academic achievement, is is likely to form and

maintain better marriages and friendships. These skills may be acquired in family and

school contexts and are practiced in children’s play situations.

As Hinde (1976) points out, a relationship is predicated on mutual knowledge and

recognition: without these, it is simply an interaction. Habermas does not privilege one

form of social action over another, stating that instrumental and systemic solutions are

necessary to organize an efficient society in which food and resources are distributed;

however, Habermas posits that instrumental action leads to less deeply-held foundations

of action than does communicative action. To explain this, he proposes the concept of the

―binding force.‖
 31

Binding Force

Bindungsenergien, or Habermas’s idea of the binding force (Habermas, 1987) is

the route through which much knowledge gets turned into action. To Habermas, the force

that comes from instrumental action, often coercion or reward, is weaker than the binding

force that comes from communicative action. Where instrumental action operates through

interest (incentive and disincentive), communicative action operates through a set of

social and cognitive mechanisms that result in compacts, conventions, and commitments

for future action. Habermas suggests that this mechanism is based on an internal

cognitive process. He sees no discontinuity between his theory of social action and the

psychodynamic theories that explain internal processes (Habermas, 1987). He theorizes

that the binding force of communicative action is powerful precisely because it is

developed consensually in innumerable acts of love and solidarity. This idea contradicts

the idea of behaviorism (Skinner, 1969) and developmental biology that proposes that all

action and learning is based on the reward accrued (Bus, 1987).

The essential nature of the cognitive structure underlying commitment and

convention is stability. This stability is responsible for the tenacity of the binding force to

evoke social action. Actions based on commitment or conventions are more likely to

endure over time and be generalized across different situations.

Instrumentalization and Distortion

Lifeworld and system, and communicative and instrumental action are discrete

components in Habermas’s theory, yet he allows that in real life the divisions are not

always so crystalline: people can abuse the trust of others by giving the false appearance

of consensual process. Habermas uses the term ―instrumentalization‖ to refer to the

 32

manipulation of speech/acts to achieve one’s own interests. While acknowledging the

existence of manipulation and lies, he makes the point that dishonesty can succeed only

when there is enough honesty in human interaction to provide a reasonable expectation

that the person is being honest (Habermas, 1984). Likewise, some forms of bullying may

mimic consensual action, and there is some evidence that this deceptive form of bullying

is more destructive than bullying that is being done overtly (Mynard, Joseph, et al.,

2000).

Distortion occurs when people are manipulated into taking action that is not in

their own interest and is not generated through consensual process. An example of this is

the domination of the US food industry which has largely replaced healthy nutritional

practices with processed foods through a campaign of advertisement and lobbying.

Repetition is a source of distortion within the nature of bullying. This causes

distortions in both the bully and the victim themselves and may create a distortion in the

peer culture of the school when bullying becomes an accepted function within the school

culture as a child’s thoughts, emotions and identity become shaped by repeated and

enduring experiences. Bullying within the peer ecology can become a significant source

of entertainment to a large number of children (Cheyne, 1998; Salmivalli, 1999), which

also may cause distortion in the peer culture.

Comparison of Bullying to the Theory of Communicative Action

In this project the various components of the theorized model of effects will be

brought together into a comprehensive model. The constructs and effects sizes and any

salient correlate such as developmental stage or demographic feature will be compared to

the predicted model of effect based on Habermas’s theory of communicative action.

 33

Particular attention will be paid to any data that appear to disconfirm Habermas’s theory

or that point to a mechanism of effect not immediately anticipated in Habermas’s theory.

Habermas’s concepts will be explored in this section including the idea of the

binding force, cumulative continuity, and the dynamics of respect to see if they can shed

light on the process that occurs when some children are affected by bullying and other

children are protected from its ill effects.

Table 3 contains a number of predictions based on the review of the literature and

the ideas of communicative action.

Table 3: Suggested Mechanisms of Effects and their Predicted Results

Mechanism of Effect Predicted Results

ambient stress lack of enjoyment of school
lack of safety at school
sleep disturbance
concentration disturbance
lower academic performance

deprivation of social free play less (more circumscribed) rehearsal of adult roles
(Pepler) less enjoyment of social world
less physical activity through play
mechanisms of endurance (ie. repetition and
reproduction)

identity formation distortion or lack of identity formation

and other sequelae predicted by Honneth

cumulative continuity social skills and aggression that continue due to the
(Kokko) child’s choice of friendships and social environment

role preparation bullying as a preparation for an adult role (perhaps

with the implicit or explicit support of adults or older
children) as suggested by Brunkhorst.

for taking different social roles in different contexts

deep imprinting of instrumental social skills at the
social skills expense of the skills required for consensus building
that the children who view the reasons for bullying as
attributes will be more damaged than children who
view the reasons as variables
 34

While the goal of this project is to test the theory of communicative action, it is

possible that the project will result in a modification of that theory, or the building of a

new theory derived from communicative action.

 35

PART II REVIEW OF THE LITERATURE ON BULLYING

In this section, the research on bullying will be reviewed and analyzed. Bullying

is a complex phenomenon comprised of several dimensions. The subsections below focus

on the history of the research, the characteristics of the bullying phenomenon, the

characteristics of bullies and victims, intervention strategies, and the effects of bullying.

History of the Research

Research into bullying began in the 1970s with a small number of seminal

articles. Since that time researchers have concentrated on determining the actions that

constitute bullying (Arora, 1996; Bjorkqvist, Lagerspetz, & Kaukiainen, 1992; Mynard,

Lawrence, & Joseph, 2000); the characteristics of both children who bully and their

victims (Bosworth, Espelage, & Simon, 1999; Coie, Dodge, Terry, & Wright, 1991;

Olweus, 1980; Perry, Williard, & Perry, 1990; Schwartz, Dodge, & Coie, 1993); the

development of effective intervention and prevention strategies (Hawker & Boulton,

2000; J. D. Smith et al., 2004); and the effects of being bullied (Ambert, 1994; Mynard,

Joseph, et al., 2000; Roland, 2000) As has been stated above, no research has been done

to determine the latent, unseen, processes and mechanisms at play when bullying affects

children (Ladd & Troop Gordon, 2005; Morrison, 2006; Rigby, 2002b; Rivers et al.,

2007).

The Actions that Constitute Bullying

A large part of the bullying research attempts to delineate and define the bullying

phenomenon itself (Mynard, Lawrence, et al., 2000; Rigby, 2000; P. K. Smith et al.,

1999; P. K. Smith & Sharp, 1994). In early research (Olweus, 1978, 1980), the concept of
 36

bullying was not differentiated from peer aggression in general. The variety of gross and

subtle behaviors that comprise bullying and the nature of the bullying interaction

emerged as researchers and clinicians became more familiar with the phenomenon.

Bullying behavior is also evolving and with new communications technology, new types

of bullying are emerging (Li, 2006; Mishna, 2007).

Dan Olweus, a Swedish psychologist, in the first large-scale and systematic

studies of bullying among boys (1973, 1978), limited his operational definition of

bullying to physical aggression and verbal threat without concern for power disparities

(Olweus, 1978). Olweus’s initial definition has been used as the operational definition in

a great many subsequent research projects which has led to the conflation of bullying

with other forms of aggression in many studies.5 Since the publication of Olweus’s early

work, several researchers, including Olweus himself, have expanded the definition of

bullying to include power disparities and a range of harmful actions taken strategically

and indirectly (Arora, 1994; Crick & Grotpeter, 1995; Crick, Grotpeter, & Bigbee, 2002;

Lagerspetz, Bjorkqvist, & Peltonen, 1988; Lowenstein, 1978).

Arora and Thomson (1987) continued elaborating the phenomenon of bullying by

including a wider range of indirect and strategic actions including ganging up on children

and being rude about skin color. Lagerspetz, Bjorkqvist, and Peltonen (1988) furthered

the understanding of the actions that constitute bullying by including relational forms of

aggression, such as social manipulation and status assaults. Subsequent research

continued to elucidate subtle forms of relational aggression (Crick, Casas, & Mosher,

1997; Crick & Grotpeter, 1995; Owens, Slee, & Shute, 2000; Simmons, 2000) and the

5
This conflation of bullying and aggression compounds itself throughout the early bullying literature as the
definition used in early Olweus studies is repeated uncritically in later research.
 37

use of contemporary technology such as cell phones and the Internet (Kowalski &

Limber, 2007; Li, 2006; Mishna, 2007) that bullies can use to coordinate elaborate

campaigns to undermine the reputations and social friendships of their victims.

The Internet and texting capacities of cell phones provide a new forum for

unsupervised communication among and between children which has led to this new

phenomenon called ―cyberbullying‖ (Kowalski & Limber, 2007; Mishna, 2007; Strom &

Strom, 2005; Williams & Guerra, 2007). Cyberbullying refers to bullying done through

any electronic communication medium. Some of the technology available is new and

leads to hitherto unknown forms of bullying. For example, social networking technology

allows peers to rate each other using graphic indicators on such variables as popularity

and attractiveness. A child who is being subjected to an orchestrated campaign against

her social system can now literally watch her own popularity plummet on a bar graph

within minutes,

Bullying and Developmental Stages

The quality and nature of bullying is not stable over the developmental stages of

childhood (P. K. Smith & Levan, 1995). Ample evidence shows that bullying behavior

begins in the early school years, reaches a peak in middle school, and then drops off

sharply in later adolescence (Long & Pellegrini, 2003; Nansel et al., 2001; A.D.

Pellegrini & Bartini, 2000; Salmivalli & Lagerspetz, 1998; P. K. Smith & Levan, 1995;

P. K. Smith, Madsen, & Moody, 1999). Bjorkqvist, Lagespetz, and Kaukiainen (1992),

Smith and Levan (1995), and Long and Pellegrini (2003) found that the bullying

phenomena increases in quantity and sophistication up to middle school among children

ages 14 and 15, and then sharply drops off in later adolescence. Younger bullying
 38

children were found to use more direct physical aggression and threat, gradually

increasing their repertoire of actions as they age to include orchestrated social exclusion

(Stanley & Arora, 1998), assaults on reputation, and strategic manipulation of other

children to carry out bullying activities. The drop off in bullying in late high school may

be due to simple proportion: the fact that older teens have a diminishing cohort of older

and bigger children in their environment.

 39

Bullying as an Interactional Phenomenon

From the beginning of the research into bullying, the phenomenon was described

as an interpersonal dynamic (Heinemann, 1972.; Pikas, 1975; Salmivalli et al., 1996);

however, research shows that bullying most often occurs in the context of a group

dynamic as opposed to a one-to-one interaction (Bukowski & Adams, 2005; Macklem,

2003; D. J. Pepler & Craig, 1995; Salmivalli et al., 1996). Bullying is increasingly

typified as an interactional phenomenon and much of the bullying research being done in

the United States (Blackburn et al., 2004; D. L. Espelage, Holt, & Henkel, 2003)

interprets bullying from the ecological perspective (Bronfenbrenner, 1979), as a

phenomenon that occurs over time as the result of the interplay between and among

intrapsychic and environmental variables. While these authors show how bullying

behavior affects the school community, they do not suggest an explanation of how

environmental factors promote or prevent bullying.

Pepler and Craig (1995), using videotapes of children in play and classroom

contexts, reported that in 85% of observed incidents of bullying at least one uninvolved

peer was present. Salmivalli and her colleagues (1996) undertook the first systematic

study into the ancillary roles children take in bullying situations. In addition to the direct

bully and victim roles, these authors found children taking such roles as the ―assistant,‖ a

child directly abetting a bully through behaviors such as making suggestions to the bully

or holding the victim; the ―reinforcer,‖ a child who comes to watch, laugh, and actively

incite the bully; the ―defender,‖ a child who attempts to help the victim by actively

attempting to stop the bullying or by comforting the victim after the incident; and the

―outsider,‖ the child who intentionally absents him or herself from the bullying. A later
 40

study by Sutton and Smith (1999) found that younger children tended to be more actively

involved in the bullying experience with fewer children taking the outsider role or no role

at all.

Cheyne (1998), in a conceptual analysis of bullying, suggests that children who

bully may victimize others as a form of mimesis, the deliberate imitation by one group of

another group’s behavior--in this case, children imitating adult roles in a process similar

to theatrical productions, sporting events, or pageants. Like theater, bullying comes with

assigned roles and coordinated action. The victim plays the protagonist role, as his or her

emotions are the subject of intense interest to the audience.

Characteristics of Bullies and Victims

A large number of studies attempt to identify the factors and characteristics that

put children at risk for being involved in bullying either as victim or bully (Ahmed &

Braithwaite, 2004; Baldry, 2004; Bowers, P.K., & Binney, 1994; Ladd & Burgess, 2001;

Loeber & Dishion, 1984; Olweus, 1980; Perren & Hornung, 2005; Smetana, 1995).

Research into the correlates and characteristics of bullies and their families has been

uncoordinated and contains some very significant gaps. For example, researchers have

only recently begun to explore the co-occurrence of other forms of victimization in the

lives of bullied children (Finkelhor, Ormrod, & Turner, 2007a, 2007b; Legkauskas &

Jakimaviciute, 2007) and the association between victimization at school and family

violence (Mohr, 2006).

Personal attributes may put some children at risk for being chosen as targets of

bullies (B. J. Kochenderfer-Ladd & Skinner, 2002; Ladd & Troop Gordon, 2003; Perry et

al., 1990; Schwartz et al., 1993). Kochenderfer and Ladd have done significant research
 41

into the interplay between such variables as aggression and withdrawal with peer

rejection and bullying. Their body of work includes a number of longitudinal studies

intended to develop causal models of processes that put children at risk for bullying and

aggression. Aggressiveness, reactive aggressiveness, and social withdrawal tendencies

are associated with being victimized by bullies (Salmivalli & Helteenvuori, 2007).

Gender, Ethnic, and Linguistic Difference in Bully Involvement

Bullying appears to be evenly distributed across national, linguistic, ethnic, and

class boundaries. Both genders bully, although there is a mildly significant difference in

the rates at which girls and boys use relational and direct bullying (Lagerspetz et al.,

1988).

Reacting to Olweus’s original work that excluded girls (Olweus, 1978),

Lagerspetz and Bjorkqvist and their colleagues (Lagerspetz, et al., 1988; Bjorkqvist, et al.

1992) included both genders in their samples and expanded the definition of bullying to

include relational aggression. The authors found only one minor gender difference – a

greater likelihood that boys will engage in direct physical aggression than would girls.

Despite attempts by the authors to clarify the purpose of their research and dispel

misunderstanding (Bjorkqvist et al., 1992), their studies have been widely misinterpreted

to mean that girls and boys bully in categorically different ways (Crothers, Field, &

Kolbert, 2005; Owens, Shute, & Slee, 2000; Owens, Slee, et al., 2000). One particular

study (Storch et al., 2003) finds no significant differences in the nature and extent of

bullying by gender.

The studies that address bullying among different ethnic communities also have

contradictory results. Moran, Smith, Thompson and Whitney (1993), in studying children
 42

ages 9 to 15, found no differences in the ethnic affiliation of children who are involved in

bullying. Nansel et al. (2001) found a slightly higher level of bullying among Latino and

African- American children as compared to the dominant culture. Peskin and colleagues

(2007) found slightly higher levels of bullying among African-American children, while

Estell, Farmer, and Cairns (2007), in a study of rural African-American youth, found no

difference in the rates of bullying when compared to studies of children in dominant

communities. DeVoe (2004) found that white children were slightly more likely to be

bullied than African-American or Latino children. Spriggs, Iannotti, Nansel & Haynie

(2007) surveying 11,033 children ages 11 to16 found lower rates of bullying among

African-Americans as compared to white and Latino students. A systematic review

comparing definitions, methods, and results among these seemingly contradictory studies

may be helpful in determining why these studies came to differing conclusions.

The variation in rates of bullying among ethnic, rural, and linguistic communities

is not as significant as the variation among schools in individual studies (Sharp et al.,

2000)6. Olweus (1990) found that bullying occurs less frequently in schools that do not

tolerate it. The degree of ambient violence--the range of behaviors that constitute

bullying -- fluctuates across context and depends, to some degree, on adults’ attitudes and

responses.

Family Characteristics

Most of the studies into the family characteristics of children involved in bullying

focused on the social context of the bully rather than the victim. Bullying behavior is

associated with certain negative parenting practices, studied as a variety of variables

6
These differences were noted in every multi-schools study reviewed for this paper where data for
individual schools were provided and will be a part of the data mining study proposed below.
 43

including emotional distance, punitive or authoritarian practices, family conflict, and

family violence (Ahmed & Braithwaite, 2004; Baldry, 2004; Bowers et al., 1994; Loeber

& Dishion, 1984; Smetana, 1995). Olweus (1980) in a study of boys, ages 9 to 11, found

a link between mothers’ negativity and fathers’ power assertive methods and boys’

aggression. Again, as with many of the early bullying studies, the definition of bullying

used in this study did not distinguish bullying from other forms of aggression.

Olweus, in a later study using his more contemporary definition of bullying

(1993), found an association between bullying children and a power-assertive parenting

style and a lack of warmth in the parent. In that study, no association was found between

bullying and maternal negativity, however. In a similar vein, Bowers, Smith, and Binney

(1994) found that bullies perceived their parents as distant, lacking in warmth, and either

over-protective or under-involved; they also perceived themselves as lacking in power

within their own families (Bowers et al., 1994). Loeber and Dishion (1984), looking at

aggressive children in general, reported that the children of parents who used both

inconsistent and power-assertive discipline techniques were more likely to be aggressive.

From a positive perspective, studies show that positive outcomes in adolescents,

such as low substance abuse (Baumrind, 1991), social and psychological adjustment, and

social competence, are associated with such positive parenting practices as warmth,

discipline, and adequate levels of supervision (Smetana, 1995; Steinberg, Lamborn,

Darling, Mounts, & Dornbusch, 1994).

Ahmed and Braithwaite (2004) examined a wide variety of family and school

variables that could contribute to a child’s role as a bully or victim. The authors found

that a combination of family and school factors predicted 61% of bullies and 76% of
 44

bully-victims, the children who are both bullies and victims; however, the authors do not

attempt to determine whether some of the school factors such as ―school liking‖ and

―school hassles‖ pre-existed or arose as a direct result of the bullying.

Aside from the power differential, victims are not always chosen because of a

particular attribute or quality, as many in the public believe. Some children are chosen as

targets at random, or because they are thriving socially and academically and are

perceived as smart (Evenson, 1999).

There is less evidence of an association between negative parenting practices and

a child’s vulnerability to victimization (Haynie et al., 2001); however, since some

children are selected as victims at random this dilute the statistical likelihood that family

characteristics will be strongly associated with victimization, since many random victims

will have strong positive family relationships.

Bullying Behavior and Social Relations

Bullies vary in terms of popularity among other children (L.D. Hanish & Guerra,

2004). Aggressive children in general are not usually popular (Lopez & DuBois, 2005;

Mynard, Joseph, et al., 2000; Mynard, Lawrence, et al., 2000; Wentzel & Asher., 1995);

however, some bullies are considered popular among their peers (Estell et al., 2007; P. C.

Rodkin, Farmer, Pearl, & Van Acker, 2000) and this group appears to have the largest

aversive effect (Mynard, Lawrence & Joseph, 2000; Lopez & DuBois, 2005). Estell et al.

(2007) found that both bullies and victims tend to be socially marginalized in schools;

however, a subset of bullies was also identified, in their study, as popular. The authors

argue that deficits of social learning and social skills do not explain the incidence of

bullying by otherwise popular children who have well-developed social skills.

 45

School and Community Characteristics

Olweus’s extensive research project (1992, 1994) on bullying in a school context

indicates that the characteristics of the school environment may determine much of the

variance in the prevalence of bullying. Certain characteristics of a school may reduce

bullying, including rapid and firm adult response to bullying, positive, non-punitive

discipline, and professional supervision of children on the playground, in lunchrooms,

and corridors (Olweus, 1992, 1994; Boulton, 1991). This is supported by studies that

compare the rates of bullying among schools (Boulton, 1991; Sharp et al., 2000). Sharp

and her colleagues (2000) found that the prevalence rate of bullying varied by as much as

17% among five regional schools.

Gumpel and Meadan (2000) show that the prevalence of bullying is sensitive to

and possibly even determined by the response of adults within the school community:

bullying will increase where adults systematically ignore it. Dixon and colleagues

qualitative research about the integration of a class of hearing impaired children into one

British school serendipitously uncovered an extensive pattern among teachers of denying

and minimizing instances of bullying (Dixon, Smith, & Jenks, 2004). Thus, the attitudes

and prejudices of parents and teachers should be the focus of a coordinated research

effort into the correlates of bullying.

Noguera (1995) proposes that the structure of the school system replicates an

oppressive power structure in that it is organized using the same model as prisons with an

emphasis on surveillance and order (Foucault, 1977a). Noguera suggests that this

structure undermines the efforts of well-meaning educators who want to interact

humanely with children. Twemlow, Fonagy, Sacco & Brethour Jr. (2006) note that some
 46

teachers themselves are using bullying tactics in the classroom and that this reinforces

bullying among the children.

Community factors in school bullying have been studied less than family and

individual factors (Bowen, Bowen, & Ware, 2002). In the one study that specifically

looks at community violence, Schwartz and Proctor (2000) found no correlation between

community violence, experienced directly or witnessed, and bullying behavior.

Factors that Put Children at Risk for Victimization

Bullies do ―not distribute their aggression evenly across all available peer targets

but instead selectively direct their attacks toward a minority of peers who serve

consistently in the role of victim‖ (Perry et al., 1990, p. 1310). There is a popularly held

belief that victims of bullying are chosen for this role because of a particular physical or

personal quality, such as obesity or wearing eyeglasses (Olweus, 1993), or being gifted.

One study (Janssen, Craig, Boyce, & Pickett, 2004) did find, contrary to other research,

that overweight teens were somewhat more likely to be chosen as a target. However,

Sunde-Peterson and Ray (2006) found that such ―outstanding‖ features were evenly

distributed between bullies and victims. In other words, a child with glasses, obesity, or

braces is as likely to fill the role of bully as the role of victim, but these children are not

as likely to be in the non-bully-involved group of children.

Whereas physical characteristics do not solely appear to predict victimization,

certain behavioral and social characteristics put children at higher risk of being bullied.

These include passivity (Schwartz et al., 1993) and aggression, particularly reactive

aggression (Schwartz, Proctor, & Chien, 2001), and the tendency toward social

withdrawal (Ladd & Burgess, 2001).

 47

Victims may signal an inability to defend themselves, by simply having a smaller

stature or by a behavioral signal that indicates an unwillingness to defend their own

interests (Egan & Perry, 1998). The victim may also reinforce the bully by easily

relinquishing resources (Olweus, 1978; Perry et al., 1990) or by exhibiting an emotional

reactivity such as crying easily.

Ladd and Kochenderfer-Ladd have conducted a large number of studies into the

quality of children’s social relations (B. J. Kochenderfer-Ladd & Skinner, 2002; B.J.

Kochenderfer & Ladd, 1996; Ladd & Troop Gordon, 2003) and found that social skills

were both a cause and a consequence of bullying and that both were associated with a

downward spiral of self-esteem among victims. In other words, children who started their

school career with poor social skills were found to be at higher risk of bullying, and

bullying experience damage social skills even for those children who started school with

good social skills. Social skills, then, need to be understood as dynamic and evolving

through the developmental stages of childhood.

Low self-esteem also appears to function as both a risk factor for victimization as

well as a consequence of victimization (Egan & Perry, 1998, Troop-Gordon & Ladd,

2005) (Egan & Perry, 1998; Ladd & Troop Gordon, 2005). Egan and Perry (1998) found

that children who were ―protected‖ by a high self-regard were victimized less than

children with low self-regard. The authors also found that the experience of victimization

further lowers the child’s level of self-regard.

Children who are either withdrawn socially or reactively aggressive are at higher

risk for bullying than children with better social skills (L.D. Hanish & Guerra, 2004; B.

J. Kochenderfer-Ladd & Skinner, 2002; B.J. Kochenderfer & Ladd, 1996; Ladd &
 48

Burgess, 2001; Ladd & Troop Gordon, 2003). The victimized child may lack the social

skills that are valued in children’s peer relations, such as a sense of humor or familiarity

with popular culture (Egan & Perry, 1998).

Swiss researchers Perren and Hornung (2005) provide confirmation that victims

tend to have lower levels of peer support than either bullies or non-bully-involved

children. Having and keeping friends protects children from victimization (Davies, 1982;

Ladd, Kochenderfer, & Coleman, 1997). Friendships may function to decrease the

imbalance of power and increase self-esteem (Schmidt & Bagwell, 2007).

Certain minority groups such as gay, lesbian, and bisexual youth in contemporary

America tend to suffer more bullying than other groups, but it is not yet clear whether the

phenomenon is part of bullying or indicative of a separate social problem of heterosexism

or homophobia (Poteat & Espelage, 2007; Warwick, Chase, Aggleton, & Sanders, 2004;

Young & Sweeting, 2004)Further research needs to be done to identify children who

otherwise have few risk factors and rich resources of resilience and are still targeted by

bullies, some of whom may have been chosen at random, .

 49

Entry and Exit from the Victim Role

Figure 1: Role Transitions Over Two Years(adapted from Smith, Talamelli, Cowie,
Naylor, & Chauhun, 2004)
Time 1 Time 2 (two years later)

Non Victims
Non Victims

New Victims

Continuing Victims
Victims
Escaped Victims

Bullying, according to the research, peaks in middle school. Absent in the

literature is the simple mathematic formula that determines this large jump in prevalence

in middle and especially junior high school: the proportion of children on the lowest end

of the age range is one-third in middle school and fully one-half in junior high: this

compares with one-sixth of children in a K-5 school. Middle and junior high school

children find themselves suddenly on the smaller end of the school after being on the

larger end for two years or more.

As was discussed above, the acceleration of growth at middle and junior high

school age explains the statistical jump in bullying prevalence in middle school.

Children’s bodies may increase as much as 35% in bone and muscle mass over a two-

year period (Ruff, 2003). In addition to the likelihood of becoming a victim increasing

with the proportion, children may experience extreme variation in physical size. In some

cases, children who were formerly the smallest in their cohort will grow in size and

strength to equal or surpass their peers: conversely, a child who was a bully in elementary

school may suddenly find himself in middle school dwarfed in size by his former victim.
 50

As with the peak in middle school, the fall off of victimization in high school may also be

the result of simple mathematics: as children reach the end of their high school years, the

pool of potential bullies vanishes as the older and larger cohort graduates or leaves high

school.

Co-occurring Victimization (Poly-victimization)

Little research has sought to determine whether and to what degree children

involved in bullying have experienced other forms of abuse or traumas (Finkelhor,

2007b; Legkauskas & Jakimaviciute, 2007). Legkauskas and Jakimaviciut (2007) found

that for both girls and boys the experience of abuse at home was highly correlated with

the perpetration of bullying behavior; emotional and physical abuse at home correlated

with both victimization and bullying at school. In several studies, Finkelhor and his

colleagues (Finkelhor et al., 2007a, 2007b; Finkelhor, Ormrod, Turner, & Hamby, 2005b)

surveyed a national sample of over 1,000 children for the occurrence of different types of

victimization: unfortunately, they conflated peer victimization and sibling victimization

into the same category despite the fact that they are different phenomena, albeit with

similar characteristics. The authors found that bullied children often experience several

forms of abuse including sexual, physical, and emotional abuse and peer victimization as

well as being witnesses to violence. While victimization was correlated with other forms

of abuse, 50 per cent of the children who were bullied did not have other forms of abuse

or aversive family characteristics. This finding lends credence to the idea that there is a

cohort of children who are chosen at random as targets of bullying. Holt, Finkelhor, and

Kaufman-Kantor (2007) confirm the existence of a group of youth victims who have not

experienced related traumas.

 51

Effects of Bullying

A great deal of research has been done into the association between bullying and

its negative effects7. It has been established that bullying and other problems within the

peer system are antecedents of global problems with adjustment and psychopathology in

children (Bukowski & Adams, 2005). Children who have been bullied or who themselves

engage in bullying for an extensive period of time are at risk for problems during

childhood such as lower academic performance and school attendance (Clarke &

Kiselica, 1997; Cunningham, 2007; Holt et al., 2007; Sharp et al., 2000; Shellard &

Turner, 2004; Wentzel & Asher., 1995); lower self-esteem (Mynard, Joseph, et al., 2000;

O’Moore & Kirkham, 2001; Rigby & Slee, 2001; Salmon, James, & Smith, 1998);

depression (Haynie et al., 2001; Erling Roland, 2002; Seals & Young, 2003); generally

poorer mental health (Petersen & Rigby, 1999; Rigby, 1999, 2000); anxiety and poor

physical health (Baldry, 2004; Craig, 1998; Slee, 1994); suicidal ideation and gestures

(Brunstein Klomek et al., 2008; Carney, 2000; Kaltiala-Heino et al., 1999); and

withdrawn behaviors (Baldry, 2004).

Bullying contributes to problems with both academic performance and school

climate, often depressing academic scores for an entire school (Ballard, Argus, &

Remley, 1999; Clarke & Kiselica, 1997; S.W. Twemlow et al., 2001). Bullying has a

negative impact on learning for children with all active roles in bullying, including

bystanders for whom the bullying of peers may be at best a distraction from learning and

at worst a source of ongoing anxiety (Lumsden, 2002). Victims were more often absent

7
The terms ―effects‖ and ―sequelae‖ are used interchangeably. The term ―sequelae‖ will be used in this
paper when the term ―effect‖ is used to denote effects other than bullying sequelae.
 52

from school than non-victims and expressed less enjoyment in life (Ballard et al., 1999;

P. K. Smith & Sharp, 1994; S. W. Twemlow, 2004).

Another effect of bully involvement may be disturbances in the process of identity

formation (A. van Hoof, Raaijmakers, van Beel, Hale, & Aleva, 2008); however, to date,

no research has been done into the formation of identity in children experiencing

bullying. Nevertheless, the totality of the effects literature shows a close corollary to

identity formation as the items in identity formation questionnaires involve most of the

same measures of social life, cognition, emotion, and behavior as are used in the effects

literature.

Effects on Physical and Perceived Physical Well-Being

The physical effects of bullying have been firmly and clearly established (Baldry,

2004; Faust & Forehand, 1994; Hodges & Perry, 1999; Legkauskas & Jakimaviciute,

2007; Paquette & Underwood, 1999). The immediate physical effects include stomach

aches, headaches, sleeping problems, and other physical complaints. Srabstein (2006) in a

national study found that victims as well as bullies report a wide set of physical problems,

including headaches, stomach aches, sleep disturbance, and backaches. These symptoms

are due, in part, to a somaticization of anxiety or due to hypochondriacal attempts to

avoid school.

Victimization experiences, for a large percentage of affected children, have

immediate physical consequences that have a negative impact on the quality of the

child’s life. Empirical evidence supports an association between victimization and a set

of immediate physical symptoms, especially in studies that measure contemporaneous

 53

bullying. This is a clear indication that children are suffering as a result of bullying, and

that the suffering is felt in the body.

Effects on Cognitive and Affective Well-Being

Even in the absence of other risk factors in the life of a child, a single or short-

term experience of bullying and peer rejection can negatively affect the child’s self-

esteem and emotional well-being (Arora, 1996; P. K. Smith & Levan, 1995). For

example, the victim of orchestrated relational aggression may experience the sudden loss

of a peer network after a lifetime of stable and healthy peer relations, precipitating a

sudden, severe depression (Simmons, 2000).

Sourander, Helstela, and Piha (2000) found that victimization and bullying were

both associated with lower levels of social competencies and high externalizing behavior,

such as aggression and hyperactivity in children who bully; and, in victims, high

internalizing, including depression and anxiety problems as scored on the Achenbach

Youth Behavior Checklist (Achenbach, 1991; Achenbach & Rescoria, 2001). The

children themselves reported feelings of ineffectiveness and interpersonal problems.

Victims were found to exhibit both high levels of psychotic symptoms and internalizing

behavior (Forero, McLellan, Rissel, & Bauman, 1999).

Children involved in bullying are at higher risk for suicide than their non-bully-

involved peers (Cleary, 2000; Finkelhor et al., 2007a, 2007b; Kaltiala-Heino et al., 1999;

Kim & Leventhal, 2008; Omigbodun et al., 2008; Rigby, 1998; E. Roland, 2002). In a

systematic review of the 39 studies on bullying and suicide, Kim and colleagues (2008)

found that, despite the methodological problems of comparing studies that used differing
 54

measures of victimization, both bullies and victims are at greater risk for suicidal

thoughts and attempts, with children who are bully-victims at the highest risk.

Kaltiala-Heino et al. (1999) also found that children at highest risk for suicidal

symptoms were bully-victims, with a 12:1 odds ratio. An odds ratio is used to evaluate

risk in epidemiological studies; it is the number of events divided by the number of non-

events (Cook, 2002). Bullies were seen as at risk with an 8.7:1 odds ratio; victims were

5.7:1 times more at risk8. A fuller study might gather data on other risk factors in the

children’s environments, such as abuse and neglect, that may also explain these elevated

levels of suicide and depression (Finkelhor et al., 2007b; Legkauskas & Jakimaviciute,

2007).

In another study of depression and suicide, Brunstein-Klomek and colleagues

(2008) studied 2,342 high school students and found that both victims and bullies are at

higher risk for depression and suicidal ideation as well as actual suicide attempts.

Bullies and victims were found to be more frequently referred for psychiatric

consultation than their non-bully-involved peers (Kumpulainen et al., 1998;

Kumpulainen, Rasanen, & Puura, 2001; Rigby, 1998, 2000), with bully-victims almost 7

times more likely to be referred for psychiatric consultation (Kumpulainen et al. 1998).

Involvement in bullying in early childhood predicted psychiatric symptoms in

adolescence (Kumpulainen & Rasanen, 2000) for both bullies and victims. Roland (2002)

conducted an extensive examination of the effects of bully involvement on mental and

physical health and social support. His study investigated bullying among 2,088

Norwegian children in the 8th grade. Roland selected a sample of children who identified

8
While the lower rates for victims appear counterintuitive, they may result fromthe significant percentage
of victims who are chosen at random and may have significant resources for protection.
 55

themselves as having been bullied within the past week: 5% met the criteria; this narrow

lens makes it difficult to distinguish transitory victims with a brief experience of

victimization from those whose bullying had been ongoing. However, within that small

group, the author found a very strong association between peer victimization ( as Roland

tends to term the phenomenon of bullying) and both anxiety and depression. Bullies and

their victims scored higher on depression and measures of suicidal ideation and gestures

compared to a group that was not involved in bullying. The sample of those

contemporaneously bullied shows that anxiety is high during episodes of bullying, as

those children being bullied may experience symptoms of stress, such as trouble sleeping

and eating, that are directly related to bullying as an environmental condition.

The Effect of Bullying on Social and Geographical Space

Few bullying researchers have looked at bullying from the perspective of its

spatial effects. An exception is the mapping of high school spaces by social work

researchers in Michigan (Astor & Meyer, 1999; Astor, Meyer, & Behre, 1999; Astor,

Meyer, & Pitner, 2001). These authors identified ―unowned spaces‖ where children are at

higher risk for violent encounters; the authors, however, did not distinguish bullying from

other violent acts. A small number of geographers have started to look at bullying as

creating ―tyrannical space,‖ referring to the changes that bullying and other forms of

victimization make in the environment (Andrews & Chen, 2006; Percy-Smith &

Matthews, 2001). Several studies, without explicitly defining the issue as spatial,measure

children’s perceptions of their physical space, by determining the children’s sense of

safety or anxiety about being in school (Berthold & Hoover, 2000; Glew, Fan, Katon,
 56

Rivara, & Kernic, 2005; Juvonen et al., 2000). This element of bullying will be more

fully explored in the systematic review below.

Social Space

Where physical space is bounded by clear, physical limits, social space is more

diffusely bounded and involves social skills and social bonds without physical

manifestations that can be detected directly, though researchers can infer lack of

connection by measuring friendship networks through interviews and observation. The

social space of children victimized by bullies appears to be profoundly affected by their

bullying experiences and leads to, or is correlated with, problems of social connection

and social competence. Several separate studies (Mouttapa, Valente, Gallaher, Rohrbach,

& Unger, 2004; A. D. Pellegrini, Bartini, & Brooks, 1999) as well as a canon of studies

conducted by Gary Ladd and Becky Kochenderfer-Ladd (Buhs & Ladd, 2001; Buhs,

Ladd, & Herald, 2006; B.J. Kochenderfer-Ladd & Wardrop, 2001; Ladd & Burgess,

2001; Ladd et al., 1997; Ladd & Troop Gordon, 2003, 2005) shed a great deal of light on

the social lives of victimized children.

Social skills, is another diffuse concept that refers to a large set of qualities,

actions, and perceptions that defy easy definition (Segrin, 2000). Not only is the concept

diffuse, it is highly context-sensitive and changes over time. A person may need to use a

different set of social skills depending on her role and that role’s orientation to various

social contexts (Brunkhost, 1986), so a child may need to use certain skills with adults

and another, far more dynamic set of skills with peers. Peer contexts vary and fragment

as children create communities of interest that become richer as the children progress
 57

through adolescence: each context may have a jargon, a dress code, and a set of behaviors

that are unique at least in some feature.

Victimization and social competence are highly entwined concepts as the one

affects the other. Problems with social competence and social connectedness are both a

cause and a consequence of victimization (B.J. Kochenderfer & Ladd, 1996). Pellegrini

and colleagues (1996) found that victimized children were significantly more isolated:

they had the fewest reciprocated friendships (the independent agreement between two

children that they recognize each other as friends), and were generally least liked by their

peers. An isolated child is at higher risk for being chosen as a target, and the

victimization limits or eliminates the child’s opportunity to learn social skills. Since

social skills within childhood are constantly evolving, even a child who enters school

with a good set of social skills may quickly fall behind her peers if she finds herself

isolated within the peer group.

The research of Ellis and Zarbatany (2007) finds a similar pattern of withdrawal,

and the authors also conclude that there is a social cost to befriending or maintaining a

friendship with a targeted child: the friend of the target may be putting herself at risk of

becoming a target as result of the friendship and may experience a lowering of social

status.

Under some circumstances, withdrawal can be used as a strategy to avoid further

bullying (L.D. Hanish & Guerra., 2000). Kogan and Chandan (unpublished data)

observed victims using this strategy as well as non-victims who quietly and routinely

absented themselves from situations in which they were at risk for being bullied.
 58

Bullying as an Antecedent of Future Problems

Several studies examine the role bullying experiences play in the formation of

later problems, particularly later aggression (Coie, Lochman, Terry, & Hyman, 1992;

Newman, Fox, Harding, Mehta, & Roth, 2004; Vossekuil et al., 2002). The cohort of

children taking part in the longitudinal studies begun in the 1980s in Scandinavia is now

entering late adolescence, and the first study has recently been published (Sourander et

al., 2007); it links early bully involvement with criminality in late adolescence and early

adulthood. Earlier US studies found that early peer rejection contributes to adolescent

aggression (Coie, et al., 1992) , lower psychological well-being and academic

underachievement (Chen, Chang, Liu, & He, 2008; Glew et al., 2005; Ialongo, Vaden-

Kiernan, & Kellam, 1998; Schwartz, Gorman, Duong, & Nakamoto, 2008; Schwartz et

al., 2005; Wentzel & Asher., 1995). Both peer rejection and aggressiveness are predictors

of several disorders that appear during adolescence (Coie et al., 1992). From an inverse

perspective, positive peer relationship is a predictor of future social adjustment (Crick,

1996; Hodges & Perry, 1999).

Persons who were bullied and/or peer rejected as children may experience

negative sequelae that endure into adulthood (Crick, 1996; Kokko & Pulkkinen, 2000).

While there are, as yet, no longitudinal studies that have followed victimized children

fully into adulthood, Ambert (1994) coincidentally found evidence showing that bullying

experiences in childhood are still deeply felt in early adulthood. The author gathered

1,350 autobiographical sketches over 15 years from her undergraduate students at a

public university. Students were asked to describe what made them happy and unhappy

during their childhoods, pre- and post-puberty. The students reported that negative peer
 59

interactions, including bullying, racist, and sexist actions, had more detrimental impact

lasting into the present day compared to abuse at the hands of a parent. While many

students disclosed other forms of abuse, including physical and sexual abuse, the

consensus was that the bullying experiences were more emotionally damaging.

A few studies show some evidence that bullying and other negative peer

experiences have sequelae that last into adulthood. In an oft-cited longitudinal study,

Caspi and colleagues (1998) delineated several variables associated with chronic adult

unemployment and underemployment in the United Kingdom. The authors found general

negative peer experiences in childhood to be associated with labor market under-

performance in adulthood. Miller and Vaillancourt (2007), in a retrospective study, found

that girls who were indirectly bullied tended to exhibit perfectionist tendencies as young

adults, and Sourander, et al. (2007) found that early bullying behavior was associated

with criminal behavior in youths 16 to 19 years old, while victimization was not

associated with later criminal activity.

Kokko and Pulkkinen (2000) advise caution, however, in drawing conclusions

about associations and correlations between early bullying and peer rejection and later

sequelae; they suggest that the mechanisms of ―interactional cumulative continuity,‖ the

selection of relationships and environments that support and propagate anti-social (or

pro-social) behaviors, would be as likely explanations of the presence of adult adjustment

problems as early bullying. Cumulative continuity operates when children with an

aggressive and anti-social peer group accumulate negative experiences, including

constricting and punitive disciplinary actions; these children tend to maintain the same

friends or similar aggressive friendships over their entire life span. Kokko and Pulkkinen
 60

say that the process by which these children are affected is likely to be continuous rather

than discrete.

Sourander and colleagues (2007), in a longitudinal study of Finnish males, found

no association between the experience of being bullied and youth crime and some

evidence that the bullies were at higher risk for adult criminality. Garbarino (2002) and

Gilgun (2001) both claim that school bullying is a contributing factor to the mass murders

of students by other students. Both authors base their assertions on qualitative research

done with the perpetrators of highly publicized school shootings as well as the survivors

of the shootings who describe the murderers as having been extensively bullied for years

prior to the shootings.

One extensive data-mining study supports Gilgun and Garbarino’s assertions in

that bullying was a significant factor in the lives of many children who perpetrated school

shootings. The study, commissioned by the US Secret Service (Vossekuil et al., 2002),

analyzed hundreds of transcripts of interviews undertaken by police officers,

psychologists, teachers, and social workers with 41 children who perpetrated school

shootings between 1974 and 2000. The content was coded and analyzed in an attempt to

uncover the intrapsychic or familial variables that could help identify potential shooters

and thereby prevent future school shootings. The most salient variable discovered in that

study is that 71% of school shooters were victimized by school bullies. The authors

conclude that educators need to make each school a place where students feel safe and

valued, and that neglecting problematic social interaction among students could yield dire

consequences.
 61

Noted educational researcher Cheryl Graves (2004), in a conversation about

bullying in the media, stated that parental neglect may be a more significant factor than

bullying in school shootings, pointing out that the parents of some of the child murderers

claimed they did not know that their child had a collection of assault weapons though the

police discovered the weapons in plain sight in the children’s bedrooms.

Bullying plays in long-term sequelae, bullying appears to be a chronic, although

not always acute, environmental stressor. Finkelhor (1995) states that bullying is a source

of ongoing stress for the victim as well as an obstacle to adequate development of social

skills and cognition for both the bully and the victim.

Evidence of Policy Support for Bullying

Kamenetsky (1996) indirectly contributes s to the bullying literature with her

review and analysis of children’s literature during the years that the National Socialist

(Nazi) party dominated Germany and much of occupied Europe. German educators of the

Nazi era directly and explicitly encouraged bullying, while not using that term itself.

Kamenetsky points out that this was most evident in the curriculum they used to train

children enrolled in Hitler Youth groups9. The Nazis ―successfully‖ used the Hitler

Youth groups to develop a corps of adults to take jobs preparing countries for Nazi

occupation or in the death camps (Office of United States Chief of Counsel, 1946); it can

be assumed that the children’s literature was developed and disseminated expressly to aid

in that goal.

In one particularly frightening example of Nazi children’s literature disseminated

through the Hitler Youth, in a children’s book whose title was translated as Never Trust a
9
Kamenetzky (1996) demonstrates that all children’s literature of that period was developed according to
specific guidelines from the Nazi Department of Education.
 62

Jew on the Green Heath (Kamenetsky, 1996), cartoon depictions show children how to

bully and exclude Jewish children on the playground. These illustrations bear a striking

resemblance to cartoon depictions of bullying Smith and colleagues (2000) showed their

subjects in order to study bullying across linguistic and national barriers. Kogan and

Chandan (2004) found that some children, parents, and teachers support childhood

bullying, in part for its perceived positive role in preparing children for adult roles of

enforcement and domination. While an analogy to the Nazis may seem distasteful, their

idea that bullying prepares children for adult roles of coercion is identical to the ideas of

the Nazi Department of Education.

Anti-Bullying Interventions

With federal governments and states implementing policy and laws against

bullying (Limber & Small, 2003), intervention programs are proliferating (Blakely, 2004;

McGrath, 2007; Scaglione & Scaglione, 2006); however, there is scant evidence for the

effectiveness of these programs (Rigby, 2002b; J. D. Smith et al., 2004; P. K. Smith, D. J.

Pepler, et al., 2004; Stevens et al., 2000). Some evidence suggests that certain

interventions may actually increase bullying or simply make it less visible to adults (D.J.

Pepler et al., 1994; Vreeman & Carroll, 2007).

Despite Olweus’s assertion that individual school characteristics may determine

the prevalence of bullying, only a small number of bullying intervention models have

been in existence long enough to have been evaluated, and none of the bullying programs

evaluated would meet standards of evidence-based practice(Rigby, 2002a; J. D. Smith et

al., 2004; J. D. Smith, Stewart, & Cousins, 2003). Rigby (Rigby, 2002a) in his meta-

analytic study of anti-bullying intervention programs shows that none of the studies
 63

included in his analysis sought to determine the relative effects of their multiple

components or their dosage; evaluation methodology varied in quality and criteria; and

many evaluation studies were applied in only one geographical area often with a

homogeneous population.

While there are several models of anti-bullying interventions, most utilize three

program components: whole school policies, social skill-building curricula, and social

and behavioral group work (Vreeman & Carroll, 2007).

This section will review a few of the major intervention models that have been

implemented over the past three decades and recent models that offer a significant

variation or innovative program component. All programs reviewed have one or all of the

abovementioned three components, and some recent programs have added new

components, for example, legal accountability, in states where schools can be held liable

if bullying is not dealt with effectively (McGrath, 2005).

Olweus (1991, 1994), surveying the results of several intervention programs using

his Olweus Anti-Bullying Program (now called the Olweus Bullying Prevention

Program) over the previous decade, reported a reduction of bullying and an increase in

teachers’ willingness to intervene. Since most of the data were gathered before the more

specific contemporary definition of bullying was articulated, it can be assumed that

general aggression is conflated with bullying in the results. Several future evaluations of

the same program model were unable to replicate Olweus’s results (D.J. Pepler et al.,

1994; Vreeman & Carroll, 2007); however, all these used a contemporary definition of

bullying as opposed to the original study’s wider definition. The Olweus model was

implemented and evaluated in four Canadian schools (Pepler, Zeigler, & Charach, 1994)
 64

over a period of a year and a half. The results were mixed at best with a small decrease in

the number of children reporting abuse within the week previous to the post-test, but with

a small increase in the proportion of children who reported being frequently bullied. No

control group was used with which to compare rates of bullying.

Vreeman and Carroll (2007) in a meta-analysis of bullying intervention programs,

show that, while the programs may reduce the number of bullying incidents, the number

of bullies and victims may still remain stable or actually increase. These seemingly

contradictory results may be explained by a number of concerning reasons: perhaps

children who were formerly only overtly aggressive had started to bully in a de facto

changing of the guard, or children may have changed the way they bullied in order to

remain below the attention of ever more vigilant adults.

One US study purports to show that the Olweus model significantly reduces

bullying. Black and Jackson (2007) report observing a very large reduction of bullying in

one school after four years of intervention. The authors used an interrupted time-series

observation methodology implemented by external evaluators who reported a 45%

decrease in observed bullying. While the authors claimed to be implementing the Olweus

model, they added significant program components, including the separation of ages and

genders and the inclusion of highly adult-supervised and organized social activities at

lunch and recess. These new program components would logically account for the large

reduction of observed bullying simply because of the limitations they place on the

autonomy of all children including bullies. The Black and Jackson model may in fact be

iatrogenic, with the cure being more costly than the problem, due to the children’s lack of

opportunity to practice social skills in an autonomous environment.

 65

Salmivalli (2001) showed modest reduction in bullying by adapting the Olweus

model in a different way, by including program components that are intended to limit the

ancillary roles of bullying. In a posttest after six months of intervention, students reported

a 15% decrease in bullying. Frey, et al. (2005) studied a similar intervention program that

incorporated Salmivalli’s approach which also showed some success in reducing the

incidence of observed bullying. These studies suggest that intervention programs that aim

to reduce children’s ancillary roles in bullying are promising and worthy of future study.

Ducharme, Folino, and DeRosie (2008) suggested that the minimal effectiveness

of social- skill building intervention programs (J. D. Smith et al., 2004; P. K. Smith, D. J.

Pepler, et al., 2004) result from too weak a dosage, and, in response, the authors

developed an intensive social-skills program which they implemented with two groups of

four children with severe bullying involvement. The authors theorized, after intensive

time-series observation of children at play, that the ability to give in to the will of others

was the core component skill for positive peer interaction which they term

―acquiescence.‖ The researchers hypothesized that teaching acquiescence would lead to a

broader range of positive social skills and positive interpersonal experiences. While the

authors show some moderate improvement in bullying, the intensity of the intervention

makes it costly both in terms of labor and time, with 60 academic hours taken up by the

program each year.

Legal considerations are also entering the field of bullying prevention and

reduction and should be considered in developing future intervention programs.

McGrath’s (2007) model uses the whole school policy and social-skills building

components and adds in procedures to protect the schools from liability in jurisdictions
 66

where there are State laws against bullying. McGrath, a lawyer, adds program

components that require the schools to maintain procedures and records that will protect

the children against bullying and protect the schools in any future legal procedures.

McGrath offers no empirical evidence of her program’s effectiveness –a program that is

widely marketed --nor have any outcome studies been published to date.

The few program models that show positive results demand a tremendous

concerted effort on the part of school staff, parents, and students. This puts anti-bullying

and social-skills promotion in direct competition with academic curriculum. In the US, at

the time this paper is being written, schools are under pressure to raise standardized test

scores as part of the No Child Left Behind initiative and schools may not be amenable to

introducing programs that would take time away from academic teaching unless the

program positively affects test scores. While there is a correlation between bullying and

low academic performance, none of the intervention programs’ effectiveness studies cited

above provide data showing significant improvement in either academic performance or

test scores.

Despite the lack of significant intervention results for individual intervention

strategies, the effects of victimization may be mitigated in certain countries with

powerful policies and well-funded national intervention efforts. Due (2005) directly

studied the effects of bullying cross-nationally and found that US and Israeli children

experienced more severe levels of symptoms than countries with long standing anti-

bullying efforts. The same comparison can be extrapolated from the results of Wolke’s
 67

(2000)10 UK study that shows markedly less severity of anxious and depressive

symptoms than the results for those variables in this review of US studies.

It is likely that victimization is experienced less severely in environments where

children are supported and, if necessary, treated as they are in the UK, Scandinavia and

Commonwealth nations in which bullying has been seen as a significant social problem

worthy of dedicated and relatively well-funded policies.

Conclusion

The effects of bullying are diffuse. A cursory look at the results of the studies

reviewed below prior to the more systematic analysis of the results show that bullying is

at least partially responsible for anxiety, symptoms of post-traumatic stress, depression,

suicide, school problems, aggression, future low employment and future

violence.Measures of these variables, as well as the prevalence of bullying itself, vary so

widely that making comparisons between studies is not always plausible. Bullying is

measured variously by direct observation, peer report, self report, teacher report;

operational definitions of bullying may be limited to one aspect of aggression or inclusive

of a range of bullying actions; and the temporal criteria for determining the degree to

which bullying occurs differ from contemporaneous bullying to retrospective periods of

up to three years. Measures used to investigate variables such as depression and anxiety

range from standard instruments such as the Achenbach Child Behavior Checklist (1991)

to author-created items on questionnaires. A perusal of the studies into the consequences

of bullying would lead a reader to conclude that being bullied is associated with a

bewildering complex of negative sequelae.

10
Unlike Due, Wolke did not perform a cross-national study but her results indicate a less statistically
significant level of emotional distress among the children studied than any comparable US study.
 68

While bullying is an emerging social problem in the US and an established social

problem in Commonwealth and European countries, there is no evidence that bullying

can either be reduced or prevented on a school-wide or district-wide basis (Rigby, 2002a;

J. D. Smith et al., 2004). Dubin (1978) points out that for interventions in the applied

sciences to be effective a theoretical explanation for the underlying phenomenon is

required. A theoretical model that can explain both the etiology of bullying and the

mechanisms through which bullying affects children is also likely to result in more

effective interventions .

There has been very little research seeking to build or test a theory within the field

of bullying research. None of the theoretical models currently reflected in the bullying

literature explicitly consider the dynamics and effects of power abuse; yet, this single

variable distinguishes bullying from aggression. The next section will survey the

literature on power with the aim of finding parallels to the bullying phenomenon by

exploring the links between the literature on bullying and the literature on power.

Review of the Literature on Power

The power qualities of bullying are the central concern of this writer’s project, yet

a thorough archival search of the bullying literature has yielded no study specifically and

directly linking the literature on power with the literature on bullying. In this section, the

power literature will be briefly reviewed and then related to the topic of bullying.

Theory and research of power involve a range social processes and are important

fields of study in several social science disciplines, such as sociology (Dahrendorf, 1958;

Parsons, 1969), organizational theory (Kearins, 1998), political science (Gaventa, 1980;

Giddens, 1995; Luhman, 1990), social and community psychology (Rappaport, 1987),
 69

marriage and family studies (Safilios-Rothschild, 1970; Straus & Gelles, 1979, 1990),

and social work (Guittierrez, 1990; Guittierrez & Ortega, 1991; Kondrat, 1995; Levy,

1990). Contemporary power theorists and researchers tend to cross disciplines such as

philosophy and psychology (Foucault, 1973, 1977a, 1977b, 1978, 1997); sociology,

philosophy, and linguistics (Habermas, 1971, 1979, 1984, 1987, 1996, 2000, 2001a); and

literary criticism (Borradori, 2003; Derrida, 1974, 1988).

Definitions of Power

The first problem in the study of power is that there has yet to be a consensus on

the definition for the concept of power itself (Dahl, 1957; Keifer, 1984; Lukes, 1974;

Parsons, 1969). Just as the concept of bullying is inclusive of a great many qualities and

actions undertaken by individuals or collectives for a myriad of motivations, the concept

of ―social power‖ is comprised of a myriad of actions and processes. Also as with

bullying, the physical aspects of power can be more easily quantified by examining the

control and distribution of resources, personal attributes, and access to the political

process, than can the socially constructed aspects of power that lie within our subjective

experience of the world (Bachrach & Botwinick., 1992; Foucault, 1973, 1978, 1997;

Kearins, 1998; Luhman, 1990; Pfeffer, 1997).

Parsons (1969) says that there are three contexts that explain the difficulty in

developing a consensus for the definition of power: the concept itself is diffuse; the

relation between the coercive and consensual aspects of power has not been clearly

established11; and the notion that power is a zero-sum problem poses difficulty. In

11
Parsons wrote this before the ideas of Foucault, Derrida, and Habermas, among others, enriched the
dialogue on power.
 70

reality, certain aspects of power can be seen as having finite properties, called zero-sum,

while other aspects may not be bounded.

Parsons (1969) subscribes to the zero-sum idea of power, which is consistent with

his structural functionalist idea that uses as its dominant metaphor for social relations the

flow of goods and services in the marketplace. The zero-sum idea proposes that, like

money and most resources, there is a finite supply of power, and that increasing the

power of one group or person will, somewhere in the system, decrease the power of

another. In fact, Parsons equates money and power as similar symbolic media in that

paper or coin is invested with a meaning based on consensus.

Weber (1947) defines power as "the probability that one actor in a social

relationship will be in a position to carry out his (sic) will despite resistance, regardless of

the basis on which this probability rests" (Weber, 1947, p. 152). Weber provides a

schema through which power is exercised through authority based on different qualities

broadly described as tradition, rationality, and charisma. Parsons (1969) presents a

framework for how complex social systems function by adapting to new situations, by

attaining goals, by integrating or harmonizing with other systems, and by including a set

of latent patterns that reproduce culture and institutions.

Though Foucault’s encyclopedic research project concerns the evolution of the

technology of power, he rarely uses the term ―power‖ itself. When he does, he defines

power neutrally as a productive network that includes qualities, processes, or activities

that promote or resist change in human systems (Foucault, 1977b). Habermas also rarely

uses the term ―power‖ but, in elucidating the instrumental and communicative sources of

social action, he covers the gamut of power relations (as will be fully explained below).
 71

Luhman (1990), while acknowledging the objective manifestations of power, points out

that power is often situated in social relationships in which at least some part of the action

could have been different, no matter what the state of the power relations; with this

symbolic or subjective aspect of power, power would not be considered a finite quantity.

Models and Theories of Power

Bachrach and Baratz (1962) developed a model with two dimensions of power

that is still widely used in the political theory literature today and will be helpful in

locating the various forms of bullying catalogued by bullying researchers. The first

dimension of power is the most easily quantified: it includes observable behavior, such as

overt use of physical strength and resources. The second dimension of power includes

techniques that are employed to deter access to political power by keeping issues off the

political agenda; these less obvious uses of power include ―non-decision-making‖ and

―mobilization of bias.‖

A third dimension of power exists on what would now be termed a socially

constructed (Burger & Luckman, 1966) plane, as in Foucault’s analysis of how power

influences behavior through knowledge, including our conceptions of health (Foucault,

1973), mental health (Foucault, 1976), justice (Foucault, 1977a), and sexuality (Foucault,

1978). For example, in Foucault’s analysis of the mechanics of state power, he traces the

evolution of the techniques of power: from coercive forms exercised directly on the body,

through public displays of torture and execution, to more remote and subjective

techniques of state control.

Foucault (1977) examines the evolution of the mechanics of power over time across

the dimensions described by Bachrach and Barantz (1962). Initially, power was
 72

exercised directly by the monarch and his proxies against the individual body, the first

power dimension. Before there were codified laws, local lords meted out justice usually

in the form of capital punishment: torture, bodily humiliation, or execution. With the

codification of laws, the sovereign began to be distanced from the body, and power

mechanics moved into the second, interpersonal, dimension.

Foucault (1977a) describes the architectural innovation of the ―panoptikon,‖—a

plan in which prison barracks are built around a central atrium from which the inmates

can easily be watched --as one of the most significant inventions in Western history. The

dossier, a system of retaining files on individual citizens in a central location is another

form of panoptikon in that it allows authority easy surveillance of large numbers of

people. Subsequent civil engineering used the panoptikon principle in city planning, as in

Napoleon III’s redesign of Paris in which miles of medieval alleys and crooked streets

were razed and replaced with wide straight avenues amenable to surveillance and easy

access of troops (Saalman, 1971). Foucault sees the panoptikon and the dossier as a

revolution in the technology of power, allowing control to be exercised remotely.

Foucault furthers his analysis of how power is socially constructed when he

describes that power is exercised through knowledge (Foucault, 1977b) and can extend to

our own bodies in the form of sexual constraints and stereotypes (Foucault, 1978) and the

forms and qualities of mental processes (Foucault, 1976).

Habermas’s early work closely mirrors Foucault’s in two meaningful ways but

differs in an even more significant way: Habermas, like Foucault, investigated how

knowledge is formed by human interests, and, like Foucault, he surveyed an enormous

body of literature. Foucault, however, exhibits a flare for dramatic narrative, drawing the
 73

reader into his highly abstract philosophy with stories taken from history; at times,

Foucault’s enthusiasm for narrative leads him, at times, towards doubtful associations

and poor scholarship (Kaiser, 2000). By contrast, Habermas’s scholarship is meticulous,

though his work may lack narrative verve. Habermas abandoned his interest-based

theory, knowledge constitutive interests, in favor of communicative action which he uses

to pursue his project of developing socially-just constitutional states.

Derrida’s contribution to the study of power derives from his textual project.

Derrida changed the frame of literary analysis by making all characters in narratives

equally important as protagonists and by determining the assumptions embedded in the

text that reveal the social forces acting on the writer to constrain the narrative, the actions

of social power. These techniques of ―decentering‖ and ―deconstructing‖ the narrative

can help social workers and their clients see the world differently.

Perhaps Derrida’s most profound contribution to social work will be the idea of

“differance.” Derrida says that differance is a ―non-concept‖ as it does not refer to

anything that exists in reality. It is a pun on the French terms ―to differ‖ – differer – and

the English word ―to defer,‖ and refers to things that exist out of time, that are deferred,

but that would exist if not for constraints in the socially constructed world (Derrida,

1988). While Derrida avoided theory and method in his work, it is this writer’s opinion

that his literary analysis techniques will be able to provide the social work profession

with a very clear way of helping clients to analyze the social world in order to detect and

unearth the hidden workings of power.

Habermas’s theory of communicative action (2003; 1984, 1987) is directly based

on the work of Weber and Parsons and purports to solve the problem of how systems can
 74

experience such cataclysmic disruptions as the Nazi genocide. By adding an

intersubjective paradigm, in the form of the idea of the lifeworld, Habermas shows how

systems designed to solve problems in can develop a life of their own and diverge from

their originally intended functions.

The idea of power entered the social work dialogue in the form of empowerment

(Guittierrez & Ortega, 1991; Kondrat, 1995; Moreau, 1990). The meaning of

empowerment is as diffuse as the meaning of power and will be examined in greater

detail below; however, one can say that the common feature of the various definitions

encountered in the literature is that ―empowerment‖ is a process which increases the

power of groups that are in some way oppressed. Kondrat (1995) cautions, though, that

blindly empowering an individual or group may result in the oppression of another

individual or group, thus he unwittingly incorporates ideas of Habermas to show how

social workers can ensure that empowerment is mutual by including the interests of other

individuals or groups in the communication process.

Research into Power

In this section, several research studies on power will be reviewed in an attempt to

lay groundwork for the study of power qualities in bullying. The methods discussed range

from studies of community power structure (Gilbert, 1972); observation of the activities

of small groups (Gamson, 1982; Milgram, 1974); use of survey instruments with

families; and structured interviews with members of community action groups

(Guittierrez & Ortega, 1991; Keifer, 1984).

Dahrendorf (1958) expanded Marx’s notion of capital’s control of the means of

production to a mechanism of social control through what he called ―imperatively

 75

coordinated associations‖ in which the relations of power get institutionalized as

authority that has a right to dominate others. Bachrach and Botwinick (1992) suggest

that there is a strong current in society that sees compliance as essential to the stability of

the society; that belief mitigates the movement towards participatory democracy. Several

researchers develop this theme.

During and after World War II, a number of critical theorists living in the US

(Adorno, Frenkel-Brunswik, Levinson, & Nevitt Sanford, 1982) used social-science

technology to develop an instrument that they hoped would help identify individuals at

risk for becoming dangerously oppressive. The so-called ―F-Scale‖, or Fascist Scale, was

updated by Aletmeyer and his colleagues (1981) by including items from cognitive

research into inflexible thinking that they considered to contribute to authoritarian

personality and action.

Timothy Leary (1956) integrated some of Adorno’s items into the MMPI

(Minnesota Multiphasic Personality Inventory), the preeminent personality instrument of

its time, to include measures of interpersonal power. Leary developed two sub-scales to

measure dominance globally, the DOM and LOV: the DOM scale measures -- items were

grouped into a series of behavioral categories which he classes as either maladaptive of

adaptive: autocratic/managerial, dependent/ docile, masochistic/self-effacing; the LOV

scale measures behaviors such as acceptance of others, affection, and the tendency and

desire for affiliation. Leary’s scale has not been used widely12 but is one of the few

12
In fact, the scale was only available through the Harvard Library which only owns one copy. This may be
largely due to done to Leary’s academic reputation by his later activities promoting the use of psilocybin
and LSD for which he lost his academic position.
 76

examples of early power research that is predicated on a neutral rather than either a

positive or negative view of power.

Two prominent researchers on social power, Milgram and Zimbardo, conducted

experiments that were subsequently determined to have put their subjects at risk for

emotional and psychic damage. Milgram (1974) conducted a series of experiments

intended to test the limits of obedience in which subjects were made to believe they were

administering electric shocks to ―learners.‖ An actor representing an authority figure, a

researcher at Yale University, enjoined the research subjects to administer greater and

greater levels of shock to ―learners‖ whenever they made mistakes in learning word pairs.

The experimenter assured the subjects that there would be no permanent damage that

could result from any shock they administered. Subjects believed they were actually

administering shocks, while in fact, the learners were actors paid to sham reactions of

pain and protest. Despite clear labeling on the devices used to administer the ―shock,‖

fully two-thirds of the subjects administered shocks at levels that would have been deadly

had the shocks been real. The degree to which people were willing to obey authority was

far higher than the author predicted.

Zimbardo (1972), in a similar obedience, The Stanford Prison Experiment,

recruited subjects to act as prisoners and prison guards and created a make-shift prison in

which the roles were played out. Zimbardo found that acting out the role of prison guard

elicited sadistic and abusive behaviors in research subjects.

Milgram’s and Zimbardo’s experiments were widely believed to have traumatized

the subjects-- some of whom have lived for decades with the guilt that they are capable of

harming others-- contributing greatly to the movement to promote ethical

 77

experimentation on human subjects (Blass, 2000); however, their research speaks to

issues of power and, although unintentionally, its psychological effects in that some

subjects experienced emotional damage.

Gamson (1968) used an elaborate technique of misdirection to study what he called

―Rebellion Careers.‖ He secretly audiotaped volunteers discussing whether they will

comply with an unethical request by researchers to argue in favor of corporate interests in

a fictional trial in civil court. One-third of the participants were requested to argue in

favor of an oil company’s position and attempted to sway the others to make false

statements on videotapes that they were told would be used in a civil trial. The researcher

then left the room for a period of time. Gamson was trying to create a scenario conducive

to rebellion but not easily spotted by his subjects as a hoax. A successful rebellion career

was seen as a unanimous decision not to sign the affidavit allowing the fictional oil

company to use the videotape in the fictional trial. Gamson then analyzed and classified

the rebellion activities into seven categories of relevant action: compliance, evasion, rim

talk (questions and discussions about the scenario), dissent, resistance, direct action, and

preparation for future action. The author was originally attempting to study the actions of

defiance against authority; the high levels of compliance to authority he found had not

been anticipated.

Milgram, Zimbardo, and Gamson’s studies do not provide a methodology to study

bullying among children due to the obvious ethical constraints of putting children into

potentially damaging situations. However, the entirety of the literature on the effects of

bullying can be viewed as research into the effects of power imbalance on target children.

What would not be ethical in a research scenario or experimental design is available

 78

through the study of the secondary data into the social, emotional, physical, and cognitive

effects of bullying. Finding ethical and safe ways to study the conditions under which

children resist oppressive and abusive authority could help in the design of bullying

intervention programs; narratives of successful rebellion against bullying may be

available within virtually all victimized cohorts.

In a study more aligned with today’s ideas of protecting human subjects, Keiffer

(1998) looked at the process of community activism and the emergence of individual

activists and their ―transition from powerlessness to sociopolitical empowerment.‖

Keiffer chose a sample of 15 community activists, all of whom acknowledged that their

involvement in community activism precipitated a personal transformation. The

participants were active in more than one issue area, took a proactive rather than a

reactive position vis-a-vis social action, and had an ongoing commitment to social action.

Keiffer conducted interviews with the participants using the technique of dialogic

retrospection whereby the researcher involved the participants in a process of analyzing

their own data and elaborating the meaning of their personal transformation.

Keiffer found that there are a set of skills that include an awakening consciousness

of power dynamics through praxis, or a ―circular relationship of experience and reflection

through which actions evoke new understanding, which then provoke new and more

effective action,‖ and a set of skills to deal with constructive conflict. The author

describes empowerment as a ―continuing construction of a multi-dimensional

participatory competence‖ (p29). Keiffer suggests that further research with larger

samples will reveal a set of competencies related to community empowerment.

 79

A study of children who either overcame or avoided bullying could employ a

variant of Keiffer’s methodology. One could use Keiffer’s method to study

―emancipation careers‖ akin to Gamson’s ―rebellion careers.‖ It is essential to learn from

the experience of people who have successfully negotiated bullying and other oppressive

power experiences. This could lead to intervention programs that promote the safe

resistance to abuses of power.

Empowerment and Social Work

This section will discuss the history of empowerment and look at the research that

has directly studied empowerment.

As already mentioned, the concept of empowerment has yet to be defined in a

universally accepted way. Keiffer defines empowerment (as cited above) as the

―continuing construction of a multi-dimensional participatory competence‖ (Keifer, 1984,

p. 29)(Keiffer, 1984, p. 59). Gutierrez defines empowerment as ―a process of increasing

personal, interpersonal, or political power so that individuals can take action to improve

their life situations‖ (Guittierrez, 1990, p. 149); Rappaport states that empowerment is ―a

process, a mechanism by which people, organizations, and communities, gain mastery

over their affairs‖ and that social problems arise from the inequality of access to

resources and power among different groups in society (Rappaport, 1987). Simon (1994)

sees empowerment as the development and implementation of strategies to obstruct the

operation of power blocks, or, at least, to diminish their effects. Empowerment, in these

authors’ definitions, is the use of power in ways that address inequality across groups and

minimize oppressive structures. While all of these authors speak of increasing or

balancing power, they do not address how that power can be increased or balanced in an
 80

ethical way. As has been discussed above, the symmetry inherent in Kondrat’s (1995)

idea of ―mutual emancipation‖ provides a constraint: to increase power for one group

while not infringing on the autonomy of other groups.

Power is essential to make change in the social world. Gutierrez (1990) claims that

social workers can best create change by sharing power with their clients. She suggests a

number of strategies and principles to apply to that end: accepting the client’s definition

of the problem; identifying and building upon existing strengths; engaging in a power

analysis of the client’s situation; teaching specific skills; and working with the client to

mobilize resources. The goal is to reduce self-blame while increasing a sense of efficacy

and personal responsibility.

Gutierrez and Ortega (1991) studied empowerment strategies among Latino college

students. The authors hypothesized that an experience of group empowerment through

consciousness-raising would lead to greater commitment to further participation in

community activism. Gutierrez and Ortega (unpublished material referenced in Gutierrez

1991) developed their Change Strategy Scale based on subjects’ responses to an anecdote

about a person overcoming oppressive power.

Gutierrez and Ortega’s change strategy methodology(1991) is a significant

variation on Milgram’s and Gamson’s small group technique, in that, rather than having

subjects participate in a hoax, the authors gave the subjects an anecdote and asked a set of

questions designed to elicit what power strategies they would use to deal with that

situation. This methodology could be more easily adapted for use with children,

adolescents, in particular.
 81

Power and Bullying Experiences

A small body of research within the bullying literature touches, directly or

indirectly, on issues of power. In this subsection, these few studies will be discussed.

Power imbalance differentiates bullying from other forms of aggression. Power

differential is a factor determining the consequence of an aggressive act: Different types

of bullying experiences have different effects, qualitatively and quantitatively (Crothers

& Levinson, 2004; Mynard, Joseph, et al., 2000), and the element of power abuse may

increase or even determine the toxic effect of an aggressive action (Mynard, Joseph, et

al., 2000). Non-bullying aggression, such as physical confrontations among equals,

accidents, and social misunderstandings may be far less traumatic than aggressive events

that occur when there is an imbalance of power.

Selman and Demorest (1984) documented the negotiation strategies of two

children considered unmanageable in school settings during 50 hours of dyadic play

sessions led by a therapist. They examined the strategies children used to ―transform‖

themselves or others. The authors coded behaviors that they consider to reflect four

conceptual levels of social skills development and were rated as either self-transforming

or other-transforming. Using a modified event sampling procedure in which the actions

of the research subjects are observed intermittently for brief periods, two observers,

trained to identify negotiation strategies, took detailed notes of their observations from

behind a two-way mirror. The verbal content was transcribed, coded, and analyzed. The

authors offer this insight into the process of these two children:

In our view any strategy, regardless of its orientation or its developmental level,
represents an attempt to exercise some kind of control over a situation. A self-
transforming strategy is a particular way of controlling a situation in which the
medium through which control is achieved is self-adaptation; and conversely, for
 82

other-transforming strategies control is achieved by changing the other. The

emphasis on control, however, is not limited to one developmental level; control
is at the heart of all negotiation strategies. However, whereas the way that control
is asserted varies between orientations, the nature of the control that is sought
differs as a function of developmental level (1984, p. 303).

Selman and Demorest’s methodology could be adapted to gather detailed data on

children’s power tactics, including instances of bullying and the responses that bullying

evoked.

Mynard, Joseph, and Lawrence (2000), while not making explicit links to the

theory or literature on power, explicate the specific differential effects of power

imbalance and bullying. The authors studied 331 adolescents at one English secondary

school in an attempt to determine whether specific types of bullying are associated with

specific adverse effects. In particular, the authors tested the effects of relational, direct,

and indirect bullying on self-esteem and post-traumatic stress symptoms. All the

responding children were administered Mynard and Joseph’s Peer-Victimization Scale

(2000), the peer-victimization sub-scale of the Self-Perception Profile for Children

(Harter, 1985), and were also asked to respond ―yes or no‖ to the question of whether

they had been bullied. Forty percent of respondents were identified as having been

victims of bullies, fairly evenly divided between boys and girls. When the data were

analyzed for this group, bullying was found to be correlated with posttraumatic stress.

Relational bullying and social manipulation were found to predict posttraumatic stress.

Physical bullying and low self-worth were also found to be associated, with the subscale

of verbal victimization predictive of markedly lower self-worth than physical forms.

Mynard and colleagues’ results would need to be replicated in order for one to

draw definitive conclusions based upon them; yet, their evidence that different bullying
 83

experiences correlate with ill effects supports two important assumptions: 1) that

bullying, independent of preexisting factors, can be aversive; and 2) that social

manipulation, among other bullying experiences, tends to be experienced as more

aversive.

Raven and his colleagues (Raven, 1993; Schmidt, Raven, Pastorelli & Caprara,

1993; Schmidt & Raven, 1985) attempted to apply Raven and French’s (1960) adaptation

of Weber’s (1947) power analysis to children’s peer relations and bullying. The authors

undertook pilot studies using a cartoon instrument that depicts French and Raven’s

categories of administrative power in ways that apply to the world of children’s peer

relations (Raven, 1993; Schmidt, Raven, Pastorelli & Caprara, 1993; Schmidt & Raven,

1985). The authors conclude that there are parallels between the power used in childhood

and adult forms of power but did not pursue their research further.

Selman and Demorest’s (1984) idea of self- and other-transformation as a social

skill shares important qualities with Ducharme, Folino, and DeRosie’s (2008) more

recent study of acquiescence. The authors found acquiescence to be a fundamental social

skill required for all negotiation and cooperation. Bullying may be an other-transforming

strategy that children use when they are lacking in or disdainful of higher order skills: in

other words, bullying is an attempt to change the social environment.

In the following section, ideas that shed light on bullying effects indirectly or

tangentially will be explored.

Related Literature on Theory

Almost all the theory discussed in the bullying literature concerns the etiology of

bullying rather than the nature of its effects. In this section, ideas and theories from a
 84

variety of sources that could potentially be adapted to explain the mechanisms of effect of

bullying will be briefly outlined.

Categorical Thinking

Categorical or essentialist thinking (S.A. Gelman, 2003; S. A. Gelman &

Heyman, 1999; Giles, 2003) explains how bullies identify a child to victimize and may

also explain how bullying experiences can be internalized by the victim, thereby causing

distortions. Macrae and Bodenhausen (2000), in reviewing literature on cognition and

neuroscience, suggest two separate cognitive mechanisms at play in human thought.

Categorical thinking is a stable set of ideas and beliefs through which the individual

learns to negotiate the invariant features of a complex world, thought to correspond to

activity in the neo-cortex-- these ideas are learned slowly over time and tend to resist

change; flexible thinking, possibly located in the hippocampus, involves faster learning

and is sufficiently flexible to process novel events as they arise.

Macrae and Bodenhausen explore the ways that categorical thought processes can

lead to stereotyping when they state that ―…perceivers may use the contents of the

activated knowledge structure (the theorized neo-cortical structure) to derive evaluations

and impressions of a target, a process that commonly gives rise to stereotype-based

judgmental outcomes (Macrae & Bodenhausen, 2000, p. 95).‖ That children in general

exhibit the features of categorical thinking was established by Gelman (2003) who

conducted an experiment in which children were identified descriptively or categorically

in order to determine whether categorical thinking is a predisposing factor for

victimization, as is asserted by critical theorists (see below). Gelman designated children

into four groups: two groups were described as ―children who eat carrots,‖ and ―children
 85

who believe in creatures,‖ thought to be neutral stimuli; and two groups were labeled

―carrot eaters‖ and ―creature believers,‖ non-neutral, categorical, stimuli. Both labeled

groups, carrot eaters and creature believers showed evidence of stereotyping but the

descriptive groups, children who eat carrots and who believe in creatures, did not.

Categorical thinking may be responsible for much of the internalization of

victimization. That some bullies choose a target based on a categorical perception seems

clear in research on homophobic bullying (Dijkstra, Lindenberg, & Veenstra, 2007;

Poteat & Espelage, 2007; Warwick et al., 2004), but categorical thinking may also

explain some of the effects of bullying. The child who is affected by the bullying and

internalizes the experiences may be applying categorical thought processes to herself in a

deeply-embedded self-blaming process. To date, no research has been done on the

categorical thinking of victims, but it can be hypothesized that both bullies and victims

will exhibit less flexible thinking than their non-bully-involved peers. The presence of

consistently negative self-evaluations among victimized children may be indirect

evidence that these children have internalized a negative self-categorization. In the

systematic review below, variables such as low self-esteem and low self-worth may serve

as proxy indicators of categorical thinking internalized as self-blame.

Categorical thinking is a core concept of critical theory and was theorized as a

component of oppressive behavior; however, none of the researchers studying the

phenomenon of categorical thinking cite critical theorists in their literature reviews.

In the above discussion, categorical thinking has been viewed as a component of

internalization and victimization; however, there are most likely positive outcomes of

categorical thought processes as well. It seems likely that categorical processes are
 86

involved in the deeper illocutionary bonds about which Habermas theorizes.

Commitments, covenants, and contracts have a great deal of force within human relations

as breaking an oath or bond means jeopardizing one’s own social connections. Much

research would need to be done to determine whether the phenomena of illocutionary

bonds and categorical thinking are related.

Resilience and Strengths

Resilience is another concept that has implications in this review. The construct

of resilience, like power, social competence, and bullying, comprises a wide array of

behaviors, qualities, and conditions (Keyes, 2004; Luthar & Cicchetti, 2000), and the

study of resilience is in its infancy. Resilience factors are those whose presence in a

child’s life will predict positive outcomes in adulthood: social competence is one of the

most powerful resilience factors (Search Institute, 2007).

The primacy of social skills in the lifeworld gets empirical support from John

Gottman’s extensive research into marital breakdown (Gottman, 1994). Gottman, a

psychologist as well as a physicist with highly developed mathematical skills, coded the

speech/acts of couples as they interacted over a series of taped interactions collected over

decades. Gottman and his colleagues discovered a simple mathematical formula that

predicted divorce: couples who committed more than one negative speech/act in every six

interactions were statistically likely to divorce; those that were able to maintain positive

skills stayed together. These coded speech/acts included a range of interactions from

subtle facial expressions to overt and explicit expressions of support or derision.

Gottman’s work can be seen as evidence of the importance of primary social skills in

negotiating marriage, a primary relationship in adulthood can be deduced from Gottman’s

 87

work that children able to gain and maintain social skills will also have success in

negotiating relationships in childhood, reiterating the idea that social competence is a

major resilience factor. There is also a cohort of resilient children who have been bullied

but do not experience ill effects: these children will likely show attributes and actions that

would be considered resilient.

Unfortunately, within the bullying literature, considerations of resilience are

virtually absent. Research on the factors that protect a child from victimization or the

factors that mitigate the effects of victimization would be very helpful in designing

effective interventions.

Resilience, in the form of social support and social competence, is at the heart of

Habermas’s theoretical project. His conceptualization of ethical action is based not on the

relative basis of social construction but by possibly universal13 constraints of human

interaction. The most competent people will be able to put themselves into the standpoint

of others, no matter which culture she is part of. In fact, evidence will be sought from the

existing empirical literature for mechanisms of effect that alienate the social sources of

resilience, in this case, the source of resilience is peer support.

With one exception (P. K. Smith, L. Talamelli, et al., 2004), bullying research has

been conducted using a medical frame and based on the assumption that bullying

processes are necessarily pathological. Change to a positive and strengths-based

perspective would not require the assumption that bullying, the experience of multiple

humiliations and aggressive acts, is beneficial; but rather would make visible the many

13
Habermas tends to use the term ―universal,‖ yet is very clear that he is studying contemporary Western
cultures. This writer uses the qualifier ―possibly‖ before ―universal‖ because a great deal of research
would need to be undertaken to determine whether the social skills described by Habermas are truly
universal or are constrained by Western ideas of human rights that have emerged post Enlightenment.
 88

positive social experiences that bullying displaces, replaces, or prevents. In fact, a large

part of the effects of bullying explained by the theory of communicative action and

derived concepts involve the systematic alienation of the victim from sources of support

and repair available within the non-coercive peer environment.

Deeper bonds, such as family relations and personal friendships, may survive

bullying experiences, if only because they are not primarily based in school where the

bullying occurs; weaker bonds, however, within the peer environment, may be instantly

decimated depending on the type and severity of the bullying. Granoveter’s (1973) notion

of the importance of weak bonds, discussed above, will be more fully elaborated within

the framework of Habermas’s theory of communicative action.

Bullying and Play Deprivation

The concept of play, for this age group meaning forms of social free play, will

figure prominently in this paper. Play and Habermas’s key theoretical concept of

―communicative action‖ share several features.

The term ―play‖ was defined by Lever in 1978 as ―a cooperative interaction that

has no stated goal, no end point, and no winners (1978, p. 473); formal games in contrast,

are competitive interactions, aimed at achieving a recognized goal ‖ (Lever, 1978, p.

473). Later definitions echo Lever’s and suggest that, at higher levels of socialization,

play includes such behaviors as the rehearsal of adult roles and serves as a venue in

which children can practice a wide set of social skills, including the ability to influence

each other without resorting to coercion (Christie & Johnsen, 1987; Cole-Hamilton,

Harrop, & Street, 2005; Frost & Jacobs, 1995; A.D. Pellegrini & Smith, 1998; Rubin,

Fein, & Vandenberg, 1983; Rubin, Maioni, & Hornung, 1976; Spinrad et al., 2004).
 89

The definitions of ―play‖ and Habermas’s ―communicative state‖ share similar

features and may be essentially the same phenomenon14. Attempts to influence another

person, as are found in play, are characterized by Habermas as ―validity claims,‖

influencing others based on logic, norms, and relationship. Victimization puts an affected

child into what Habermas would call an instrumental state; for some children, this may

last only in the minutes they are being bullied, but for other children, the bullying may

more substantially alienate them from social support and play.

The play behavior of victimized children is yet another important area of research

only marginally featured in the bullying literature (Boulton, 1999; A.D. Pellegrini,

Blatchford, Kato, & Baines, 2004). Prominent bullying researchers Pellegrini and Smith

performed a meta-analysis of studies into the benefits to children of play and found that

the only benefits of play that can be empirically demonstrated are physical strength and

flexibility (A.D. Pellegrini & Smith, 1998). This lack of finding is likely due to the

methodological difficulties of tracking change over time. This project makes the case that

much can be learned from bullying research based on the fact that bully-involved

children are deprived of play to a greater or lesser degree.

While play deprivation has been identified in the conceptual literature at the time

of this writing, there has been no empirical study attempting to measure the effect of play

deprivation. This writer’s project may shed some light into the effect of play deprivation

by showing that some of the ill effects of victimization result from play deprivation by

virtue of the theorized idea that victimized children experience less communicative social

contexts which can be shown by demonstrating that the children play less.

14
A research protocol directly based on Habermas’s theory of communicative action would need to be
implemented to determine whether and what kind of play is similar to communicative action.
 90

Autonomy

The concept of autonomy has not figured in any theory of bullying or research. It

is this writer’s contention that restriction of autonomy is, as imbalance of power, a

common feature of all bullying interactions. Threat and coercion are overtly physical but

relational bullying, including status attacks, effectively limits the autonomy of the

victimized child within the social world. Although autonomy is implicitly subsumed into

Habermas’s theory of communicative action, it was the cornerstone of his earlier idea of

knowledge-constitutive interests which will be reviewed below.

Bertschinger and his colleagues present a relatively complex mathematical model

of autonomy that assumes that an actor is operating in a complex social environment and

needs to negotiate with that environment in order to exercise freedom of action

(Bertschinger, Olbrich, Ay, & Jost, 2006). In other words, autonomous action does not

mean taking action that is self-derived without consideration of the effect one will have

within the social world. A successful interaction will not only leave the environment

intact, but could, as Kondrat (1995) described it, be mutually emancipating.

Shapiro (1981, 2000) sees the restriction of autonomy as the essential component

for the creation of many forms of psychopathology, personality disorders in particular.

Yet another example of the interdisciplinary fragmentation Shapiro does not cite or

appear to acknowledge Habermas’s or any critical theorists’ contribution to the climate of

ideas from which a focus on autonomy clearly belongs. Shapiro shows how the

obstruction of autonomy distorts character in rational ways that fit perfectly with and

extend Habermas’s ideas of cognitive distortion. This also meshes with Kokko’s (2000)

notion of cumulative continuity in that these experiences accrete over time and pervade
 91

the individual’s choice of social life. For example, Shapiro would predict a strong

correlation between childhood bully involvement and adult sadistic and/or masochistic

relationships as at least some bully-involved children replicate the oppressive social

environment over time and spatial environment.

Where Shapiro shows the restriction of autonomy as essential to the formation of

psychopathology it is a thesis of this project that restriction of autonomy also deprives

children of curative and supportive experiences. This idea will be explored at length as it

involves the interpretation and extension of theory of communicative action.

A child whose autonomy is globally affected by her victimization experiences is

theorized to be more globally impaired. The existing literature will be plumbed for

evidence, if only indirect evidence, for this assertion.

This Writer’s Unpublished Research

Kogan and Chandan (2004), in an unpublished study involving extended

observations, interviews, and data-mining of clinical files from a pilot violence

prevention program in an elementary school, found data that both confirmed and

challenged the conventional wisdom present in the bullying literature. The authors found

that community violence was a factor in how one goes about reducing and preventing

bullying. The cousins, sibling, friends, and even parents of victimized children would

frequently directly intervene with the bully or his family, sometimes directly threatening

violent retaliation. There appeared to be a code of conduct toward violence, although this

researcher was unable to find an informant willing to discuss this frankly, that

stigmatized the brutalization of weaker children. The study took place in an urban
 92

community that, despite a rapidly decreasing crime and murder rate, still had many

incidents of physical violence and threat.

Teaching children to tell teachers about bullying may not be an appropriate

strategy in high violence communities. Informants came forward voluntarily in both

research and non-research venues to express vehement opposition to one aspect of the

violence prevention program being implemented in classrooms. Both youth and adults

made it clear to our staff that teaching children to go to a teacher for help if they were

being bullied or threatened was an unwise and potentially lethal strategy in their

community that had a viable drug trade. They explained that they did not trust that the

young children being taught to report bullying behavior to teachers would not do the

same in the more dangerous community context such as the drug trade. Community

members who have complained to police or had given evidence against drug dealers had

been violently intimidated and even killed.

Evidence in a small number of victims of bullying confirmed the toxicity of

categorical thinking (Macrae & Bodenhausen, 2000). One child who had been bullied

extensively revealed her deeply-held belief that some unknown quality inherent in her

nature was to blame for the bullying. Individual, classroom, and group interventions, as

well as an arts-based recreational program, helped her find a unique identity and adopt a

more flexible self-perception.

Gaps in the Bullying Literature

The foregoing review of the bullying literature suggests several gaps in the
literature and directions for future research. Despite the primacy of power imbalance in
the definition of bullying, no research has specifically focused on the power bases used
by bullies. Rigby (2004) suggests that bullying can be understood as a socio-cultural
 93

phenomenon; however, he does not suggest a specific social theory that might be applied
to the phenomenon
Other ideas or paradigms do not rise to the level of an expansive social theory but
may also elucidate the bullying phenomenon. For example, research into the idea of
interactive and cumulative continuity (Kokko & Pulkkinen, 2000) could elucidate the
processes of bully involvement: in particular, how children who enter school at risk for
bullying involvement may be less likely to transcend their victim or bully role. Kokko’s
theory of cumulative continuity would predict that children involved in bullying, in
whatever role, might repeat their experiences throughout childhood, staying with the
same friends or friends who share the same tendency toward bully-involvement.

The bullying literature does not provide a comprehensive picture of the entry into

and exit from the victim and bully roles. The literature does imply that some children

with few social skills are at risk for bullying the moment they enter the school system.

Although there is evidence that bullying is a stable phenomenon, it is also apparent that

children enter and exit bullying at different times in their school careers. Future research

may collect data on entry and reentry into victimization to determine the differences in

context and constitution of symptomatic victims from those children who do not become

symptomatic after being bullied. As touched on above, community violence has been

largely neglected as an environmental variable that may affect the quality and quantity of

bullying.

To effectively measure the effects of bullying, researchers should isolate the

cohort of bully victimized children who were not poly-victimized and determine bullying

effects in the absence of significant confounding variables.

Very little research has been done to determine any benefits or perceived benefits

of bullying. No study directly or indirectly asked parents and teachers whether they

supported bullying or accepted it as a normal part of childhood. Some parents, teachers,

 94

and older siblings of children may well support bullying as a way for the children to gain

a competitive advantage while other adults may value the submissiveness of victimized

children. As described above, Nazi policy appeared to directly promote bullying as a way

of preparing young people for adult roles of coercion. An historical study of pedagogy

and children’s literature may unearth other examples of techniques used to promote the

social reproduction of coercive roles.

Only two studies in the literature investigate whether teachers themselves have a

tendency to bully or scapegoat children (Omigbodun et al., 2008; S.W. Twemlow et al.,

2006). It seems likely, given the fact that local factors often determine the prevalence of

bullying, that the power tactics of teachers and school administrators could determine the

presence of bullying. A large- scale ethnographic study examining leadership styles in

high and low bullying schools could shed light on these local factors15.

Play is another neglected area of bullying research. It is likely that both bullies

and their victims are deprived of social play given the isolating nature of the problem.

While there is research to show that bully-involved children tend to be less successful

socially, there has been no research specifically correlating bullying involvement and its

effects on play behavior. Buhs and Ladd (2001) contend that children with compromised

social systems and poor social skills lack the opportunity to play. Aside from rough and

tumble play, which mimics bullying and aggression in a play context, bullying cannot be

considered play. Children involved in bullying are not playing, at least while the bullying

is taking place. It would be important to determine the degree to which bullying

involvement deprives children of play during and after the bullying occurs. Play

15
Such a research project would present some methodological challenges. In a host setting it would be
challenging to collect data that has the potential to embarrass the host of that setting.
 95

deprivation may, in fact, explain some part of the negative effects of bullying. If, as

theorists predict, play helps children learn social skills and rehearse contemporaneous

situations and adult roles (Cole-Hamilton, Harrop & Street, 2007; Frost & Jacobs, 1995),

children deprived of play would have a serious impediment to successful adaptation.

Research could determine whether bullying changes the quality or quantity of play

activities.

To date, no research has been done to adapt critical theory to the study of

bullying. The theory has largely been used in legal and political theory (Habermas,

1996); however, Habermas’s theory of communicative action offers a full social theory

that is intended to explain a full range of social actions and could as easily be applied to

the life of the individual as to the life of a nation. In fact, Habermas himself often

illustrates his ideas with mundane examples of interpersonal relations – his mechanic, or

a person staying late at a party (1984) within his own lifeworld (in addition to giving

examples from larger contexts). Play behavior resembles Habermas’s description of

communicative action as an environment in which meaning is made in the absence of

specific goals. While variables may be used as stand-ins, it would be important to

undertake empirical studies to determine the salience of such Habermasian concepts as

illocutionary bonds and communicative processes of support and repair by directly

operationalizing these concepts.

Theory-building has been identified by Morrison (1994) and Rigby (Rigby, 2003)

as a serious gap in the bullying literature. Despite the lack of a theoretical foundation,

school boards are mandated to intervene to end bullying. As Dubin (1978) makes clear,
 96

an intervention is unlikely to be successful until a good theory is found to explain and

predict the phenomenon to be treated.

In light of Dubin’s observation, this writer’s project will attempt to tackle this

lack of theory building in the bullying literature by reviewing the current knowledge base

and by applying Habermas’s conceptual framework to see whether and to what degree his

theory can explain all or part of the bullying phenomenon.

Habermas’s Theory of Communicative Action

In this project, the German philosopher and critical theorist Jurgen Habermas’s

theory of communicative action will be tested as a conceptual framework with which to

understand the effects of bullying. This section includes a brief summary of the aspects of

Habermas’s social theory that will be applied in this project. In the subsequent research

design section, the theory will be adapted to the study of bullying and used to derive a

conception of some of the mechanisms through which children are affected by bullying

experiences. An in-depth analysis will be undertaken in that section which will include

and interpret Habermas’s own words.

While Habermas does not tend to use the terms ―power‖ or ―power imbalance,‖

his project can be understood as an analysis of different forms of power as applied to the

creation of just social action. Bullying will be conceptualized, according to Habermas’s

schema of social action, as instrumental action that limits the autonomy of the target (and

the other children involved). Different types of bullying limit the children’s autonomy in

one or more domains of life, physical, emotional and social domains in particular. By

contrast, consensual activities such as those that occur in free play fit the schema of
 97

communicative action in which children are highly autonomous and influence each other

in non-coercive ways.

As previously discussed, no research study has operationalized the concept of

play deprivation; however, that victims of bullying and other bully-involved children

have lower levels of social play can be deduced from studies that show children involved

in bullying are more isolated socially: at least during the minutes when the bullying is

occurring, the children involved are not ―playing.‖ This deprivation of play may explain

some of the continuing and cumulative effects of bullying in that the bully-involved

children experience less communicative action (autonomous) and more instrumental

action (coercive) than their peers. The children, through this lack of communicative

action, therefore, may not be learning essential social skills negotiation, compromise and

informal forms of positive leadership.

Habermas theorizes that communicative action is a realm of human interaction

that consists of a myriad of often small acts of solidarity, consensus-making, and

meaning creation. Bullies’ and victims’ limited access to this domain of interaction may

explain a large part of the negative sequelae of bullying and other forms of power abuse.

In light of the fact that there is no empirical literature showing the positive effects of

communicative action, a systematic review of the bullying literature may be able to

demonstrate the deleterious effects of the absence of communicative action. In addition to

limiting the actual opportunity to interact in a consensual environment, bullying may also

lead to ―distortion of cognition.‖

Habermas’s (1987) notion of the distortion of cognition corresponds with the

ideas of Dodge and Frame (1982) in which bullies are often considered to be acting on a
 98

hostile attribution of social cues due to an inaccurate internal representation of the social

world. Distortion of cognition may explain some of the effects of bullying, not just the

reasons for bullying, in that a child who internalizes a negative self-concept as the result

of bullying will have distortions of self-concept, identity, mood, and have fewer

opportunities to learn and practice social skills. Several studies show that such variables

as self-esteem, identity, and self-beliefs mediate the effects of bullying and other forms of

aggression (Burt, Obradovic, Long, & Masten, 2008; Schmidt & Bagwell, 2007). These

mediating variables, then, can be examined to see if they are consistent or not with

Habermas’s theory.

The following section will clarify the research design and methods chosen for this

project.
 99

PART III TYPE OF RESEARCH DESIGN

Opening Statement

This dissertation is a systematic review of the research literature into the effects of

school bullying published in the United States within a ten-year period, with the goals of

developing a model of these effects and of testing German philosopher and critical

theorist Jurgen Habermas’s theory of communicative action with the real world social

problem of school bullying. Systematic reviews and meta-analyses for theory building

are relatively new techniques done in the fields of organizational development (Hassard,

1991; Lewis, 1999; Viswevaran & Ones, 1995; Yang, 2002), psychology (Cheung & Au,

2005), and, under the term ―conceptual ontologies data mining‖ (Blagasklonny & Pardee,

2002), in the fields of information technology and biomedicine (Gottgtroy, Kasabov, &

MacDonell, 2003). Dubin’s (1978) schema of theory building is being used because it is

designed to build theory from empirical research and practice.

In this section, the research design and methodology will be outlined as will the

rationale for the design. The goals of the project and the feasibility will be briefly

outlined. The systematic review technique will be described and related to the problem of

bullying and to the process of testing and building a theory.

Current State of Knowledge

The research into the effects of bullying, as with most bullying research, has been

conducted using different and, at times, incompatible definitions, measurements, and

methodologies (Grills & Ollendick, 2002), also called ―shared method variance‖ (Haynes

& Hayes O'Brien, 2000). Both the dependent and independent variables in effects studies
 100

have been measured with a variety of techniques. In consequence, the studies into the

effects of bullying present a confusing picture of differing results. While there is a small

number of theories consistently applied in the bullying research, these theories have

tended to explain only the etiology of bullying involvement. No significant work has

been done to create or adapt a theory to explain the effects of the bullying phenomenon.

Intervention in bullying in the absence of a clearly articulated theory

underpinning it is unlikely to be successful simply because when a phenomenon is not

adequately understood the practitioner does not know how to effectively intervene

(Dubin, 1978; Kaplan, 1964; Van de Ven, 1989). In spite of this lack of theory, bullying

intervention programs are proliferating in the US and elsewhere. There is no evidence

clearly showing that any particular bullying intervention or prevention program is

consistently effective, while some evidence shows that certain intervention programs

actually increase bullying or help bullies hide their actions from adults (Dixon et al.,

2004; Rigby, 2003).

Despite the now universally accepted definition of bullying, which incorporates

the feature of power imbalance inherent in the bullying phenomenon, no research has

been done to determine the similarities and differences between bullying and other forms

of power. Nevertheless, bullying studies can be seen as a rich data for the study of the

effects of power, particularly those studies which were conducted using the contemporary

definition of bullying (which includes the feature of power imbalance). Among

researchers into power and power relations, only Schmidt and Raven, reviewed above

(1993; 1985), have looked at children’s relationships to deepen the understanding of

power.
 101

Research Question

The essential question for this project, ―What are the effects of bullying on the

target children (victims16)?‖, will be asked at several levels of conceptualization from the

implicit or stated theory in the research through the component concepts, the variables

measured, and the instruments and techniques used. Jurgen Habermas’s theory of

communicative action will be used as a conceptual framework with which to understand

all or part of these effects.

Goals
The goal of building knowledge within critical social work, as within critical

theory, is to emancipate people from active oppression in the present and to promote the

conditions under which they can fully participate in the lifeworld. The primary critical

objective of this project is to develop a theoretical understanding of the effects of

victimization in order to free children from the oppressive experience of being victimized

by bullying and, possibly, other forms of victimization. A second goal is to empirically

test the efficacy of Habermas’s theory of communicative action in reference to the study

of bullying and of peer relationships.

It is this writer’s contention that the emancipatory goal of this project is identical

to Jurgen Habermas’s goal of creating the conditions for a peaceful democratic society.

Whereas Habermas applies his theory to nation building and the creation of constitutions,

in this project, this writer looks at the conditions under which children can become free of

oppression at a local level. This writer contends that children able to handle peer

16
For the research project, an attempt will be made to avoid using the terms ―bully‖ and ―victim‖ to refer to
children as the categorization itself may be iatrogenic.
 102

oppression effectively will grow into adults able to participate and contribute positively

to the socially just constitutional state.

The review of US empirical studies will be used to create a typology of the effects

of bullying on target children and to propose at least one mechanism by which children

are affected by school bullying. The measures and definitions used in the research will be

organized into typological categories in the interest of conceptual clarity, since, as noted

above, definitions and measures used in bullying research vary widely. This writer

intends that this project will make a contribution to the theoretical understanding of the

ways that the bullying phenomenon and, perhaps, other forms of power abuse, affect the

target child.

Systematic Review for Theory Testing and Building

This project will attempt to build a theory by using data from research studies

published in the US within a ten-year period. A systematic review--a targeted review of a

set of studies (Petticrew & Roberts, 2006)--will be conducted on research studies into the

effects of bullying to build a model that can explain the mechanisms of effect. As Lehnert

(2007) suggests, this review will organize the array of variables already studied in the

bullying literature to build a model of the effects of bullying.

McGuire (1997) asserts that reorganizing an existing body of literature can itself

be a method of original discovery by bringing together complementary studies and

reconciling studies with conflicting outcomes. In the case of the bullying literature, this

will entail clarifying the concepts used in the research studies that were conducted using

different methodologies and different definitions to determine whether insight can be

gained about the mechanisms of effect through aggregating these results.

 103

Only US studies are being selected for this review for several reasons. The

primary reason is that US schools do not yet have significant anti-bullying initiatives is a

confounding variable that makes it difficult to compare Scandinavian, European, and

Commonwealth study result. US children, in the absence of such programs, are likely to

experience the aftereffects of bullying more strongly than their European counterparts as

the US children are much less likely to have access to informal and professional sources

of support. The research of Wolke and colleagues (Wolke et al., 2000) indicates that

victimized children living in places where there is a stigma against bullying and services

are less symptomatic than children living in places where bullying is largely ignored (i.e.,

the US): the results of their cross-national study show that children in countries where

there is a great deal of service and support in the schools for bullied children.

Due and colleagues (2005) compared the symptoms of children identified as

victims of bullies from 28 nations and found that the victims in the US and Israel ranked

significantly higher in symptom rates and severity than those in the European countries

with which they were compared. While it would be hard to determine all the factors that

explain this sharp difference, the US and Israel share two features that constitute

confounding variables: (1) both countries had significant wars during the period in which

most of the studies were conducted and published (US—Iraq and Afghanistan; Israel—

Lebanon and Palestine); and (2) both countries occupied other countries during this

period, which has raised human rights’ and other concerns. The US engaged in

systematic methods of torture in Iraq (Bybee, 2002) amid a media climate that was, in

some cases, tolerant of torture and other oppressive tactics carried out by government

officials, and Israel has used its military to attack terrorists (or freedom fighters using
 104

terror tactics depending on one’s frame of reference on the Middle East) in the West

Bank and Gaza, harming civilians in the process.

Another reason for limiting this review to US studies is that the term ―bullying‖

may be understood differently across cultures. Definitions of bullying vary among

languages and cultures (P. K. Smith & Brain, 2000; P. K. Smith, Cowie, Olafsson, &

Liefooghe, 2002). Despite regional linguistic differences, most US children will

understand the term in the same way, while speakers of other languages understand

connotations that differ from American English usage.

 105

Systematic Reviews and Meta-Analyses

Meta-analyses are often paired with systematic reviews. The research studies used

in meta-analyses need to measure the same phenomenon, using the same or similar

techniques. Various meta-analytic techniques are often used to determine the most

effective technique or clinical procedure in several outcome studies. Several statistical

techniques can be applied depending on the nature of the data and the question being

asked: the cumulative effect size can be generated; the homogeneity of the collected

results can be determined; compatibility with a theory or model can be measured using

complex factor analysis and path analysis techniques both separately and together in the

form of structural equation modeling. The advantage of meta-analysis where there is

homogeneity of method is that the use of several studies can dramatically can increase the

sample size and, therefore, increase the power of the statistical relationship.

A statistical meta-analysis is not feasible for this project because of the

tremendous heterogeneity among the variables in the selected studies. All of this project’s

selected studies report data on the correlation between victimization and at least one of its

hypothesized effects but very few use the same instrument or methods with a comparable

sample. In addition to the difference in method used to determine the presence of

victimization, different studies among those selected sought to determine the presence of

bullying during different time periods. Some asked only if the bullying was happening

contemporaneously, while others asked about differing retroactive periods.

Contemporaneous samples would be more likely to have accurate measures of the

immediate effects of bullying but may miss some of the longer term effects by
 106

eliminating escaped victims, children who were no longer being victimized and who may

be asymptomatic.

The most basic technique of presenting data is the ―vote-counting‖ technique

which simply tabulates and compares the relative effect sizes on their face value without

using a statistical technique to determine aggregated effects sizes. This review will use

this vote-counting technique to examine the results on their face value. This method

allows for simple, descriptive techniques. The range, significance, and direction of the

study results for each variable will be presented in table form, and these scores will be

averaged. Where feasible, averages will be generated to allow a comparison between age

groups. The method will be more fully elaborated in the data analysis section below.

Vote counting does not allow the researcher to weight effect sizes according to

the number of subjects used or the type and acuity of the instrument. More advanced

techniques will have to wait for a more homogeneous set of studies to be undertaken.

Lipsey and Wilson (2001) coined a term for the variation in context and method

among different studies purporting to measure the same variables: ―the apples and

oranges effect.‖ This project by aggregating and analyzing the various measures and

concepts in the disparate studies, intends to bring some clarity to the apple and orange

effect of the results of the studies. The writer’s main method for bringing about this

clarity is the use of constructed typologies discussed in the next section.

Constructed Typologies

A systematic review can be used to determine the state of knowledge and to begin

the process of building or improving on an existing theory. This is achieved by clarifying

 107

and rethinking the variables used in various studies, in this case, to develop a model to

begin building an understanding of how bullying affects the target.

The call for conceptual clarity is consistent with the evolution of the social

sciences, as succeeding generations of researchers used different concepts and measures

to describe and explain the same social action in different contexts (and used sometimes

used the same concept to describe different phenomena). Lazarsfeld (1937) credits

Hempel and Oppenheim (cited in Lazarsfeld, 1937) with describing and codifying the

ways social scientists borrowed techniques from the hard sciences by developing ―types‖

which Lazarsfeld describes as ―special compounds of attributes‖ (Lazarsfeld, 1937, p.

121), or the organization of numerous ideas or concepts according to a shared feature.

Lazarsfeld (Lazarsfeld & Henry, 1968; Lazarsfeld & Rosenberg, 1955) and McKinney

(1966) adapt classification methodologies from the hard sciences through the use of

―typologies‖ that reduce complex groups of concepts to more manageable dimensions.

The use of powerful computers has extended this process by freeing conceptual

clarity from the constraints of human intelligibility in the development of typologies.

Conceptual ontologies data-mining allows computers to compare data across dimensions

of human action, including numbers of factors beyond the capacity of most human beings

to manipulate (Blagasklonny & Pardee, 2002; Gottgtroy et al., 2003; Gruber, 1993); these

conceptual connections are then made intelligible to humans through concept mapping

(Trochim, 1989).17 This project does not have data sources large enough to require

sophisticated data- processing techniques, but the process of building a theory is the

same. Given the current public interest in the topic of bullying and the availability of

17
Conceptual ontologies data mining is beyond the scope of this project due to the limited scope of
available data; however, with web technology, the available data could rise exponentially in the near
future.
 108

―crowd sourcing technology‖-- the use of the Internet to gather large amounts of data

very quickly--it is possible that this and other projects like it could be the beginning of a

much larger process of data gathering and theory building on bullying.

This project will be limited to the creation of explanatory and descriptive

typologies. This writer’s rationale for choosing Habermas’s theory of communicative

action as the conceptual framework for the project is based largely on Habermas’s

typology of instrumental action which will be shown below in Figure 2. As will be

discussed, the typology of strategic action, a type of instrumental action, corresponds to

the types of indirect bullying that involves the manipulation of the social environment

that have emerged from over 30 years of research. Briefly, Habermas proposes a set of

categories of instrumental action, including direct action and strategic action, in which

others are recruited to elicit an action: this is consistent with the ways that bullies

victimize their targets.

Figure 2: Habermas’s Typology of Social Action

(adapted from Habermas, 1971)

Social Action

Communicative Instrumental
Action Action

Concealed Open Strategic

Strategic Action Action

Conscious
Deception

Unconscious
Deception
 109

The question in this project is whether Habermas’s theory of communicative

action or a derived theory can also provide an explanatory typology--a typology of the

concepts and relationships between concepts that explain a phenomenon (Lazarsfeld &

Henry, 1968). To that end, the concepts and measures used in existing bullying studies

will be catalogued and abstracted into typologies. The creation of a typology involves the

compression of large amounts of data into a smaller set of concepts. These typologies

will be compared with Habermas’s theory of communicative action. This will entail

organizing the many variables studied in the disparate studies into logical categories.

Theory Building and Testing

Theory building and testing is largely the work of the imagination (McGuire,

1997; Weick, 1989), often the imaginations of multiple contributors (Kaplan, 1964)

working directly or indirectly on the same phenomenon. According to Kaplan, theory

formation:

stands for the symbolic dimension of experience, as opposed to the apprehension

of brute fact. The content of our experience is not a succession of mere
happenings, but a sequence of more or less meaningful events, meaningful both in
themselves and in the patterns of their occurrence…the device for interpreting,
criticizing and unifying established laws, modifying them to fit data unanticipated
in their formulation, and guiding the enterprise of discovering new and more
powerful generalizations (Kaplan, 1964, pp. 294-295).

Theory building is an attempt to explain processes that are not evident or

necessarily visible (Brante, 2001). These processes may be ―latent variables‖, either

unseen or unnoticed, whose relationships are laws or rules that can explain and predict

outcomes and relationships among ―manifest,‖ or observable, variables. Where the

relationships among variables and properties cannot be perceived, the researcher must

rely on the creative imagination, albeit based on real world experience. As Dubin (1978)
 110

states, all theory begins with our understanding of the world, whether this takes the form

of remembered experience (Habermas, 1987) or extended observation with detailed notes

and recordings of events and conversation. For the research into bullying, processes exist

at different levels of observability.

Dubin (1978) suggests that there is a dynamic process that occurs among theory,

lived experience (for which he uses the German term verstehen, or understanding of

human activity), and empirical research. Each one informs and enriches the other. He

sees no discontinuity or inconsistency among qualitative paradigms, conceptual

processes, and empirical research. He states that there are three results possible in the

testing of a theory or model: 1) the theory and empirical results are identical; 2) the

theory is not sufficient to explain all the results; and 3) the results go beyond the

boundaries of the theory. In this case, the empirical literature on the effects of bullying

will be compared to predicted results based on the research.

Theory itself can bring phenomena into human consciousness. Research and

theory serve to make seemingly obscure processes visible. Latent, theorized, variables

are often considered to be invisible, but in fact, the visibility of theorized variables is a

relative phenomenon (Price, 2007) as are the ways that theory can make social processes

visible. Marx and Freud and their colleagues and, later Einstein created revolutions in

thought by theorizing about invisible processes in the social world and the physical

universe, respectively. For example, Einstein rethought the nature of light asking himself

what if light were neither a particle nor a wave but instead had qualities of both. This was

a theoretical idea up until 1919 when observers found that light from a distant star

passing the sun during an eclipse was bent by the sun’s mass. Not only did the light
 111

curve, it curved exactly as Einstein’s theory predicted. Light clearly had a small amount

of mass. Subatomic physics experimentally proves the existence of elementary particles

by measuring their collision with measurable particles.

In the case of bullying, despite the calibration of bullies which maintains their

activities sub rosa (Gumpel, 2000), adults also substantially ignored the bullying, perhaps

assuming that it was a rite of passage that could not be stopped. One can say that

subjective psychodynamic processes, such as Sigmund Freud’s ideas of the subconscious

process, are invisible; yet Rensis Likert (1932) developed a simple method of asking

people to rate their subjective process on a numeric scale: the Likert Scale makes

subjective processes visible by simply asking the individual to rank their subjective

experience. In an example from the bullying literature, the cognitive effects of bullying

occur within the mind of the child and are accessible only to the degree that the child is

developmentally able to express herself; however, any lessening of social play is visible

to the researcher who can observe and record the child’s and others’ actions.

Time is an essential element of theory. Victimization can have both immediate

and long-term effects. The studies selected for this review measure a large number of

effects which may or may not have a temporal relationship with each other. It is likely

that a logical model will emerge from this review showing how being victimized affects a

child over time. It is hoped that such a model will be the first step in answering

Finkelhor’s (1995) call to enrich our understanding of how victimization affects the

longer term development of a child.

 112

Components of the Research

Though the methods adapted for this research are varied and drawn from different

sources, they have been chosen for their capacity to answer a series of questions that will

give a comprehensive picture of the present state of knowledge in the field of bullying

and how that knowledge relates to Habermas’s theory of communicative action. The

research will involve two stages: a systematic review of selected existing studies, and the

testing and building of a theory.

Stage 1: Systematic Review of Selected Existing Studies

This stage of the research will analyze selected existing studies to create a

typology and model(s) of the effects of bullying. Conceptual clarity will be sought in

order to reconcile or differentiate data gathered using different definitions and methods to

create a comprehensive understanding of the present state of knowledge.

The questions guiding Stage 1 are:

 What are the concepts being applied to the effects of bullying?

 How are the variables that comprise the concepts being measured?
 What are the different effects of bullying, the results in terms of effect size?
 How can the many disparate variables be organized into typologies?
 Does the aggregated literature suggest a model or models of the mechanism of
effect?

Stage 2: Theory Testing and Building

The second stage of the research will relate the concepts and research results to

Habermas’s theory of communicative action and derived concepts and constructs. This

will take place in the analysis section of the project.

The questions guiding Stage 2 are:

 What model can be established based on Habermas’s theory?

 Do the concepts in the studies relate to those in Habermas’s theory?
 Are there relevant concepts that are not explored in the studies?
 113

 Can the results of existing studies be compared with a conceptual model based on
Habermas’s theory?
 Are there data that disconfirm Habermas’s theory?
 Are there data that suggest an alternative model to Habermas’s?
 Are there data that deepen or extend Habermas’s theory?

Examples of Studies Using Similar Methods

Hawker and Boulton’s (2000) review of 20 years of studies into the adjustment

effects of victimization very much inspired this project. The overall method of this

project is similar in several key ways. Like the Hawker and Boulton review, this project

converts study results to a common metric; also like this project the authors used the

vote-counting method18, presenting the results and a group mean for each category that

was not weighted according to the sample size. Hawker and Boulton limited their

descriptive analysis to concurrent studies and grouped their selected studies according to

the type of bullying definition used. Hawker and Boulton’s contribution was a milestone

as it helped to clarify and summarize the disparate definitions that had been used in early

bullying research.

Kim and Leventhal (2008) performed a systematic review of the research studies

that looked at bullying and the single effect of suicidal ideation and behavior. For the

same methodological reasons as are used in this project, the authors felt that a full meta-

analysis was not feasible as the dependent and independent variables were measured in

very different ways. This is termed high ―method variance‖.

The theory-building aspect of this project has close parallels to the work of Bai

Yin Yang (2002) and Lewis and Grimes (1999) who proposes the use of meta-analysis as

18
Hawker and Boulton use the term meta-analysis in their study. As in this study the studies available to
Hawker and Boulton lacked homogeneity. Hawker and Boulton can be said to have provided a set of
statistics describing the body of study results including their ranges and averages.
 114

a theory-building tool. Yang applies his technique to the literature on organizational

development. Lewis and Grimes’s (1999) meta-analysis of organizational literature

shows the literature in that field is rich in theory. A pilot study reviewing the bullying

literature in preparation for this project showed that the bullying literature is very thin on

theory. Where Yang and Lewis and Grimes were building on a rich tradition of

theorizing, this projects is taking place at the preliminary stages of theory building,

identifying potential latent constructs and processes within the empirical literature and

comparing them with Habermas’s theory of communicative action.

The theory analysis aspect of this project will be modeled upon three studies:

Margolin and Goldis (2000), Painter and colleagues (2008), and Abramovitz (1981).

Margolin and Goldis’s (2000) systematic review of the literature on the effects of

community violence on children resembles this project in that it surveys a range of

studies that show a range of effects (although not nearly as rich a source of data as the

bullying literature provides). The authors, however, do not include any theory in their

review. Painter and colleagues (2008) surveyed the theories used in a selected review of

empirical research articles into behavioral health, the field that promotes a healthy

lifestyle. They found that less than 5% of researchers clearly articulated any theory in

portraying their research. This project will similarly survey the selected studies for the

direct use of theory, but will also attempt to detect latent theories at play.

Abramovitz (1981) studied Progressive Era documents to determine the influence

business played on the political processes and alliances that supported or thwarted

worker’s compensation and health care reform in the US. She reviewed literature from

the conceptual frameworks of three dominant socio-political theories. Similarly, this

 115

project will survey the (little) theory that exists and propose a particular theoretical frame.

However, the studies in this project are not homogeneous, and only one theory, not three,

will be used for the conceptual framework. The homogeneity of the documents in

Abramovitz’s study lent themselves to a more systematic approach in the review since

the concepts and results were more able to be organized into tabular form.

A few studies indirectly parallel this project by using existing data sources to

build a theory and knowledge in general. Hripcsak, et al. (retrieved September 2007)

examined over 45,000 paper medical files in order to create a protocol for automated

error detection and error prevention that could search medical files stored on computer

databases or in data warehouses. Sequences and patterns of events that led up to a

medical error were drawn from case files, and the patterns were then translated into

searchable terms. The intention was to create an automatic search mechanism that could

identify variables, sequences, and constellations of variables that indicate a risk of

medical error.

Another project that marries case file data mining and automated searches of

databases is the domestic surveillance campaign carried out by the US National Security

Agency. While the project is secret, evading even Congressional oversight19, much of the

technology and methodology used is in the public domain (Department of Defense, 2005;

Arkin, 2007). The data-mining aspect of the domestic surveillance project does not

actually monitor the verbal content of telecommunications, but rather searches enormous

electronic databases for patterns that resemble the patterns of communications mined

from FBI and CIA files. These patterns may include a mapping of the communication

19
This section was written during the second term of US President GeorgeW. Bush, whose administration
authorized domestic surveillance.
 116

that known-terrorist plotters used to coordinate past failed or completed attacks (Bulletin

of Military Operations Research, 2005), as well as patterns characteristic of suicide

bombers preparing to die.

Keshvala (2008), in a dissertation that comes closest in method to this writer’s

project, tests social capital theory using secondary data available through the Education

Longitudinal Study (ELS) which gathers a wide range of educational data from all US

schools. Keshvala begins with a theoretical model and tests to see how well the theory

can explain the existing data. Keshvala uses structural equation modeling to build a

theory about the effect of social capital on children’s academic success. This project will

limit the theoretical analysis and an examination of the face value of the effect sizes

through factor or cluster analysis.

Theory of Communicative Action

In this project, Jurgen Habermas’s theory of communicative action will be tested

as a conceptual framework with which to understand the effects and etiology of bullying.

While Habermas does not use the terms ―power‖ or ―power imbalance,‖ his whole project

can be understood as an analysis of different forms of power as applied to creating social

action. Bullying will be conceptualized according to this schema of social action. Within

the schema, bullying can be viewed as instrumental action that limits the autonomy of the

target (and the other children involved). Consensual activities such as those that occur in

free play, by contrast, fit the schema of communicative action i.e., the action of involved

individuals is highly autonomous. Different types of bullying limit autonomy in one or

more domains of life.

 117

No study has operationalized the concept of play deprivation; however, the fact

that bullied children have lower levels of play can be extrapolated from studies that show

children involved in bullying are more isolated socially and less active: at least while the

bullying itself is occurring the children involved are not playing. This deprivation of play

may explain some of the continuing and cumulative effects of bullying in that the

children experience less communicative action and more instrumental action. They may,

therefore, not be learning essential social skills that could enrich their lives.

Habermas theorizes that communicative action is a realm of human interaction

that consists of a myriad of, often small, acts of solidarity, consensus-making, and

meaning creation. Limiting access to this domain of interaction may explain a large part

of the negative sequelae of bullying and other forms of victimization. In light of the fact

that there is no empirical literature showing the positive effects of communicative action

a systematic review of the bullying literature may be able to show the deleterious effects

of its absence. In addition to limiting the actual opportunity to interact in a consensual

environment bullying may also lead to distortions of cognition.

As was discussed above, Habermas’s (1987) idea of the distortion of cognition

corresponds with Dodge and Frame’s (1982) idea of cognitive distortion within the

internal representation of the social world. Distortion of cognition might explain some of

the effects of bullying by adversely affecting self-concepts, identity, mood and the

opportunity to learn and practice social skills. Several studies show that such variables as

self-esteem, identity and self-beliefs mediate the effects of bullying and other forms of

aggression (Burt et al., 2008; Schmidt & Bagwell, 2007). These mediating variables,
 118

then, can be examined to see if they are consistent with Habermas’s theory. That theory

will be more fully described in the following section.

 119

PART IV ELABORATION OF THE THEORY OF COMMUNICATIVE ACTION

In this section, Habermas’s theory of communicative action will be set into its

historical and intellectual context, and will be fully elaborated and applied to the problem

of bullying.

The term ―critical theory‖ refers to a diverse body of social and political thought

that shares common interests; an interest in crossing intellectual and research disciplines,

and an interest in assuring that social science acts to emancipate rather than oppress

people; and, in some cases, a common history (How, 2003).

Critical theory began in the 1920s in Frankfurt, Germany, at the Institute for

Social Research, ―the Frankfurt School,‖ as the proponents of critical theory inside and

out of the institute came to be known. The Frankfurt School was funded by private capital

to encourage Marxist scholarship and to provide a venue for Marxist intellectuals who

had been denied a place in mainstream academia (Jay, 1973). The Frankfurt School is

associated with the University of Frankfurt which provides a professor to act as the

institute director and was itself funded by the City of Frankfurt so it could exist outside of

the conservative and, at the time, anti-Semitic, state university system (von Friedeburg,

2007).

In the years after the Russian Revolution, Germany lost its position as the home

of Marxist scholarship to the Soviet Union, although the political forces within the Soviet

Communist party tended to suppress innovation within Marxism as well as criticism of

the Soviet state. The Frankfurt School provided an opportunity for social scientists to

explore and expand Marxist theory (Jay, 1973) in a way that could not be done within the

Soviet Union. Another rationale for the creation of the Institute was to overcome the
 120

fragmentation of intellectual disciplines into separate and mutually exclusive enclaves:

the Frankfurt School sought to unify philosophy, political economy, social psychology,

psychoanalysis, and aesthetics (Horkheimer, 2003).

The evolution of critical theory cannot be explained without discussing the

context of its time: the Frankfurt School began at the same time as the National

Socialists, or Nazis, were organizing politically in Germany (Ritzer & Smart, 2001).

Many of the Institute’s members were Jewish and, despite the fact that it was still hard

for a Jewish academic to rise within the German university system and that Jews were

often denied public service positions, the Frankfurt School members felt that the

moderately socialist Weimar Republic in power at the time of the inception of the

Institute was proof that German anti-Semitism was coming to an end (Jay, 1973). Jay

points out that the Jewish Institute members had more experience of anti-Semitism, in the

form of social exclusion, while in exile in the US than they did in Weimar Germany.

Though the faculty of the Franklin School perceived that German anti-Semitism

was waning, the Nazi party, well before it took power in the early 1930s, had already

begun using systematic acts of terror against German-Jewish citizens and others they

identified as political opponents and potential political opponents; trade union leaders, for

example, were being beaten and murdered (Office of United States Chief of Counsel,

1946), Jewish academics were losing their positions, initially through a campaign of anti-

Semitism and, after the Nazis gained power, through the Nuremberg Laws of 1935 which

officially revoked the citizenship and civil rights of Jews and, subsequently, other groups

(Office of United States Chief of Counsel, 1946).

 121

In early 1933, after Hitler was appointed chancellor of Germany as part of a

supposed coalition government, German gay, lesbian, and bisexual people began to be

confined in concentration camps, as were non-Jews convicted of the ―crime‖ of having

sexual relations with Jews. These concentration camp inmates were routinely sterilized,

castrated, and killed (Kaplan, 1961). People with developmental disabilities were

sterilized as well, and, as the system of extermination camps developed, killed (United

States Office of Chief Counsel, 1946).

After the Nazi takeover of Germany in 1933, the early critical theorists, many of

whom were Jewish and all of whom were progressive and committed to social justice,

had either personal experience of Nazi repression or had family and friends directly

affected. Most of the Frankfurt School faculty members emigrated from Germany to the

US during the Nazi period, although Walter Benjamin, a critical theorist with a wide-

ranging set of interests, took his own life in 1941 in Spain while trying to immigrate to

the US. The Frankfurt School itself went into exile in 1933, initially to Geneva,

Switzerland, and then to New York, where it continued to be very much concerned with

the political and social calamity of the Nazi years.

The experience of state repression continued to be the central focus and impetus

for critical theorists (Frankfurt School members and others) in the postwar period. The

Nuremberg trials in 1945 and 1946, conducted to adjudicate war crimes after the end of

hostilities, thoroughly documented the techniques used by the Nazis to gain and maintain

power, and exposed the extent of Nazi crimes and the bureaucratic nature of their

carefully planned campaign of terror, torture, and genocide (United States Office of Chief
 122

of Counsel, 1946). Individual horror stories of concentration camp survivors continued to

emerge through the 1950s and ’60s (Yitzhak, 1987).

Jurgen Habermas, a second generation Frankfurt School member, whose theory of

communicative action will be elaborated below, had regrettably himself been a member

of the Hitler Youth and briefly served as a child soldier on the Western defenses at the

end of the war (Habermas, 1992). Habermas specifically dates his political awakening to

the immediate postwar period when he listened to the Nuremberg trials on the radio and

observed the reactions of denial and minimization among the adults in his environment

(Habermas, 1992).

Critical theorists Arendt (2006) and Adorno (1982) have addressed fascism

explicitly in their research into authoritarian tendencies, while Horkheimer (2003) and

Habermas (Borradori, 2003; Habermas, 1979, 1996) struggle with the processes of

creating an ethical democratic society with mechanisms of recognition able to preclude

the kinds of state terror they had experienced in Nazi Germany.

Early critical theorists were generally opposed to positivist social science,

preferring speculative theorizing and hermeneutic research (How, 2003). The reasons for

this tendency will be explored in the following section.

Empirical Research and Critical Theory

In the 1930s, certain Nazi academics had developed systems to identify ―non-

Aryan‖ traits based on the ―science‖ of eugenics, the study of animal husbandry applied

to human beings (Hawkins, 1997). Jewish and other peoples were classified by these

academics and the state as ―sub-humans‖ and were systematically tortured and murdered

as a ―solution‖ to a perceived social problem, that Germany’s economic and social ills
 123

were caused by the presence of Jewish and other ―non-Aryan‖ influences. Critical

theorists shied away from empirical research for fear of falling into any classification

pitfalls that would reify their research subjects and cause similar harm.

After moving to the US, however, the Frankfurt School absorbed the American

bias in favor of empirical research and produced much original research using survey

techniques (Adorno et al., 1982; Jay, 1973). Nevertheless, one of the consequences of the

critical theorists’ anti-empirical stance is the enduring scarcity of empirical research

testing some of the basic assumptions of critical theory, including Habermas’s theory of

communicative action (Brunkhorst, 2003; 2005 personal communication).

Horkheimer (2003) proposes an intersubjective research paradigm for critical

theory in order to ensure that social science plays no role in oppression. Intersubjectivity

assumes an equality of position among human beings who work together to create a

world of meaning. The idea of emancipation has been integrated into the definition of

critical theory in that it is constrained to reflect on the social world in ways that dissolve

―distorted forms of communication‖ (Brunkhorst, 2004, p. 151). Critical theory, then, in

addition to providing an explanation of the social world and a basis for predicting social

action, contains a mandate to promote social justice: in this way it shares a mandate with

social work.

Despite critical theory’s anti-empirical bias, Brunkhorst, a student of Habermas

and a prominent critical theorist, suggests that empirical research can remain consistent

with critical theory’s emancipatory mandate even when it undertakes traditional empirical

tasks such as verifying the validity of concepts and testing the salience of a theory in

experimental and quantitative studies (2003; personal communication, 2005). Brunkhorst

 124

attempts to dissolve what he sees as an unnecessary boundary between the qualitative and

quantitative research paradigms, by asserting that qualitative research is essentially a

process of categorizing social phenomenon at an order at least once removed from

traditional quantitative research. Habermas himself laments that his ideas have not been

tested in empirical settings (White, 1995).

Another objective of critical theory is to break down the barriers between

academic disciplines (Horkheimer, 1982). The central objective of critical theory is the

reordering of the social sciences into a unified project that promotes emancipation by

making visible and understandable the life experiences that obstruct personal freedoms

and reveal the possibility of new social arrangements (How, 2003).

Emancipation/Autonomy

The fascist and anti-Semitic agenda of the Nazi regime was predicated on the

classification of human beings into discrete groups have certain character traits. The

process of objectifying human beings and human activity, called ―reification‖ (Lukacs,

1971), is seen as a precursory mental process to oppression (Heath, 2001), both within

the social world and the social sciences. Marcuse (1974) defines reification as the

perception of human beings and their relationships as having the quality of an object or a

thing. The perception of a human being as a thing, or an other, who is not part of a valued

social system, allows an oppressor to mistreat a fellow human being without a

corresponding lowering of self-esteem. The typification--the idea that a certain term is

typical of a group-- of human beings as ―vermin‖ is a propaganda technique that was

used to condone and exhort genocide in Nazi Germany (Goldhagen, 1996) and more

recently in Rwanda (Gourevitch, 1999).

 125

Gelman and Heyman (1999), seemingly independently from the critical

theorists20, describe a phenomenon they call ―essentialism,‖ or categorical thinking: the

process by which people believe that certain attributes are ―rooted in the nature of‖

(Gelman, 2003, p,7), or an essential part of, the person, and that these essences explain

observable behavior. Essentialism has been proposed as an explanation of childhood

aggression. Internalized essentialist beliefs may also explain some of the ill effects of

bullying: the victim may believe that his role as a victim and the reasons that he was

chosen as a victim are essential to him and will not change over time. (As has been

pointed out previously, categorical or essentialist thinking may also be a cognitive

process involved in positive experiences, such as commitment and language

development.)

Critical theory was an attempt to develop a social theory that avoided the pitfalls

inherent in such categorizations. Categorization appears to be a key to the generation of

oppressive processes and may, when internalized, be a key to the self-oppression that

occurs when people internalize colonialist and other oppressive thought forms.

The Oxford English Dictionary defines the term ―emancipation‖ as ―setting free

from social and political restrictions‖ (2002). Max Horkheimer, a key critical theorist,

defines ―emancipation‖ as the liberation of all human beings ―from the circumstances that

enslave them‖ (Horkheimer, 1982) by a process of identifying and actively overcoming

limitation to human freedom (Bohman, 2005).

Habermas uses the term ―autonomy‖ in place of ―emancipation‖ to avoid the idea

that an interest in emancipation necessarily follows as a result of its binary opposite,

20
The bibliographies of Gelman’s book and articles show no reference to critical theorists or Lukacs.
 126

oppression (2003); autonomy, Habermas argues, is a fundamental human interest in itself.

How (2003) defines autonomy as orienting oneself towards determining one’s own life,

which is identical to the social work value of ―self-determination‖ (Freedberg, 1989;

McDermott, 1975).

The conception of autonomy will figure strongly in this writer’s interpretation of

bullying using Habermas’s theory of communicative action, in that bullying can be

considered part of a group of low or no autonomy social situations which, according to

the theory as we have adapted it, will have specific sequelae for the victim, as well as for

other bully-involved children. This writer’s project will examine the bullying literature

for the insight it offers into such components if Habermas’s critical theory.

In the next section the terms of ―postmodernism‖ and ―post-structuralism‖ will be

discussed.

Postmodernism and Post-structuralism

―Postmodernism‖ is the term used to denote a stage of history in which the

modernist promise of reason and science’s ability to ensure human freedoms no longer

holds sway (Ritzer,2002). In the 1900s, The Modernist or Enlightenment philosophy, in

which reason was to elucidate universal truths (Rasmussen, 2003) was replaced by the

idea that power relations, existing as structures within society capable of oppression,

were active in direct and indirect ways, guiding the course of science to the exclusion of

the interests of oppressed minorities or non-dominant groups. Structural social work,

which extends the ecological model to include power structures in the social world that

hinder or promote adjustment by different social groups, is an example of a postmodern

approach. Moreau (1990) attributes power to an unspecified group of the ―rich and
 127

powerful.‖ Structural social work can be seen as reductionist, analyzing a complex

phenomenon as if it were simple (Oxford English Dictionary, 2002). Post-structuralism

deepens the understanding of power dynamics.

The family of post-structural theories was highly influenced by the techniques of

textual analysis, hermeneutics, in particular, through which the discourses of power could

be understood in the lacunae, the gaps or missing portions, of text (Ritzer, 2000).

Through the work of Derrida specifically, the main character and plot lines of literature

can be seen as guided by the dominant ideology of the time which privileges some

characters and issues while rendering others invisible. By focusing off the center of the

text, called ―decentering,‖ one can consider the invisible or less visible characters. For

example, a decentering approach to the book Gone With the Wind might move the focus

onto the lives of the enslaved peoples depicted as minor characters in the original novel.

Deconstruction, another technique of post-structural literary analysis, is a process of

analyzing text for logical pitfalls, contradictions, and assumptions that may not be part of

the conscious writing process but are inherent in the dominant ideology of the writer in

his social context (Borradori, 2003).

Employing these literary techniques one considers that interwoven narratives also

construct the social world through discourse. Within US mass culture, the narratives of

the dominant culture are frequently privileged at the expense of the narratives of minority

and non-dominant groups. For example, the representation of community as a positive

force is almost totally absent in mainstream depictions of African-American life in the

popular US media.
 128

Rather than monolithic structures understood as conspiratorial, class-based cabals,

post-structuralism sees power as operating in people’s daily lives as ―multiple forms of

constraint‖ (Foucault, 1980). Habermas attempts to elucidate the conditions and interests

that create the social world. Through the method he calls ―universal pragmatics‖ he looks

for the social processes and competencies that any social group uses to create meaning.

Habermas is not always considered to be a poststructuralist because of his focus

on building theory and his belief in the power of reason; he is, however, most directly

post-structural in that he builds on the structuralist theory of Parsons (Habermas, 1987).

Habermas’s theory derives from the solution he proposes to the theoretical difficulties of

reconciling the actions of an individual actor with the action of the ―structure,‖ or

organization of individuals into institutions and collectives. Habermas, as will be

discussed in detail, proposes that structures or systems operate according to a different set

of rules from the world as it is lived, the lifeworld.

Critical Theory and Social Work

Despite the stated goal of critical theorists to integrate ideas from all the social

sciences, the profession of social work and its literature are not visible within the body of

critical theory. While social, developmental, and clinical psychology frequently inform

the work of critical theorists, as does the work of linguists, philosophers, and sociologists,

the social work profession is not presented or discussed by the major theorists writing in

or translated into English21. It is ironic that a body of work and theory that is predicated

on discovering and promoting invisible narratives has entirely ignored the narratives from

21
This writer’s thorough reading of critical theorists published in English and several theorists published in
French shows no references to social work theory or research.
 129

social work history as the social work skill set is uniquely poised to actualize the mandate

of critical theory to intervene to prevent and stop injustice.

Social work, among all the applied social sciences, has the practical skill base

with which to engage individuals, groups, and communities in reflection on the social

world. If social work has been invisible to critical theory, critical theory has become

visible to social work, particularly social workers in Canada, Australia, and Great Britain

who are embracing the tenets of critical theory and relating it to the profession. Within

the social work profession, Habermas’s theory of communicative action has been applied

to the roles and ideologies underlying the child welfare system (Blaug, 1995), as a way to

promote perspective-taking and diversity in social work education (Healy, 2001) and as a

tool of policy analysis (Reading, 1998). The theory, however, has not yet been applied to

children’s peer interaction itself or the specific problem of bullying.

The fact that Habermas’s theory has not been applied to child and family research

is surprising in light of the fact that Habermas himself stated over 30 years ago that:

―communicative action is the medium of socialization through which the influences of

familial environments are filtered and transmitted to the personality system (1974, p.

132)‖.

Habermas’s Critical Theory

In this section, both Habermas’s early and later work will be described and

applied to the problem of bullying. This description will begin with a discussion about an

early theory of Habermas’s called ―knowledge-constitutive interests.‖ Although

Habermas abandoned this project, many of its features remain extant within the

instrumental aspects of the theory of communicative action. Habermas’s technique of

 130

―universal pragmatics‖ will then be more fully described, and the final sections will

elaborate the theory of communicative action and synthesize the aspects of the theory that

will be applied to the problem of bullying.

Knowledge-Constitutive Interests

An important component of contemporary critical theory is Habermas’s construct

of ―knowledge-constitutive interests‖ (Habermas, 1971; Kondrat, 1995). In his discussion

of knowledge-constitutive interests, Habermas, assumed a rational basis for human

behavior, including conscious and less conscious processes, and suggested a direct link

between cognitive structures controlling behaviors and social skills, which he called

―interactive competencies.‖ He saw interests as powerful innate forces that guide

knowledge and, reciprocally, are guided by knowledge (Brunkhorst in Rasmussen &

Swindal, 2003) within the individual and the social group which forms the basis of action

(Habermas, 1984, 1987; How, 2003)22.

Interests constitute knowledge because they are necessary conditions for

knowledge formation and are themselves the product of social interaction (Ottman,

1982). Ottman best captures the reciprocal nature of this concept of interests by

indicating that the Latin origin of the word ―interest‖ - inter esse -- actually translates as

―to be in between.‖ Knowledge-constitutive interests can be understood as mechanisms

of attraction between a person and a desired thing or state, or the advantages or benefits

that motivate human beings to form their ideas of the world based on those benefits that

accrue to them when they act on these ideas. This knowledge can be communicated

22
Habermas’s work resembles the extensive work of Foucault although their methods differ. Habermas
surveys theory and ideas, while Foucault surveys the history of institutions. See p. 66 of this document for a
further comparison.
 131

between the environment --the lifeworld – and the personality structure itself in the form

of ―interactive‖ or ―social competence‖ (Habermas, 2001), which is labeled as ―social

skills‖ in social work. Habermas proposed a direct relationship between interactive

competence and internal behavioral controls (Habermas, 1998).

While there is a myriad of human interests, Habermas distilled all interests into

three overarching categories – material/technical, social, and emancipatory – by which all

human interests can be understood. These can be described more informally as the need

for food and shelter; the need for human companionship and communication; and the

need for physical safety and freedom of action.

Habermas’s interest in autonomy was originally framed as an interest in

emancipation. However, interest in autonomy is not necessarily predicated on having

oppressive circumstances from which one needs to be emancipated. The interest in

autonomy is virtually identical to the social work value of self-determination.

Building on the ideas of Emmanuel Kant’s Foundations of the Metaphysics of

Morals (1969), Habermas proposed that a balancing of the three universal interests would

result in an ideal social state comprised of ―autonomously acting individuals whose

decisions smoothly yield to the welfare of the whole‖ (Horkheimer, 1934, p.27).

Habermas continued this search for an ideal speech situation in his later theory of

communicative action.

Habermas’s desertion of knowledge-constitutive interests, to this writer,

represents a shift away from an exclusively material conception of the coordination of

social action. By elevating the interest in social connectedness to an intersubjective

paradigm, Habermas does not so much refute the teleological idea of action as much as
 132

subsume it into a more complex theory that assumes a source of action not exclusively

motivated by self-interest, but supplemented by an essential stratum of action

coordination based on the need to create meaning and build solidarity even in the absence

of reward23.

Overview of Habermas’s Theory of Communicative Action

Habermas’s theory of communicative action is articulated in two main volumes

and several preliminary and subsequent volumes and articles (Habermas, 1984, 1987,

2000, 2001a). In the following section, this theory will be more fully outlined with an

emphasis on those aspects that will be used to derive a theory that explains the effects of

bullying. Where possible Habermas’s own words will be used followed by an exegesis24.

In the section that follows the next one, these ideas will be directly applied to the topic of

bullying25.

23
Habermas’s theory would not be used to refute ideas of evolutionary biology but would classify these
ideas as part of the instrumentally motivated actions described in greater detail below.
24
Habermas has a dense, challenging writing style and presents the breadth of knowledge in his work. In
attempting to develop an interdisciplinary theory, he refers frequently to philosophers such as Kant,
Husserl, Hegel, and Nietzsche; sociologists and social theorists such as Marx, Parsons, Mead, and Weber;
linguists such as Searle and Chomsky, and psychologists such as Freud, Piaget, Kohlberg, and Selman. His
texts are almost completely theoretical and conceptual (Healy, 2000); however, he meticulously
summarizes each theory he uses being careful to make sure the reader understands the concepts and ideas
he is portraying.
These challenges can also be seen as strengths. Habermas rarely assumes a reader’s prior knowledge and
rewards the persevering reader with succinct summaries of the ideas of the other thinkers he discusses. His
lack of illustration allows the social work reader to apply his theory to the social work profession and such
social conditions as bullying and other forms of abuse.
25
This writer decided to separate the explanation of Habermas’s theory itself and the way that theory is
being applied to the problem of school bullying. This was done because the original theory is complex and
because the application of communicative action to the problem of bullying involves particular
interpretations of the meaning of the theory as well as the derivation of concepts not directly suggested by
Habermas.
 133

Habermas and Rationality

In his theory of communicative action, Habermas seeks a theory to explain human

processes rationally. He begins his two-volume work about his theory by explaining in

depth his views on rationality (Habermas, 1984). This extensive explanation cannot

adequately be represented here, but his project intends to increase our understanding of

human processes of communication and coordination of action. Irrationality can be

defined as anything that is not currently understood.

For this writer’s project, rather than speaking of rationality, it is more appropriate

to use the term ―visibility‖ and to look at the ability of the theory of communicative

action to make human processes visible. The idea of visibility will be revisited when

speaking of the ability of a social theory to make human processes that are currently

obscure more visible. The conditions and processes that give rise to meaning and action

are the areas that the theory of communicative action tries to make rational, including the

processes through which people acquire an ideology.

Habermas, in step with a fundamental tenet of critical theory, defines ideology as

―communication distorted by power‖ (1979) and aims to discover the tools with which

this distortion can be dissolved (Heath, 2001). According to Habermas, undistorted

communication is consensual, in that it is generally accepted and communally derived, as

opposed to being imposed by a will, even the will of a majority (Habermas, 1984, 1987).

This dissolution of distortion relates to the Marxist idea of praxis, action taken that is

free of ―externally motivated behavior‖ (Jay, 1973, p. 4). This project will show how,

using Habermas’s ideas, some of the effects of bullying can be understood as arising

from the distortion of communication.

 134

The core technique which Habermas uses, universal pragmatics, is described and

discussed in the following section

Universal Pragmatics

At the core of Habermas’s theoretical project is a process he calls ―universal

pragmatics,‖ or the search for the specific set of logically irreducible competencies and

conditions that make knowledge and social action possible--Habermas terms these the

―rules‖ of universal pragmatics (Habermas, 1979). Universal pragmatic involves the

refinement of concepts and processes that link language to the objective and subjective

reality. These rules apply equally to the verifiable nature of reality and to the subjective –

intersubjectively created – experience until they are conceptually irreducible and contain

no contradictions (Habermas, 2001c).

. Habermas’s method is to survey the ideas of Western philosophers and social

scientists including linguists, social psychologists, psychologists, and, to a lesser extent,

anthropologists, and compare these to life processes (Habermas, 1984, 1987, 2000). In

his encyclopedic survey of ideas and theories, Habermas attempts to create a social

theory that links language and the real world:

The describable reality of nature and society is formed in the interplay of

language, cognition, and action, which is governed by universal-pragmatic rules.
At the same time we ourselves produce the communicative context of the
intersubjectively experienced lifeworld through speech acts that are governed by
universal-pragmatic rules (Habermas, 2001c, p. 78).

While he does not specifically describe his project as such, Habermas tries to

build a social theory that does not rely on metaphor but is actually based on a set of rules

and conditions through which actions arise and knowledge is built. This project is similar

to Marx and Engel’s concept of dialectical materialism (Engels, 1954), but with the
 135

addition of the intersubjective, and therefore, non-material processes not anticipated by

early Marxists. Habermas examines the conditions under which communication is both

effective – the ideal speech condition – and ineffective and misleading, and how action is

coordinated. Technical accuracy is only a part of effective communication which also

includes the ability to make one’s self as well as one’s idea understood and to negotiate

meaning with one’s hearers (D. M. Rasmussen & J. Swindal, 2002). Knowledge, then, is

a combination of technical and verifiable information and information that is

intersubjectively created.

Habermas can be understood to mean that his theory can be used to understand

local human processes such as bullying when he says that ―the universal-pragmatic rule

system reveals the restrictions that the external reality of nature and society, on the one

hand, and the internal reality of the cognitive and motivational make-up of the human

organism, on the other, place on language (Habermas, 2001c, p. 78).‖ Habermas clearly

indicates that the theories derived from universal pragmatics will apply equally to the

intrapersonal experience, cognition, and motivation, and to the interpersonal nature of

society.

Communicative and Instrumental Sources of Social Action

The process of universal pragmatics has led to a number of linguistic distinctions

that represent distinctions in the real world. These distinctions have hitherto been largely

invisible. The foremost of these distinctions is Habermas’s separation of communication

from instrumental speech/acts. Habermas uses the term ―speech/acts‖ to denote the

utterances and actions intended to convey meaning and induce or coordinate action.

Within the theory of communicative action, speech/acts are divided into two categories:
 136

instrumental action and communicative action. The former includes speech/acts intended

to achieve a goal, and the latter includes speech/acts intended to achieve shared meaning

or consensus. Only communicative speech/acts meet Habermas’s standard of true

communication, which, for him, is a reciprocal process. According to Habermas, the term

―communication‖ as it is used by most people today, refers to whole classes of

speech/acts that limit, suppress, and pervert true communication and subtly or grossly

coerce others into taking actions that are often not in their own interest: as will be

discussed below, instrumental speech/acts that coerce or manipulate do not require

responses from the hearer, only compliance or non-compliance. Habermas’s theory

makes visible the fact that instrumental speech/acts give rise to little or no actual

communication.

Habermas sees consensual communication as present in the social fabric in

everyday acts of solidarity, friendship, and family life. As ordinary people engage in

daily conversation about life and the world, provide mutual support, and negotiate

common activities, they contribute to the development of conventions of shared meaning

and build the social world. This often happens through a consensual process that may not

be consciously determined or even understood by those involved. Efforts to reach

consensus are Habermas’s ideal form of communication (Habermas, 1986), and they

necessarily require the recognition of different points of view, an essential condition for

socially just communication. This mutual recognition is central to the ideal form of

effective communication (Habermas, 1987; Honneth, 1995).

 137

Habermas builds on the speech/act philosophy of the linguistic philosopher John

Austin who distinguished different forms of speech/acts as ―illocutionary‖ or

―perlocutionary‖:

Perlocutionary effects, like the results of teleological actions generally, are

intended under the description of states of affairs brought about through
intervention in the world. By contrast, illocutionary results are achieved at the
level of interpersonal relation on which participants in communication come to an
understanding with one another about something in the world. In this sense, they
are not innerworldly but extramundane. Illocutionary results appear in the
lifeworld to which the participants belong and which forms the background for
their processes of reaching understanding. They cannot be intended under the
description of causally produced effects. (Habermas, 1984, p. 293)

Habermas, summing up the difference between illocutionary, communicative

action and perlocutionary, instrumental action, uses the term states:

Through illocutionary acts the speaker performs an action in saying something…

through perlocutionary acts the speaker produces an effect upon the hearer. By
carrying out a speech act he brings about something in the world (Habermas,
1984, p. 289).

By using the construct ―states of affairs‖ to describe the effects of instrumental

action, Habermas indicates that the different forms of action coordination change the

state of the social environment. While Habermas himself does not elaborate on the

specific nature of an instrumental or communicative state as such, he does describe the

effects of instrumental and communicative action as mutually exclusive entities

throughout his work. This idea of states will be an important key to understanding the

effects of bullying in that instrumental and communicative states have unique and

particular conditions. While manipulative instrumental action can mimic features of

consensual communication, it is nevertheless an altered state: it has become

instrumentalized (See below for further elucidation on this point).

 138

Habermas describes how strategic instrumental action, the indirect manipulation

of the social world, creates a state of being ―instrumentalized.‖ An instrumental state can

at times resemble a communicative state; it will, however, contain latent effects.

Habermas acknowledges that people can be manipulated into believing they are acting

consensually but the instrumental effects remain latent in the speech/act (Habermas,

1984).

…there are countless cases of indirect understanding, where one subject gives
another something to understand through signals, indirectly gets him to form a
certain opinion or to adopt certain intentions by way of inferentially working up
perceptions of the situation; or where, on the basis of an already habitual
communicative practice of everyday life, one subject inconspicuously harnesses
another for his own purposes, that is, induces him to behave in a desired way by
manipulatively employing linguistic means and thereby instrumentalizes him for
his own success (Habermas, 1984, p. 288).

Subtle and indirect forms of bullying, such as a systematic attack on a child’s

reputation, will create an instrumentalized state; involved children who may think they

are making an autonomous choice to stop being friends with the target are, in fact, having

their own autonomy curtailed by the bully’s speech/acts. Likewise, an overt bully may

coordinate action and acquire henchmen through instilling fear among his friends that

unless they take part in the bullying process they may become a target themselves.

Habermas does not imply that the distinctions between communicative and

instrumental action are always clear in the real world. In explicating his theory, he

proposes a categorical distinction between instrumental and communicative action, an

assertion he fully supports; however, these distinctions may not always be self-evident.

While deception is a frequent occurrence and would explain how people can be

manipulated into taking actions that benefit others while weakening themselves,

Habermas goes to great lengths to explain that deception only works because there is a
 139

reasonable expectation of honest communication (Habermas, 1984). Dishonesty, then, is

rare enough in the social world that an unethical person can benefit from the expectation

of honest interaction.

Habermas argues that ―agreement can indeed be objectively obtained by force;

but what comes to pass manifestly through outside influence or the use of violence cannot

count subjectively as agreement‖ (Habermas, 1984, p. 287). Instrumental speech/acts

actually serve to limit or entirely eliminate choice, or the autonomy of individuals and

groups. That limitation of autonomy is a characteristic of the instrumental state.

Habermas defines communicative action as ―the type of interaction in which all

participants harmonize their individual plans of action with one another and thus pursue

their illocutionary aims without reservation (Habermas, 1984, p. 294).‖ The term

―communication‖ as used by Habermas is in accord with its Latin source communicare,

―to share or join with,‖ whose literal root is ―within the walls‖ – con munis.

Communicative processes account for innumerable interactions that take place free of

instrumental constraints and include, as described before, acts of meaning making,

support, solidarity, and repair, and occur in such institutions as child rearing and

friendship networks.

By separating communicative from instrumental action, Habermas does not

devalue all teleologically-induced, instrumental actions since some of these, like some

systems, are necessary tools for organizing complex tasks. Problems arise when

instrumental action is used malevolently or to meet the needs of an individual,

organization, or corporation which are at odds with the needs of the other members of the

community.
 140

The conceptualization of communication as speech/acts operating consensually

represents a paradigm shift similar to social work’s adoption of the strengths perspective

(Antonovsky; 1996; Saleebey, 1996) in which social workers are challenged to privilege

the strengths and resources of the individuals and communities with which they work and

marshal those resources to solve social problems.

Lifeworld and Systems

Habermas, using his universal pragmatics technique, makes another categorical

distinction between the ―lifeworld‖ (Lebenswelt), the world of both organic life as well as

constructed meaning, from the steering mechanisms that operate within the lifeworld with

which societies organize themselves, which he calls ―systems‖ (System) (Habermas,

1984).

By separating these phenomena of lifeworld and systems conceptually, Habermas

provides a tool that can be used to explain the ways that administrative and economic

systems work in favor of or against lifeworld needs for survival and well-being. The

lifeworld necessarily involves organic and psychological processes, while systems may

function in more linear and logical ways. Pathology arises when systems become

separated, or ―uncoupled‖ (entkoppelt), from lifeworld requirements.

Habermas sees the lifeworld as directly related to communicative action. ―Every

action oriented toward reaching understanding can be conceived as part of a cooperative

process of interpretation aiming at situation definitions that are intersubjectively

recognized‖ (Habermas, 1984, pp. 69-70). The lifeworld is the repository of culture and

language as the lifeworld ―stores the interpretive work of preceding generations‖

(Habermas, 1984, p. 70).

 141

The concept of lifeworld may be better understood in light of the French

translation of the German, which is monde vecu, or the ―world as it is lived‖ (Ferry,

1991). Lifeworld processes are often constrained by the organic nature of life, such as the

human need for survival, comfort, and social connectedness. The lifeworld operates

according to the various needs and interests of living organisms. For human beings, these

needs and interests include materials for survival and well-being, the creation of meaning,

and the maintenance of social relationships. For children and their development, these

needs and interests can be summed up as socialization. For Habermas, socialization:

secures for succeeding generations the acquisition of generalized competencies

for action and sees to it that individual life histories are in harmony with
collective forms of life (italics in the original). Interactive capacities and styles of
life are measured by the responsibility of persons (Habermas, 1987, p. 141) .

The ―generalized competencies for action‖ to which Habermas refers have direct

relevance to this writer’s project on bullying and will be more fully elaborated below.

From a social work perspective, these social competencies have a direct effect on the

individual’s quality of life, as the ability to form and maintain relationships, even through

conflict, is essential to the maintenance of marriage, the social network, and career. By

―responsibility of persons‖ Habermas means the ability to understand the truth of the

other person: this ability to take the position of the other Habermas sees as the apotheosis

of social skills.

Systems are invented and implemented by human beings within the lifeworld in

response to problems endemic to complex societies, such as the distribution of resources

where direct barter and subsistence farming is not possible. Monarchies, capitalism, and

democracies are examples of systems that organize complex societies by administering

 142

justice and ensuring some form of distribution of goods using hereditary, military,

economic, or democratic power, respectively (1987).

In describing the basis for political power, Habermas states:

The political system produces mass loyalty in both a positive and a selective
manner: positively through the prospect of making good on social-welfare
programs, selectively through excluding themes and contributions from public
discussion. (Habermas, 1987, p. 346)

As stated previously, Habermas does not favor lifeworld over systems but sees

individual and group pathology arising as a consequence of the constraints and

assumptions that occur when a system becomes uncoupled from the lifeworld, by which

Habermas means that the system is no longer responsive to or synchronized with the

lifeworld (Habermas, 1974, 1987).

This categorical distinction between lifeworld and systems solves the problem of

how and why individuals can act against their own interests and the interests of

humankind, the environment, and other parts of the lifeworld. Acting on ideology, people

are capable of undertaking roles that are not oriented to the needs of the lifeworld, rather

they can ally themselves with a system that has uncoupled from the lifeworld

(Brunkhorst, 1986). Actors taking a role within a system that is uncoupled from the

lifeworld, such as the Nazi system, will follow the system’s rules even though this may

lead to genocide or other destruction of the lifeworld.

Another example of this uncoupling process is the stripping away of local cultural

narratives in favor of mass media narratives designed to secure political compliance or

consumerism (Reading,1998), or the direct favoring of a capitalist or aristocratic class to

the detriment of the general population. Habermas also points out that when systems
 143

become uncoupled, manipulation replaces consensus as the dominant mode of interaction

(Habermas, 1987).

Habermas (1987) suggests, for example, that local systems of meaning are

stripped away when multinational corporate systems colonize the lifeworld with

narratives that promote consumerism or compliance in the population to established

political authority. This replacement of narratives results in the loss of local identities

and local wisdom that served the purpose of helping the particular social group negotiate

their environment, perhaps for centuries. Traditional stories that were meant to instruct

children may get forgotten or buried in the onslaught of television narratives that promote

the sale of superhero toys. The narratives that are prevalent on US television are ones

that at worst support aggressive social roles and at best portray successful aggressive

challenges to oppressive power (Graves, 2005).

In the following section, the building block of the consensual process, which

Habermas terms the ―validity claim,‖ will be explored.

The Validity Claim

Habermas suggests that consensual communication can be reduced to a

fundamental unit which he calls the ―validity claim,‖ the implicit or explicit claim based

on the appeal to logic, justice, and sincerity. The validity claim is the method with which

actors coordinate action among individuals without resorting to instrumental actions of

coercion or incentivization, the induction of action through the use of rewards.

Habermas describes the validity claim as the:

processes of reaching understanding (that) aim at an agreement that meets the

conditions of rationally motivated assent (Zustimmung) to the content of an
utterance. A communicatively achieved agreement has a rational basis; it cannot
be imposed by either party, whether instrumentally through intervention in the
 144

situation directly or strategically through influencing the decisions of opponents

(Habermas, 1984, p. 287).

Validity claims evoke three potential responses -- a negative response, a positive

response, or a counterproposition. An accepted validity claim results in a commitment of

some kind for some future action, if only to support the proposition. Agreement rests on

common convictions. The speech/act of one person succeeds only if the other accepts the

offer contained in it by taking (however implicitly) a ―yes‖ or ―no‖ position on a validity

claim that is in principle able to be criticized. Both ego, who raises a validity claim, with

his utterance, and alter (this writer’s italics ), who recognizes or rejects it, base their

decisions on potential grounds or reasons (Habermas, 1984, p. 287). The validity claim is

not meant to be understood necessarily as a consciously applied activity; Habermas

simply says that when we communicate a proposition, we base our claim on appeals of

various types. Through pragmatic analysis, Habermas categorizes the basis of claims as

technical, appeals based on logic; normative, appeals based on the conventions of law

and general practice; and sincerity, appeals based on an emotional connection to the

speaker. In other words, we accept a proposition from another person because we accept

their logic, we accept that the proposition fits with normal practice, and/or because we

trust the speaker.

Habermas sees the validity claim as the fundamental unit of the creation of the

lifeworld and therefore, the development of language and culture. Conventions, such as

norms and languages, develop through the action of enormous numbers of validity claims

occurring reflexively in human communication in which meaning is assigned and refined

in innumerable iterations similar to the actions of tides and currents.

 145

Applying the theory of communicative action to the initial development of

language in early human society, we can imagine a scenario in which a word becomes

associated with an object in the physical world. An early human may have made a

particular vocalization in reference to a particular edible plant, say a wild onion, in order

to share the nourishment with others. That vocalization amounted to a proposition for

others to associate that sound with that plant in the future. Such an event would be

considered a validity claim. If the person making the claim had credibility based on

expertise, perhaps as an adept provider or cook, it is more likely that the claim would be

accepted, and others would continue to use that sound to represent that plant.

As a group continues to use that sound for that plant, through successive waves of

validity claims, that sound becomes a word in a nascent language which is passed on to

the next generation or onto another group. As language develops, it is likely that

conventions also form, such as preferred cooking methods and foods that are pleasing

with wild onion. Conventions may occur as enterprising early humans found ways to

cultivate the onion, thus ensuring a steadier supply and contributing to a change from a

hunter-gathering society to an agricultural society. The names and uses of the onion come

together through the process of social interaction.

In the contemporary, postmodern period, these utterances and speech/acts may

have a multiplicity of possible meanings depending on the social context and the uses to

which knowledge is being put (Gottgtroy et al., 2003); however, the result of social

interaction, as in the example above, considered within Habermas’s framework, is a

mostly consensual process.

 146

This example is an oversimplification necessary to help readers understand

Habermas’s conceptual framework. It is entirely conceivable that the development of

language occurs in a way that mixes consensual and instrumental processes. Religion, for

example, can instrumentalize eating habits with both proscriptions against eating certain

foods and the ritualization of eating others26; commodity fetishization, the accretion of

value, of a rare food product, such as black truffles, may make it financially out of reach

of most buyers, and thereby out of reach of most lifeworld members. In this way,

consensus within the lifeworld is strategically overridden to benefit an administrative or

economic system.

Validity claims and other forms of action coordination are predicted to have

longer lasting effects within the lifeworld than action compelled through instrumental

action through an invisible, internal process Habermas calls ―illocutionary binding.‖ This

will be explained in the next section.

Illocutionary Binding

Validity claims that are accepted lead to further action through what Habermas

calls a ―binding process‖ (Bindungsenergien) (Habermas, 1984). Where instrumental

action operates through interest, communicative action operates through a set of cognitive

mechanisms that include conventions and commitments for future action. ―With the

illocutionary force of an utterance a speaker can motivate a hearer to accept the offer

contained in his speech/act and thereby to accede to a rationally motivated binding (or

bonding; Bindung) force (Habermas, 1984, p. 278).‖

26
These may have had roots in the lifeworld in that proscriptions against eating certain foods may have
arisen out of health concerns such as parasites and food spoilage.
 147

Habermas proposes that the social actions based on communicative actions

developed consensually in innumerable acts of love and solidarity are more powerful

binding agents than those that arise from instrumental actions based on the desire to attain

a specific goal. While this idea appears utopian given the volume of violence and

oppression in human history, one needs to consider that innumerable acts of healing and

creativity repair human communities during and after times of oppression and should be

considered powerful counterforces.

Acts of repair may be essential components of the natural world. Both string

theory and its more contemporary variant ―M-theory‖, in physics, contain a process by

which tears in space/time that occur in extremely minute spaces are easily repaired by a

symmetrical twin (Aspinwall et al., 1994). The complex mathematics of these theories is

well beyond the discipline of social work; nevertheless, if only by metaphor, we can

propose that repair processes are part of the fundamental nature of the universe.

Ecological and biological repair and support processes, such as the body’s innate healing

and protective abilities, are more obvious, accessible metaphors.

In the social world, Gottman and his colleagues, in a previously mentioned study,

discovered a simple mathematical relationship that successfully predicts the stability of a

marriage by the couple’s ability to generate five speech/acts of support to every one

critical or negative act(1994). While Gottman doesn’t reference Habermas, his research

can be seen as a voluminous trove of data on speech/acts within families containing 30

years of coded speech/acts and whose conclusion shows how support and solidarity are

key to family cohesion, bonding.

 148

Social history, history from the perspective of ordinary people, looks at the

actions of numerous people making change over time. E.P. Thompson (1966), a founder

of the social history movement, examines labor history from the perspective of laborers

instead of the perspective of the political elite. Thompson finds that the labor movement

began with myriad acts of sacrifice and creativity by ordinary people. Those people’s

actions are not found in the ordinary historical record. Acts of solidarity and support

undergirded the effort to feed and clothe families during strikes and to provide emotional

and physical support in the face of violent counteractions. These acts, for Habermas, are

based on a binding process.

The binding process operates through each successful validity claim which results

in a commitment to take action. The ―speaker makes an offer that he is ready to make

good insofar as it is accepted by the hearer‖ (Habermas, 1979, p. 61). When acting on

coercive power, the action may be dependent on the proximal range of the authority, the

distance after which the authority is no longer enforceable; conversely, the actions based

on validity claim remain present in the lifeworld. Habermas, then, sees communicative

actions as inherently stable.

The essential nature of the cognitive structure underlying commitment and

convention is stability. Actions based on commitment or conventions are more likely to

endure over time and may be generalized across different situations. It is through this

binding process that knowledge gets turned into action. The ideas of binding are implicit

in the work of attachment theorists who propose that much human suffering is caused by

problems in the development of nurturing relationships (Bowlby, 1999) and are also

consistent with Catalano’s related theory of social development (Catalano et al., 2004).
 149

Social Competencies, Identity, and Social Pathology

Habermas, in his one foray into intra-psychic processes, applies the theory of

communicative action to the local level of family communication published in German in

1974 and reprinted in a collection of essays on the genesis of the theory of

communicative action (D.M. Rasmussen & J. Swindal, 2002). He shows how unequal

power relations in families distort communication and leads to repression strategies

which vary in quality among differing cultures. The theme of conflicts which is

prominent in his analysis, Habermas says that ―the ego’s strength increases to the same

extent that the ego is able to do without such (repression) strategies and to process its

conflicts consciously…‖ (D.M. Rasmussen & J. Swindal, 2002, p. 136).

By conscious conflict processing, Habermas refers to the conditions of

undistorted communication. He presents a nuanced picture of how mutual understanding

is reached in a complex interplay of processes. To reach understanding involves the

ability to express oneself and one’s ideas and to take in the other and her ideas in light of

such communicative grey areas as prior knowledge, which he terms ―pre-understanding‖

misunderstandings, and lack of understanding. He also catalogues such problems with

communication obfuscation and the creation of false consensus, which he terms

―pseudoconsensus.‖

Habermas stresses that identity is the stable quality of the self. Identity is formed

within the constraints of the resources available in the lifeworld, as well as in the face of

stressful, contradictory role systems. Habermas lays out an argument for the ways that a

person is optimally self-realized in the ideal communication context, through consensual

and intersubjective processes:

 150

Corresponding to the ideal communication community is an ego-identity that

makes possible self-realization on the basis of autonomous action (italics in the
original)…In the course of the process of individualization, the individual has to
draw his identity behind the lines of the concrete lifeworld and of his character as
attached to this background. The identity of the ego can then be stabilized only
through the abstract ability to satisfy the requirements of consistency, and thereby
the conditions of recognition, in the face of incompatible role expectations and in
passing through a succession of contradictory role systems (Habermas, 1987, p.
97).

Another aspect of theory of communicative action is identity formation. Identity

is formed, according to Habermas, in a dialectical process between the child’s own

character, with its unique configuration of interests, abilities, limitations, and tastes, and

the child’s experience of both the lifeworld and the systems within it. The goal of

identity-building is to form a functional adult identity that can help the person

successfully negotiate the world. This identity needs to be capable of change:

The ego-identity of the adult proves its worth in the ability to build up new
identities from shattered or superseded identities, and to integrate them with old
identities in such a way that the fabric of one's interactions is organized into the
unity of a life history that is both unmistakable and accountable. (Habermas,
1987, 98)

Habermas introduces the idea of the formation of identity through narrative

processes in which he sees the individual taking responsibility for the formation and

adaptation of her own identity. Identity is the essential condition for recognition, by

which he means the enduring condition in which the individual remains in a stable

relationship with the lifeworld. As a relatively stable phenomenon, the identity of an

adult forms the basis through which the person interacts with the world. This idealized

self-realized ego would be able to form lasting relationships precisely because it can be

consistent as well as flexible.

 151

An ego-identity of this kind simultaneously makes possible self-determination

and self-realization…To the extent that the adult can take over and be responsible
for his own biography, he can come back to himself in the narratively preserved
traces of his own interactions. Only one who takes over his own life history can
see in it the realization of the self. Responsibility to take over one's own
biography means to get clear about who one wants to be, and from this horizon to
view the traces of one's own interactions as if they were deposited by the actions
of a responsible author, of a subject that acted on the basis of a reflective relation
to self (Habermas, 1987, pp. 98-99).

Jean-Marc Ferry builds on Habermas’s view of identity formation as a discursive

process, comparing the formation of identity with the transformative nature of narrative

through which life events acquire unique overlays of meaning (Ferry, 1991). An

observable event may appear to have a singular meaning on a concrete level, but in

reality will have innumerable potential meanings when recounted in language. This idea

of identity being formed on the basis of repeated and enduring experiences will be more

fully explored in terms of the experience of bullying victimization and the ways this can

give rise to social and intra-psychic pathology.

Habermas bases his analysis of social pathology on the ways that

―communication‖ diverges from the ideal (Habermas, 1987; Honneth, 1992). Habermas

theorizes that healthy identity is best formed under ideal speech conditions that, for

children, include free play as well as other autonomous interactions with peers and adults.

He says that ―the structures of non-alienated social intercourse (referring to

communicative action) provoke action orientations; they are aimed at filling in the spaces

for reciprocal self-realization‖ (Habermas, 1987, p. 97). In non-instrumental

communication, children can explore their interests, proclivities, tastes, skills, and

weaknesses within their own cultural and sub-cultural contexts contributing reciprocally

to the formation of identity as well as meaning.

 152

There is a very strong association between behavior and social skills, Habermas

argues. He asserts that his

point of departure is the assumption that interactive competence regulates the

construction (Aufbau27) of internal behavioral controls…interactive competence is
measured not according to the ability to solve problems of knowledge and moral
insight at the appropriate level, but according to the ability to maintain processes
of reaching mutual understanding even in conflict situations rather than breaking
off communication or merely seeming to maintain it (p 135).

The idea that communication needs to be maintained even through conflict is key

to Habermas’s view of this link between behavior and social competence. He privileges

this process of reaching mutual understanding and makes it the keystone of his project of

discourse ethics in which he proposes that the constitutional state be built based upon a

consensus among all groups on essential issues. Consensus can only be arrived at if and

when all involved parties are able to hear each other even when their positions are in

conflict. Individuals raised with the best social skills…

learn to orient themselves within a universalistic framework, that is, to act

autonomously. On the other hand, they learn to use this autonomy, which makes
them equal to every other morally acting subject, to develop themselves in their
subjectivity and singularity…Universalistic action orientations reach beyond all
existing conventions and make it possible to gain some distance from the social
roles that shape one’s background and character. (Habermas, 1987, p. 97)

Habermas (1974) proposes a schema of the development of interactive

competence that should take place in societies that have a commitment to equality for its

citizens28. At the top of the set of communicative competence is the ability to take the

perspective of the other person. This is Habermas’s key argument against the cultural

relativism that is prevalent in poststructuralist thinking. Habermas believes that children

27
The term Aufbau also translates as ―development‖ which may be more consistent with Habermas’s own
view of the process of acquiring communicative competence.
28
Habermas argues that societies that do not have universally accepted conventions for fundamental human
rights social competencies that lead to universal participation may be obstructed or suppressed. To test this
assumption would require an extensive ethnographic and anthropological study and is not germane to the
topic of bullying.
 153

will develop the advanced competencies that will allow leadership by consensus in the

absence of systemic constraints against taking the perspective of others into account. It is

this constraint against taking into account the perspective of others that is at the heart of

oppressive regimes such as Nazi Germany: genocide would not take place if all people,

including the oppressed group, were allowed to take part in the decisions that affected

their own lives.

Habermas looks at social competencies as the building blocks of a just society

with the ability to take the perspective of the other person as the highest level of

competence. While intelligence and goal-oriented behavior are keys to academic and

vocational achievement, social skills are the key to a successful social life for all

members of society. The ability to form and maintain relationships vastly improves the

quantity and quality of an individual’s life experience entirely apart from career,

education, and intelligence and will apply equally to people with significant

developmental delays as to those with significant internal and external resources.

Instrumental experiences that are decoupled from the lifeworld, which includes

victimization and abuse, can distort identity and give rise to pathologies:

This (distortion) can be seen in disturbances of the socialization process, which

are manifested in psychopathologies and corresponding phenomena of alienation.
In such cases, actors’ competencies do not suffice to maintain the intersubjectivity
of commonly defined action situations. The personality system can preserve its
identity only by means of defensive strategies that are detrimental to participating
in social interaction on a realistic basis, so that the resource of ―ego strength‖
becomes scarce (Habermas, 1987, p. 141)

In the following section the above theory will be applied and adapted to explain

the effects of bullying experiences.

 154

APPLYING AND DERIVING A THEORY

In this section, the connections between Habermas’s theory of communicative

action and the topic of school bullying will be explored. The philosopher’s concepts will

be applied directly to bullying and new concepts will be derived following through on the

logic inherent in his social theory. An important caveat must be added here in that what is

presented here is this writer’s interpretation of Habermas’s concept. This writer was

unable to communicate with Professor Habermas and is not educated in either of the two

disciplines associated with the theory of communicative action: philosophy and sociology

There has been no research in the English and French databases and search

engines directly or indirectly applying Habermas’s ideas of communicative action to

children’s development; there are researchers applying the philosopher’s ideas in

Germany (Brunkhorst, 2003 personal communication),but their results are not yet

published in English. Due to the lack of English-language research, Habermas’s ideas

will be explored conceptually and the related to the results of studies in this systematic

review.

With the exception of his essay on family communication (D.M. Rasmussen & J.

Swindal, 2002), Habermas applies his theory to the building of a just state and his

research involves constitutional law and statecraft more than individual psychology. In

this section, the connections between Habermas’s macro level theory and the topic of

school bullying within the peer context will be drawn.

The starting point for this collection of typologies is Habermas’s own (1987)

descriptive typology of social action (Figure 3), which this writer will then adapt to

bullying (Figure 4).

 155

Figure 3: Habermas’s Descriptive Typology of Social Action (Habermas, 1984, p. 333)

Social Action

Communicative Instrumental
Action Action

Concealed Open Strategic

Strategic Action Action

Conscious
Deception

Unconscious
Deception

Constructing an Explanatory Typology

In exploring Habermas’s theory of communicative action for the first time, this

writer had a sense that it could be applied directly to explain the breadth of the bullying

phenomenon, including all or most of its effects. Habermas’s ideas about the nature and

context of communication seemed to hold promise in explaining how the abuse of power

in bullying would diminish a target’s social competence and distort her cognition. In this

section, this theory will be explored for both its direct application and for the derivation

of concepts that can explain the effects of bullying. Where possible, these ideas will be

presented in the form of constructed typologies.

Latent Variables/Varying Visibility

Constructing typologies of explanation about the effects of bullying will involve

the use of latent variables, variables based on theoretical concepts, such as illocutionary

binding, that are not visible or directly measurable (Borsboom, Mellenbergh, & van
 156

Heerden, 2003; Lazarsfeld, 1937; Lazarsfeld & Henry, 1968; J. C. McKinney, 1966; J.C.

McKinney & Kerckhoff, 1962). The concept of invisibility is not monolithic. Human

processes are invisible for a number of reasons. Some invisible human processes, like

other processes in the physical world, can only be perceived through their traces; others

occur in plain sight but are ignored.

Table 4 below shows different levels of visibility of human processes. Internal

states, for example, had been seen as invisible to researchers until Likert (1932)

developed a method to measure Freudian concepts. Likert makes subjective states visible

by simply asking people to rate their self-perceptions. Still, a preverbal or verbally

inaccessible child is not able to represent herself and her anxiety and depression, so these

invisible states would need to be discerned from their visible traces: the child may stop

having fun, have somatic complaints, or wake up with nightmares. For these research

techniques that involve observable aspects of hidden phenomenon, I have borrowed the

term ―perturbative‖ from physics because it represents a term that can be applied to social

sciences. Using a particle accelerator, physicists cause a reaction among particles, and try

to learn about invisible particles through the effects of the collision on more visible

particles. Social scientists are ethically constrained not to cause harm by evoking

reactions experimentally in their subjects in order to determine effects, but bullying

researchers can be said to be examining in their subjects the perturbation that the bullying

itself has already caused.

 157

_______________________________________________________________________
Table 4: Levels of Visibility of Human Processes

Level of Ways of Increasing Examples Research

Visibility Visibility Methods
Visible Part of common Purchasing habits and Observation
knowledge as well as other reward-based
scientific inquiry behaviors

Invisible in Plain Making visible Bullying as a Observation with

Sight processes that are problem has existed the aid of a model
routinely ignored for generations but or theory
was routinely ignored
by adults.

Invisible due to Making visible Social problems such Large databases

Complexity processes that are too as poverty which and conceptual
complex and may have multiple ontologies data
multifaceted to be causes mining. Sweeping
perceived. Computers surveys as in the
are helpful in work of Foucault
organizing large and Habermas
numbers of factors

Invisible with Making visible through Cognitive and Observation or

Traces their traces. emotional states of survey of
Perturbative methods preverbal children. symptoms that
make visible processes Neurological, indicate the
that are invisible but biochemical, and presence of an
have an effect (different other physical unseen problem
types of effects) processes

Time and space are important considerations in a theory of the social world and

are often neglected by those building a theory (George & Gareth, 2000). Temporal

considerations are especially relevant to the study of children considering the changes

that occur as they develop physically, mentally, and socially over a relatively short period

of time. Bullying itself is a temporal phenomenon as it is defined in part by its repetition

or endurance over time; it also has a strong spatial effect in that it shrinks the space in

which a child feels secure (Andrews & Chen, 2006) and in which she is able to

autonomously develop. The effects of bullying are also temporal, ranging from the

immediate effects to the child to iterative--characterized by repetition--cognitive and

 158

affective processes that produce long-lasting effects on her psyche and social status.

Temporal and spatial elements will be considered in applying and adapting the theory of

communicative action to the problem of bullying.

Habermas’s Descriptive Typology

Habermas uses the technique of constructing typologies extensively in his two-

volume explanation of the theory of communicative action, although, with few

exceptions, these typologies are used to describe phenomena rather than explicate theory

(Habermas, 1984, 1987). In this section, the theory will be applied to the phenomenon of

bullying. Examples from the bullying literature and other forms of power abuse have

been added to each box to illustrate their fit with Habermas’s theory. These connections

will also be outlined in the following paragraphs as will a set of derived concepts with

which to build a theoretical understanding of the mechanisms involved when children are

bullied.

Figure 4 adapts Habermas’s typology of social action from Figure 3 by adding

examples from the bullying literature. As has been discussed above, Habermas divides

instrumental action into direct and indirect actions: actions taken by an actor directly

toward another actor (or actors) and actions taken through intermediaries, respectively.

Direct instrumental action includes a set of speech/acts that are intended to affect another

person directly, including commands and incentives on one end of the spectrum of

coercion to threats and violence at the other end. Strategic instrumental action includes a

set of speech/acts that, when used covertly, are carried out by other people without the

knowledge of the target; for example, in relational bullying, the spreading of rumors to
 159

diminish a child’s social network is both strategic and concealed: the victim may never

find out who initiated the rumor campaign.

Actions that are both open and strategic are not concealed. Neutral forms of open

strategic activities include theatre, pageantry and religious or spiritual ritual. Some forms

of bullying share features with both theater and ritual in that they take place in front of an

audience in a predetermined space and may be considered entertainment. Habermas also

considers legal and military procedures to be, at times, open, insofar as the rules of the

interaction are predetermined, and strategic insofar as the framers of the law or policy are

not present.

Figure 4: Instrumental Sources of Social Action and Types of Bullying

(Adapted from Habermas, 1987)
Social Action

Communicative Action based on Instrumental Action/Power

shared meaning (solidarity,
negotiation, and consensus)

Indirect Action Direct Action (coercion/

incentive: violence and
threat of violence)

Concealed Strategic Action

Open Strategic Action (theater,
(relational aggression,
legal procedures, overt rumor
manipulation, control of speech--
spreading, acknowledged war and
agenda, narrative, media-- ridicule,
terror tactics)
social exclusion, manipulated
terror tactics)

Conscious Deception-Pseudo-
Communicative Action (mimicking
communicative action by
manipulating others into excluding
others)

Unconscious Deception
(this results from cognitive
distortions/deficits)
 160

The Effects of Bullying

In this section, the mechanisms of effect of bullying will be examined. First will

be described the contemporaneous effects then the long-lasting developmental effects.

As has been noted review of literature above, bullying is at least partially

responsible for a variety of ill-effects. When other sources of stress and trauma in the

lives of bullying targets are controlled for, bullying still appears to be a potent force;

however, many bully-involved children have likely experienced other forms of trauma

and victimization.

Each of the following subsections ends with a list of specific effects that would be

expected to be quantified in existing or future research efforts.

Contemporaneous Geographical and Spatial Effects within the Lifeworld

Being bullied shrinks the space that is safe for the victimized child as well as

many of her peers (Andrews & Chen, 2006; Percy-Smith & Matthews, 2001): it changes

the child’s lifeworld, using Habermas’s term. Bullying may alter the way the child

experiences the bus to school, its hallways, bathrooms, locker rooms, and classrooms.

The geography of childhood is relatively circumscribed; in most cases, the young child’s

world consists of her home, a small part of the social space of her community, such as its

parks and stores, the homes of a small number of relatives and friends, and the school.

Bullying has a spatial effect within the lifeworld, further limiting the already

narrow space in which the target child lives. The target experiences a narrowing of the

safe space in her world. Depending on the nature of the child, this will have an immediate

effect on her physical organism: she may become hyper-aware of her surroundings and
 161

will likely be scared in those unsafe environments; this fear and arousal may make it hard

for her to relax and enjoy other experiences. Being continually or repeatedly afraid, she

may engage in less social and physical activity and may experience a diminishment of

attention and academic progress.

Some children may experience their entry into the victim role as a trauma and

have symptoms related to post-traumatic stress disorder, including sleeplessness, re-

experiencing the traumatic event, nightmares, and recurring physical effects, such as

increased heart rate. Bullying, however, differs from trauma in that it is by definition an

ongoing, repetitive and enduring condition. Bullying then may have two distinct

modalities of effect: an initial trauma and/or a continuous accumulation of stress. Some

percentage of victimized children will be expected to experience symptoms of trauma,

but the majority of children will be affected through a slower, more cumulative process.

Within the empirical literature on bullying, the following effects could be

measured in empirical studies:

 fear and anxiety around school

 physical symptoms associated with fear and anxiety
 diminished physical activity
 diminished attention to academic work
 sadness and depression
 sleep disturbance
 social isolation/loneliness
 symptoms of trauma

Temporal and Internal Effects of Bullying

Enduring bullying experiences in the environment may start to create ongoing and

enduring problems some of which may continue even if the bullying ceases. Finkelhor

and Hashima (1995) term these ―developmental‖ effects.

 162

A child who is experiencing anxiety and fear on an ongoing basis may start

thinking and feeling differently about her world than she did previously and compared to

non-bullied peers: these effects can be described under the rubric of cognitive and

affective distortion. For this project, these internal processes can be measured and can be

classified as temporal because due to the ongoing and cumulative nature of bullying over

time. These include temporal cognitive and affective processes, such as negative self-

esteem and self-derogation and continuing low mood or depressive symptoms.

The physical symptoms associated with fear and anxiety may start to turn into

physical complaints, and a child may have more excused and unexcused absences from

school. Within the empirical literature on bullying, one would expect to see evidence of..:

 somatic symptoms
 generalized anxiety (as opposed to fear of the school environment only)
 depression
 low self-worth, low self-esteem, high self-derogation
 externalization, such as aggression or conduct difficulties
 absenteeism from school
 lowering of academic performance
 continued and worsening patterns of social isolation and/or withdrawal

Some of the mechanisms involved when children are affected by bullying are not

as easily explained as the above variables. The mechanisms through which bullying

experiences can be expected to adversely affect a child’s identity formation and to

impede the acquisition of a full set of social competencies are not immediately apparent

in existing theory. In the next section, Habermas’s theory of communicative action will

be applied to deepen the understanding of the corrosive nature of bullying.

 163

Bullying and Communicative Action

The above sections predict a connection between experiences of bullying

victimization and both mood and cognition. Repeated and enduring stressful and

humiliating experiences will distort a child’s thinking and feeling, but a large part of the

effects of bullying victimization derives from the target’s lack of supportive and anodyne

experiences available within a rich lifeworld. The subtle and less visible mechanisms

through which the experiences of bullying victimization undermine healthy processes can

be derived from Habermas’s theory of communicative action. These latent, theorized,

variables include the deprivation of communicative states or contexts, including play;

lessening of autonomy; distortion of identity; and the distortion of cognition based on the

experience of bullying.

While adapting Habermas’s ideas for use in social work with children and

schools, it becomes clear that children overall experience a great many instrumental

actions or instrumental social conditions. The institutional arrangement of schools, as

Noguera (1995) describes it, is almost entirely comprised of instrumental actions:

children have to walk, talk, and sit when told by adults, are put into lines and ranks, and

have their behavior monitored in almost all settings. This institutional arrangement

narrows the social space available for autonomous states such as free play.

While bullying is ―instrumental‖ in its nature, it is also highly idiosyncratic and

largely ineffective. By ineffective, it’s meant that children involved in bullying are not

learning effective skills of instrumental action through the bullying in which power is

exercised in ways that positively affect the lifeworld; rather, they are learning skills that

are consistent with adult abuses of power, including coercion of innocents and
 164

humiliation in its various forms. While the effects of bullying on the bullies and

bystanders are outside the purview of this project it is clear that these non-victimized

children also experience an array of ill-effects. It is not effective for any children

involved.

Bullying would only be considered to be effective instrumental action if, as with

the Hitler Youth movement, the society deems it desirable to reproduce brutal and

ruthless adults, as Kamenetsky (1996) asserts was done by the National Socialists. In the

case of the educational policy aimed at the Hitler Youth movement, the reproduction of

adult coercive roles was effective, although not moral.

Bullying and communicative action are mutually exclusive within a single social

interchange. In general, children have relatively few opportunities to shape their own

social environment. In school, this can take place most often only during recess and lunch

breaks. When bullying, as instrumental action, enters the play environment at school, the

involved children will experience no non-instrumental environment during their school

day.29 hildren who have restrictive home and community environments may have no

social space in which they can operate autonomously.

Many of the effects of bullying can be understood to arise from the breach of

communicative contexts resulting in the child’s deprivation of many opportunities for

support and learning. This includes the deprivation of play which can be understood as

the crucible in which children learn and practice the widest range of social skills.

29
Actions that appear to have the features of bullying, such as rough-and-tumble play and face-maintaining
teasing, are still part of communicative action and may help children to develop successful strategies for
dealing with bullies.
 165

As was elucidated above, Gottman found the need for a preponderance of

supportive words and gestures to sustain a marriage; from this can be extrapolated the

importance of positive social skills as a resilience factor in relationships in general.

Play as Communicative Action

As was touched upon in the review of the theoretical literature, Children’s play

and communicative action share the common feature of speech/acts with no stated or

recognizable goal (Lever, 1978). Habermas assumes his analysis of communication

among adults to be an end state after childhood (D.M. Rasmussen & J. Swindal, 2002),

and it can be taken as given that these skills are practiced and learned in childhood. What

follows is an analysis of communicative action among children.

 166

Figure 5: Social Action Outcomes of Communicative and Instrumental Actions

Social Action Outcomes

Box 1: Communicative Action Box 5: Instrumental Action

(validity claims based on appeals to Distribution of Resources and
logic, justice, and sincerity or Responsibility and Recognition
recognition) (overcontrol, undercontrol, status)

Box 2: Social Environmental Box 6: Fixed Categories, Roles,

Effects (Social atmosphere, shared and Identities (a function of the
responsibility, informal organizational process)
distribution)

Box 3: Allows for Infinite Box 7: Allows for a Finite Set of

Responses Responses (compliance or non-
(Shared meaning in conventions, compliance. Attempts to
norms, language, repair negotiate would lead back to
processes, binding to future Communicative Action)
action)

Box 4: The Lifeworld Creation of

Instrumental Solutions
(systems, steering mechanisms
requiring instrumental action)

To illustrate the contrast between communicative and instrumental action, Figure

5 includes the nature and features of both. The first box shows the validity claim. A
 167

validity claim can be understood as any attempt to convince others to act in a particular

way when such an attempt is devoid of coercive elements.

Children try to influence their social world in myriad ways, but only among their

peers are the power dynamics sufficiently level to allow for mutual influence. Within

imaginative play, children will recreate adult roles. They learn adult skills through this

play, including convincing others to act in a certain way. When children make

suggestions and explain the logic, albeit in a childlike manner, of their suggestions these

meet the criterion for validity claims. A research agenda directly based on Habermas’s

ideas could shed light on the links between children’s pretend play and the development

of communicative competence.

The nature of the validity claim (Box 2) is that it allows a hearer the freedom to

respond in an infinite variety of ways, including making counter-suggestions, or even

simply ignoring the original actor. Children’s responses are limited only by their

developmental stage and limited life experience; however, their imaginations may make

up for these limitations.

The next step (Box 3) in the schema of social action is the development of

meaning and norms that lead to future action. Habermas’s idea of a consensual speech/act

is a speech/ act jointly formed and mutually agreed upon; it will likely have a lasting

effect, possibly enduring in some form across multiple generations. This is true in the

creation of language, particularly idiomatic language. Communicative contexts allow for

innumerable acts of support and repair.

Instrumental action itself is generated through lifeworld processes (Box 4)

through steering mechanisms intended to solve a complex problem. Box 4 is left

 168

deliberately unconnected to either the instrumental or communicative sides and indicates

an instrumental speech/act, or imperative, in whatever form it may take. The response to

an imperative, unlike a validity claim, is finite. The intention of the action may vary

greatly but the response is to obey, disobey, or pretend to obey (pseudo-consensus).

―Distribution of resources‖ is the term being used here as it can encompass such abstract

commodities as recognition and status, as well as concrete resources, such as money and

food. Roles and identities adhere to differing forms of instrumental action (Brunkhorst,

1986); the roles of bully and victim are those under discussion in the current project.

The final box on the right side (Box 8) shows the range of responses to an

established steering mechanism including a counter-response, such as a rebellion, and,

ideally, a return to communicative action, if the steering mechanism is no longer able to

meet the needs for which it was originally formed. Boxes 4 and 8 show how social action

can be understood to be circular as instrumental and communicative action function in

tandem.

In the following section, these ideas will be applied to the mechanisms of effect

specifically suggested by Habermas’s theory of communicative action.

TYPOLOGY OF EXPECTED EFFECTS

In this section, the mechanisms of effect of bullying predicted by Habermas’s

theory of communicative action will be explored. Each subsection looks at a different

type of effect and ends with a list of specific and measurable effects as well as a concept

map of the mechanism. The separate mechanisms will be related in a complete model in

the final subsection.

 169

Spatial Effects

The initial adverse effects of bullying are related to the shrinking of the physical

and social space of the target. A child who is being bullied will likely react with fear and

anxiety that is specific to the locale of the bullying. She may also, sometimes for the first

time, begin to fear people if only the bully and those children that collude with the

bullying. For the target who is bullied relationally, her social life may be suddenly

changed, and she may experience the loss of some or most of her companions. This

experience may be sufficient to catalyze a depressive episode, even a suicide attempt, and

may spark the development of an anxiety disorder that generalizes beyond the school

environment. The targeted child will almost certainly stop enjoying at least part of her

school experience and may begin to be less attentive in class: it is likely, however, that if

the bullying is stopped early enough, the contemporaneous effects will clear up quickly.

As has been discussed above, victimization may begin at any age throughout

childhood, so a target child who has experienced bullying before may develop more

resilience resources over time; however, a teenager, for whom peer relationships are of

primary importance, may experience the destruction of her rich social life as a

cataclysmic loss.

The expected contemporaneous effects of being bullied on the victim would be:

 anxiety related to the specific locations of the bullying (school, bus, bathroom,
etc...)
 physical symptoms related to stress, including somatic complaints
 symptoms of post-traumatic stress
 social withdrawal or isolation

Figure 6 below shows a concept map of the contemporaneous effects of being bullied.
 170

Figure 6: Contemporaneous Effects of Being Bullied

Breach Effects: Experiential Impoverishment or Deprivation

The term ―breach effects‖ has been chosen to describe the set of bullying effects

that is predicated upon the deprivation or impoverishment of positive experiences

inherent in communicative contexts. These breach effects include lack of opportunity for

the acquisition of communicative competence and the alienation from a myriad of

solidarity and support experiences that that entails. Such continuous deprivation

compromises the opportunities to form an identity in an autonomous peer environment.

Communicative experiences in childhood are the precursors to a healthy adult

identity and the acquisition of a wide range of social skills. Habermas stresses the role of
 171

social skills in the creation of a just state; and social skills are also seen as the key to a

rich and satisfying life for the individual. The primacy of social competence in

Habermas’s theory of communicative action is the key to understanding and predicting

the ways that bullying affects children’s development by depriving children of a major

source of such competence.

Bullying, as an instrumental state, can be expected to lessen the quantity of

communicative contexts the target experiences. The bullying affects the child’s social

world in two ways: it usually isolates the child from her peers through a systematic

undermining of the social network, and it may also change the quality of peer interactions

over time. As a result, the child will not learn the social competencies required to take

part in adult consensual processes, will not be able to take in the perspective of the other,

and will not develop an identity that fits with her individual nature. Instrumental contexts

also shape the child’s identity through the use of power resulting in an inauthentic,

distorted and weakened identity.

Not only can bullying change the cognitive and affective experience, it can

change the actual and objectively quantifiable experience of the child. The child forms a

world view based on the quality of her personal experiences. Where Burger and Luckman

(1966) propose that individuals can construct their own social world, Kokko’s idea of

cumulative continuity shows how experience of either instrumental or communicative

states can determine the actual quantity and quality of those experiences. Two children in

the same social environment, then, can have two diametrically opposed experiences of

their own childhoods and form different identities and worldviews based on those

different experiences.
 172

The damage bullying causes may be predicated on the degree to which the

victimized child is deprived of communicative social contexts. The deprivation of play –

which is the corollary to the deprivation of communicative social contexts – will have

multiple, serious, effects as described above.

A child with other social venues outside of school in which to experience consensual

interactions may well be less affected by the bullying than the child who experiences

instrumental environments in other contexts. Olweus (1980), in one of the foundational

bullying studies, showed that victimized children did tend to have more authoritarian

parenting, but that variable has not been studied in more recent effects studies.

Controlling for these multiple contexts will be important in future research to determine

whether the quantity of communicative experiences plays a role in determining the

degree of damage done by bullying.

Identity can be damaged through the limitation of communicative contexts. As

Erikson (1974) and, later, Ferry (1991) points out, identity formation is a dialectical

process requiring the freedom to test out one’s own preferences and predilections within

a complex social life. In instrumental contexts, the child does not explore a wide set of

preferences and experiences, and her identity is less likely to be adaptive to the

potentially ambiguous and conflictual aspects of the adult social world. Problems of

identity formation, as with early onset depression, are co-morbid with adult psychiatric

illness as an identity is a primary element of adaptive functioning. As will be seen in the

next section, identity can be adversely affected by bullying as well as directly affected.

Based on Habermas’s theory, one would predict that the bonding that arises from

bullying would be less effective than that arising from consensual communicative
 173

contexts. Attachments in school and with peers may be measurably weaker and would

result in less deeply felt commitments to the school and peer group.

Within the empirical literature on bullying, one would expect to see evidence of:

 lower levels of play and physical activity among victims of bullying

 lower levels of perceived autonomy of action
 a progressive decline in social skills during extended periods of bullying
or a lack of improvement in social skills over time
 lower levels of social bonding
 distorted or less-fully formed identity

Figure 7: Breach Effects of Bullying on Target

Theoretically-Derived Variables such as Identity, Distortion, and Restriction

Direct effects are those effects that directly confirm Habermas’s theory of

communicative action. It would be unlikely to find much direct confirmation of

 174

Habermas’s ideas within this systematic review as none of the studies explicitly base

their research on his theory. These direct effects include the verification of his concepts

such as illocutionary bonding and the acts of repair and solidarity that he asserts make up

communicative action. Direct effects also include the continuing exposure to maladaptive

instrumental social skills which may be directly limiting and distorting. The most

significant source of distortion most likely comes from the role and label victim itself.

The child begins to think of herself in light of that limited role and well-meaning anti-

bullying strategies may perpetuate this label by identifying children as victims and

treating them as such.

In children’s peer relationships, power may have direct effects inherent in its

nature. The children obeying a more powerful child are not exercising their own interests.

A child habitually bullied will have fewer chances to take leadership roles and practice a

variety of social skills. While there has been no published study quantifying the amount

of play engaged in by bullies and targets, it can be assumed that both the bully and victim

roles take away from the time available for free play. If bullies and targets are playing

less, then they may acquire fewer of the benefits of play, such as rehearsal and skills

acquisition. Instead, they are replaying preexisting roles. It would be an important

addition to the research into children’s play to characterize the dynamic flow of power

relations within free play situations, coding the speech/acts as either instrumental or

communicative (consensual) over time.

Instrumental states are not negative in themselves; they are necessary elements of

a complex social world. The instrumental actions that constitute bullying, however,

involve coercion, manipulation, and humiliation. These behaviors are not adaptive to
 175

success in the lifeworld and do not prepare children for a sufficiently rich variety of adult

social roles. Neither the bully nor the target learns effective instrumental leadership

strategies through the process of bullying. Bullying, then, is a distorting mechanism that

propels children away from consensus and toward coercion. Bully-involved children are

less likely to learn effective instrumental techniques, such as allocating tasks and

resources to optimum effect and issuing imperatives in respectful and logical ways.

Children involved with younger siblings and schoolmates have ample opportunity to use

effective instrumental techniques. As has been discussed above, children often act to keep

other children safe from cars, strangers, and potentially dangerous adults.

Instrumental action functions to limit choice; it attempts to create predictable and

repetitive scenarios; communicative action, on the other hand, maximizes choice. This is

an important distinction in that there are many circumstances where choice needs to be

limited and others where choice is essential. In complex and dense social environments,

such as schools, it is necessary to use power in ways that allow smooth flow of activities

that maximizes safety. These are inherent, as Noguera (1995) points out, in the power

arrangements of the mass education system.

Power in the social world, from an instrumental action perspective, is exercised

primarily to reach a specific goal, usually involving the apportioning of finite resources,

where those resources are social or material. The person in power assigns roles and

distributes resources. This is as true for complex social organizations as it is for children

exercising power in the school environment. It is understandable that children would

need to rehearse roles that they see around them and that children exposed to a surfeit of

power within their environment might try to exercise power in their own social contexts.
 176

However, for power to be exercised wisely, the person in power needs to have a great

deal of information and skill both to determine the circumstances under which power

should be exercised and the appropriate ways to apportion resources. A child who bullies

may be in the extremely uncomfortable position of trying to apportion resources without

the requisite skills set.

For many children, the victim role is stable over time. While some children have

short or multiple episodes of being bullied, most children in victim roles, particularly

those who have internalized that role, continue to accumulate victim experiences even in

discrete and unrelated social environments. A child may reflexively repeat behaviors and

signals that trigger bullies and may not have the social skills to join cohorts of children

that successfully avoid bullying interactions.

Empirical evidence to support these theorized processes would include:

 direct confirmation of theorized processes such as quantifying the amount of play,

observing attempts to influence others consensually – autonomy, validity claims
among children, direct measurement of bonding, and direct observation of role
rehearsal
 iteration of negative internal cognitive and affective experiences
 distorted identity based on the habitual role of victim
 fewer effective instrumental skills
 tendency for the victim role to become active across geographic contexts

Figure 8 below depicts a model of the active effects of bullying within the theory

of communicative action. The first box describes the iterations of bullying

experiences including iterations of negative thoughts and moods; the second box

shows the instrumental nature of bullying leading to fewer consensual social

situations leading to distortions in social bonds. The third box describes the

absorption of a victim identity including the generalization of victimization (for some

 177

children) of bullying across social and geographic contexts. The result of this

cumulative experience is a distortion of identity.

Figure 8: Active Effects of Bullying Experiences on the Target

The physical symptoms associated with fear and anxiety may start to turn into

physical complaints and a child may have more excused and unexcused absences from

school. Within the empirical literature on bullying one would expect to see evidence of:

 somatic symptoms
 generalized anxiety (as opposed to fear of just the school environment)
 depression
 low self-worth, low self-esteem, high self-derogation
 externalization such as agg
or school social work, with overt acts of violence that invoke disciplinary and juvenile

justice intervention absorbing a great deal of school social workers’ limited resources

(Allen-Meares et al., 2000).

Bullying prevention has not been part of school social works’ mandate because

prevention has not been universally accepted as consistent with the traditional mandate of

school social work of intervening only where an issue directly affects academic

functioning (Miller et al., 2003). Evidence is emerging, however, that bullying does in

fact depress academic functioning both for individual children as well as classroom
 178

groups (S. W. Twemlow, 2004; Wentzel & Asher., 1995; Woods & Wolke, 2004);

therefore, intervening in bullying should fall within the traditional school social work

mandate (Astor, 1998; Mishna, 2003).

A strong case, based on the academic and personal effects of bullying, can be

made for an expanded mandate for clinical professions to provide direct services to

children and their families within schools affected by bullying, including prevention and

primary treatment (Adelman & Taylor, 1998; Frey et al., 2005; SAMSHA, 2003). Such a

mandate, however, will require funding for social workers and increased faculty time to

implement anti-bullying and social skills building interventions.

Mishna (2003) asserts that social workers are in the best position to provide

counseling, consultation, and training for school personnel to deal specifically with

school bullying. The social work profession is already established within the host setting

of the schools and has a set of conceptual skills with which to analyze multiple social

contexts and a set of practical skills with which to intervene to improve the school

environment.

Bullying, as a set of oppressive experiences, can be seen as an issue of social

justice. Intervening against bullying is consistent with the mandate of critical social work.

Critical social work (Fook, 1993; Healy; Ife, 1997) is a dominant paradigm in this

writer’s native country Canada and shares with Critical Theory an emancipatory mandate

that requires taking direct action against oppressive forces and ameliorating the effects of

that oppression. Scandinavian anti-bullying initiatives share a similar concern in that

reducing bullying is seen as a way to increase social justice by reducing marginalization

among adult citizens (Roland, 2000).

 179

Related Issues

Alternative Views of Bullying and Victimization

Most research into bullying is predicated on the assumption that bullying is a

malignant phenomenon; however, there is some evidence that some children who make

the transition away from the victim role, called ―escaped victims‖, find that the bullying

experience strengthened their character (Dixon et al., 2004; P. K. Smith, L. Talamelli, et

al., 2004). Bullying experiences during childhood may lead to the social reproduction of

adult roles of submissiveness and domination (Dixon et al., 2004; A. Sutton et al., 1999)

which may be seen as a positive or a negative depending on one’s standpoint on roles of

social control. One study (S. Brown & Taylor, 2008) shows that adults who were

victimized as children are employed in subservient roles more frequently than those who

bullied or those who were neither bullies nor victims. Percy-Smith makes the point that

systematic abuses of power are endemic to capitalist societies which require people to

take roles with varying degrees of coercion.

In contemporary social constructions, bullying is increasingly seen as a problem.

Yet a paradox exists in that, for example, the systematic abuse of power has
throughout history characterized many processes of social interaction, for instance
in the workplace by bosses, teachers in schools or by parents in the home, and
continues to do so in some cases (Percy-Smith & Matthews, 2001, p. 51).

Kogan and Chandan (2004) in their qualitative research into school violence

found a small number of teachers and parents who expressed opinions in favor of

bullying. Most of this small group felt that bullying was an important part of childhood

and helped the children become tougher and more prepared for an adult world. One father

threatened to hit his child himself if he did not learn to stand up to his tormentor.
 180

No research studies have yet been published that examine the relationship

between bullying and adult violence and aggression30; however, the example of Nazi

tyranny may extend this notion into the reproduction of coercive and oppressive adult

roles. Kamenetsky (1996) shows how bullying was promoted and taught as a way to

reproduce adult roles of domination and even cruelty. Bullying among adults in the

workplace is another emerging topic of research (Cowie et al., 2002; Hoel et al., 1999; P.

K. Smith, 1997). However, it is not yet clear that workplace and school bullying are

related. Longitudinal studies begun in Scandinavia in the 80s and 90s are just beginning

to measure adult roles to determine correlations with childhood experiences of bullying

and victimization and their research cohort is just entering the labor market.

Bullying and Social Competence

The systematic review of bullying is intimately entwined with social competence,

another fundamental concept in Habermas’s theory of communicative action. Social

competence is a diffuse concept that comprises a wide range of behaviors. Segrin (2000),

in an attempt to capture the common feature of all social skills, defines social competence

as the ability to interact in a way that is appropriate and effective. Within the empirical

literature social competence – or social skills- are measured in several ways including the

number of reciprocated friendships, subjective measures of the quality of the child’s

interaction, the ability to negotiate, sense of humor, perceived attachment to among

others.

30
There has been a longitudinal study that has followed Norwegian children into young adulthood.
 181

Definitions and Methodology in Bullying Research

Despite the fact that bullying has been the subject of research for 30 years,

bullying researchers have not yet coordinated definitions or methodology. To date, there

have been no international conventions of leading researchers and no attempts to develop

conventions that would allow studies to build logically on one another. The result of this

lack of coordinated activity is a very confusing picture of bullying and its effects.

With governments and states implementing policy and laws against bullying

(Limber & Small, 2003) intervention programs are proliferating (Limper, 2000; McGrath,

2007; Scaglione & Scaglione, 2006); however, there is scant evidence for the

effectiveness of these programs (Rigby, 2002a, 2003; J. D. Smith et al., 2004; P. K.

Smith, D. J. Pepler, et al., 2004; Stevens et al., 2000). Three meta-analytic studies

concluded that there is no evidence to show that any bullying intervention program has

been effective and there is even some evidence that certain interventions may actually

increase bullying (D.J. Pepler et al., 1994; Vreeman & Carroll, 2007). By contrast, there

is evidence that violence, as disaggregated from bullying, can be both prevented and

reduced (Mytton et al., 2006): this leads to the conclusion that bullying is an intransigent

and far more damaging subset of aggression.

Dubin (1978) suggests that interventions in the applied sciences are unlikely to

succeed when there is no adequate theoretical explanation for the underlying

phenomenon. This appears to be the case with bullying. Many programs are created with

little or no reference to any supporting scientific literature (Stassen Berger, 2007). None

of the theoretical models currently used in the bullying literature explicitly consider the

dynamics and effects of power abuse; yet, power abuse is the single variable that
 182

distinguishes bullying from other forms of aggression. Imbalance of power clearly needs

to be a significant feature of a theory that explains the bullying phenomenon and form the

basis of effective intervention to improve outcomes for children involved in bullying.

 183

PART V SYSTEMATIC REVIEW

In this section, the systemic review will be discussed including a discussion of the

data collection and data analysis methods employed. The results will be reviewed and

related to the theory discussed above.

DATA COLLECTION

This section will elaborate the tools employed for gathering data and how these

data will advance the understanding of the effects of school bullying. The systematic

review of bullying will be a process of isolating the variables used in the existing

literature as well as the results found. Data were gathered on the measures and results on

the selected literature and as well as on the claims and theoretical assumptions being

made by the researchers who conducted the studies.

Inclusion Criteria

The following is an outline of the strategy used for identification of appropriate

studies and articles in the bullying literature. This is the American sub-set of the literature

reviewed above and includes articles from US databases. Specific information on the

databases used are identified in Appendix I. All studies that met the requirement were

included. The inclusion criteria can be summarized as follows:

 quantitative research conducted in the US

 studies published in juried journals, or books by established experts
(whose research substantially appears in juried journals)
 studies that measure the correlation between victimization and at least one
effect variable
 studies that measure bullying and its effects either contemporaneous or
within a three year period for longitudinal studies (where the data from
longitudinal studies are presented according to different time spans one
year will be the preferred span
 184

 study whose results were in a form that could be converted to a common

effect size metric either a ―Cohen’s d‖, correlation effect size; a
―Pearson’s r‖ effect size.
 studies with an ―Odds Ratio‖ effect size will be included as most studies
of suicide use this metric and is, theoretically possible to convert it to a
continuous metric such as an r or d
 studies whose sample included children and adolescents representing the
characteristics of the general population (since there is no significant
difference between economic status and ethnicity in bullying prevalence
studies whose samples came from minority communities were included)
 studies whose sample included children and adolescents of both genders

The decision to include longitudinal and cross-sectional studies was made based

on convenience in that there would not be enough studies in this review if one or the

other sets were eliminated. As well, both types of studies will capture a somewhat

different peer context: longitudinal studies will capture more of the effects of longer-term

victimization while contemporaneous studies will emphasize the immediate effects of

bullying. Since the goal of the study is to determine the range and type of effects the

contextual factors are less important. Where possible, the time comparisons will be

limited to one year or less in longitudinal studies due to the fact that children who are no

longer being victimized will begin forgetting their experience (this is the same reason that

all retrospective studies were excluded).

Exclusion Criteria

Studies will be excluded for the following reasons:

 the definitions of bullying used were out-of-date (some researchers still

conflate bullying with other forms of aggression)
 studies that measure effects without measuring bullying
 retrospective studies that measure bullying that occurred more than three
years in the past
 studies that reported only statistically significant results excluding the
results of measurements taken (in these cases an attempt was made to
obtain these data directly from the authors).
 185

A number of studies were excluded because they lacked data that would make

it possible to convert to a standard effect size. In one case the means were

presented in graphic form without either the SDs, standard deviations, or the

number of subjects in the comparison groups. (Cohen’s d or Pearson’s r can be

calculated with the mean, standard deviation and N).

Another set of excellent studies developing more complex models of the

effects of bullying were excluded because they did not provide correlation data. In

studies that present regression or path analyses the correlations could not be

extrapolated from the data presented. Several studies reported the results of latent

class analyses or structural equation modeling; such studies were included as long

as the correlation data were also included (authors of studies that lacked these

data were contacted but were not able to provide the correlation data).

DATA COLLECTION METHOD

Quantitative data were collected in a Microsoft Access database designed

specifically for this project. Textual data on the theory used in the study were tabulated in

a word processor as, likely due to the limited space in journals, very few studies related

their results to a comprehensive social theory. The database used to collect data will be

discussed below.

Quantitative Database

The following data was gathered in a Microsoft Access database specifically

designed for this project:

 the theories informing the studies as described in Table 2 (where these are made
explicit)
 the actual measures used
 186

 any established statistical validity of the measures cited

 operational definitions,
 the sample characteristics and size (N)
 the results of individual studies in whatever metric was used

Protection of Human Subjects

The project has been subject to review by the Hunter College Institutional Review

Board and data was gathered under the protocol #HC-11091453. The project did not

gather data directly from human subjects but from secondary sources and is considered to

be exempt research under category (45 CFR 46.101b):

Research involving the collection or study of existing data, documents, records,

pathological specimens, or diagnostic specimens, if these sources are publicly

available or if the information is recorded by the investigator in such a manner

that subjects cannot be identified, directly or through identifiers linked to the

subjects.

Data Management and Storage

Since the data collected are considered exempt from human subject protection

concerns due to their lack of identifying information secure storage is not an issue. No

data containing identifying information, such as survey data with identifying material,

was used in this study.

Reliability and Validity

Since the unit of observation for this meta-analysis is the individual study, issues

of reliability and validity have been dealt with by each set of authors. Reliability is the

consistency of a measurement item or instrument to produce similar results in different

contexts; validity is the ability of a measure to accurately represent a particular construct

(Yang, 2002). In several studies existing measures were used whose validity and
 187

reliability of the instrument had already been established. In other cases the authors of the

selected studies borrowed and adapted items from existing measures. In only a few cases

did the authors develop original measures for their study; and in these cases the items

were not substantially different than those items from established studies.

DATA ANALYSIS

A single question is being asked of each level of data: what are the proposed or

implied mechanisms that link the actions of bullies to the effects on bullying? This

question is asked at different levels of conceptualization from the level of theoretical and

conceptual analysis to the empirical level at which research has been performed.

In this section, the analysis techniques to be used will be described, beginning

with the analysis of the theoretical data, the qualitative data and the theory of

communicative action.

Analysis of Concepts and the Formation of a Typology

The purpose of any analysis of multiple studies, as Yang (2002) points out, " is to

synthesize and organize the existing empirical findings into a coherent pattern…and to

seek general conclusions across studies" (p. 297). This review takes all the measurements

and attempts to find the commonalities and differences in order to begin to build a model

that can explain the effects of victimization as they pertain to the dynamics and nature of

bullying.

To accomplish this using existing literature as a starting point the first order of

business is to bring some clarity to the many disparate concepts used in the selected

review. Some concepts are similar. While the statistical possibilities of this review are

limited, using the typological method (Hegar & Yungman, 1989; Lazarsfeld, 1937;
 188

Lehnert, 2007; J.C. McKinney & Kerckhoff, 1962), it is possible to determine the

consistency of concepts, variables and propositions inherent in the empirical studies with

the theoretical framework and to begin the process of determining a coherent model of

the mechanism of effect.

The different concepts used for the variables measured were classified and

categorized according to their qualities. The variables were classified was rated by four

independent raters to establish consistency and reliability. Each rater was given a list of

all the variables as they were termed by the authors themselves and were asked to put

them into logical categories. The classified variables were be turned into typological

categories based on shared features and, where feasible, consistency with the

communicative conceptual frame. For example, all the measures of self-esteem, self-

worth, self-derogation and self-derogation were categorized into a cognitive typological

category and all mood related process were put into an affective category.

Temporal considerations were added since none of the raters used spatial or

temporal considerations in their ratings. As will be seen below, fears and avoidant

behaviors that were specifically related to school were put into a spatial category and

repetitive cognitive and effective processes were put into temporal categories. These

considerations were considered to have face validity and not be open to interpretations.

These considerations will be important to the development of a theoretical model of

bullying as a life process (George & Gareth, 2000).

As was discussed in the analysis section, a plan had been put in place to use

qualitative methods and a qualitative software program to gather, code and analyze a

great deal of textual content to determine stated and implicit theory. In reality the selected
 189

studies were very thin on theory to explain bullying and negligible in explaining the

effects of bullying and could be collected almost verbatim in a few pages of prose.

In the following section the choice and rationale for the use of the vote-counting

statistical technique will be discussed.

Statistical Analysis of Results

The statistical technique that will be used in this systematic review is an informal

method of comparison called vote counting (Bangert-Drowns, 1986). In this sub-section

the rationale for that choice will be re-iterated, other statistical techniques that were

considered will be described, and the vote-counting technique explained.

Due to the variation in measures, also called the ―shared method variance‖

(Hawker & Boulton, 2000; Haynes & Hayes O'Brien, 2000) used for both victimization

and all the effects outcomes several statistical methods were eliminated. For example,

such statistical analyses as measuring the homogeneity of results, the degree to which the

results of separate studies of the same phenomenon appear to be consistent, and changing

the power of the result as according to the cumulative sample (Glass, McGaw, & Smith,

1981) are not feasible due to the lack of homogeneity of method and measures. In this

systematic review that cumulative sample size is over 90,000 so that, if traditional meta-

analytic techniques were possible, the statistical power would be increased exponentially.

Typically this kind of meta-analysis is done on studies that were conducted using the

same measures and under similar conditions making the results more easily compared.

The results of the separate studies selected for review are presented in different

effect size metrics. An effect size metric is simply the choice of a statistical metric that is

used to compare the degree and strength of an association between two variables. The
 190

selected studies depicted their results using different effect size metrics, the most

common being the Pearson’s r. Other studies compared the means between comparison

groups, used chi square comparisons or Odds Ratios (the standard technique in suicide

studies). ―Pearson’s r‖ correlation coefficient, the most commonly used metric, measures

the degree to which two continuous variables correlate; ―Cohen’s d‖ measures the degree

to which the standard deviation of two continuous variables differs; and an ―Odds Ratio‖

measures the correlation of dichotomous variable in terms of a numeric ratio, as in a

horse race. (A continuous measure allows a range of scores between whole numbers as in

scales; dichotomous effect sizes measure exclusive, either/or, categories that do not allow

scaling: in the case of suicide ultimately the dichotomous choice is between alive or dead

or an attempt to take one’s own life or no attempt.) A few selected studies presented

descriptive statistics only and were included if they presented sufficient data for

conversion to a common metric; specifically, the means, the number of subjects in each

comparison group, and the standard deviations of the scores.

In order to make it possible to compare results, each discrete result of all the

studies was converted to a common metric, Cohen’s d. An exception was made to include

studies using Odds Ratios for which data conversion is not feasible: this exception was

made so that suicide studies could be included in this review. An on-line effect size

calculator was used to convert the data into Cohen’s d (Lyons & Morris, 2010).

The ―Cohen’s d‖ effect size was chosen because this metric allows easier visual

comparison than other measures of effect size. It is a 3 decimal point metric that has a

range of plus or minus 3, representing the convention of 3 standard deviation measures

above or below a mean, By contrast, while Pearsons r varies between -1 to + 1 with one
 191

being an absolute correlation. The Odds Ratio metric is a far more unwieldy measure of

effect size; it presents effect sizes as a ratio, a confidence interval – an upper and lower

range of likelihood that a particular score will fall within that range (Hopkins, 2001). For

example, Brunstein (2008) shows that victimized girls are at higher risk for suicidal

ideation measured as 27.38/1 ratio with a lower limit of 8.93/1 and an upper limit of

83.53/1 95% of the time. This is represented as 27.38 (8.98-83.53) 95%: an unwieldy

metric for comparing several effect sizes.

Studies using Odds Ratios were included in this review in order to include suicide

in the model of effects. As is standard in suicide research all but one study measuring

suicide used the Odds Ratio metric. Odds Ratios can also be converted to r and d if the

standard deviations are known (Chinn, 2000). In fact, Cohen’s d and Pearson’s r effect

sizes can also be converted into Odds Ratios. This was not done as the Odds Ratio is such

an inefficient metric for the purpose of comparing effect sizes visually: they are not as

easily to compare as they use more space; and almost no suicide studies provided the

standard deviations necessary to convert the odds ratio to Pearsons r.

Where possible the direction of effect sizes was changed to further assist

comparison. The direction of a measure shows the presence or absence of a particular

quality in the form of a plus or minus (+ -). For example, a measure of self-derogation

was changed to a negative direction and included with studies of self-esteem and self-

worth which were determined to be essentially measuring the same cognitive process (see

the following section). In a similar way peer acceptance, a positive direction, was

changed to a negative direction in order to be included with measures of peer rejection.

 192

A visual inspection of these scores will determine the consistency of results in

terms of direction and strength, and allow for the identification of outliers, scores that are

radically different from others in a set. An unweighted mean score will be created by

averaging the scores in each dimension. An unweighted mean is calculated by adding the

scores and dividing by the number of scores; a weighted mean would take into account

the size of the sample and other considerations such as the measures used: weighting is

being avoid again due to the lack of homogeneity between studies. As will be evident

below the homogeneity and consistency of some results such as those for loneliness and

self-esteem can easily be seen using this informal method.

In a very small number of cases results that authors considered to be non-

significant were turned into significant results. This was only done when the studies

being compared showed low levels of shared method variance (where variables were

measured in similar ways) and when the particular non-significant score fell within a

range of significant scores. In all cases the smaller but significant correlation came from a

study with a larger sample than the one with higher correlation and non-significant result.

This was not done in cases where there was high shared method variance since the non-

significant result may have been measuring an unrelated phenomenon.

The original studies collected data on several different factors. Some presented

data on gender, others distinguished between scores and data collected from different

informants such as peers, teachers and parents. For this study none of these distinguishing

factors were retained as the fundamental similarity between genders and between raters

has been amply established in previous research (Hawker & Boulton, 2000; O’Connell et
 193

al., 1999) . Hawker and Boulton (2000) distinguished between informants in their meta-

analysis and found only marginal differences between results.

The only distinguishing factor of the sample that has been maintained has been

the general age of the subjects of the study. These have been grouped into two groups

designated L for early school and U for later school. (Middle school children are often

included in samples of high and elementary school student and therefore span both

categories.)

Results will be presented in three ways. The range of scores for each dimension

will be presented; the unweighted average of scores will be presented; and, for

dimensions with 2 results from at least 2 studies or 3 or more discrete results, the

unweighted mean scores for the two age cohorts will be presented. This will allow the

reader to determine whether the scores appear consistent and whether there is a change

between lower and upper school ages.

The acuity of the victimization measures was collected but did not play a

significant role in the data analysis. The vote-counting method was used to show that low

acuity studies were more likely to produce insignificant or less significant results. Future

detailed statistical exploration would need to be done to develop a system for weighting

the results to compensate for the lack of instrumental acuity but that would require a

larger number of studies where other variables were measured using the same methods.

The vote-counting method chosen for this review is sufficient for an early

exploration of the typology in that an informal analysis of the results as they are

aggregated into typological category would be able to rule out the salience of the
 194

category. Vote-counting meta-analysis can identify outliers, and determine whether all or

most findings are significant and in the same direction (Bangert-Drowns, 1986).

A table for each dimension will show the range of results, the percentage of

results deemed significant, the mean unweighted result, and, where possible, the means of

the scores for the lower and upper grades. Several sets of scores showed 100%

significance which is an indication that all studies measuring a particular variable showed

a significant correlation; the range of scores indicates the relative strengths of the

associations, and the mean shows the cumulative average of scores.

Drawbacks of the Method

The methodology has a few drawbacks. The most significant obstacle is that the

research is constrained by the data available in existing studies. Variables for which no

research has been conducted, such as the play behavior of bully-involved children or the

state of their social and emotional bonds, cannot be directly measured using this

approach. Another problem is the limitations inherent in using aggregate scores as

opposed to individual scores: since the unit of analysis in a systematic review is the

individual study rather than the individual children that are the subjects of the studies it is

possible that some results can show relationships that exist solely between studies rather

than in a human population: regression analysis to determine the relationship between

and among different variables is not possible as the individual scores are not available.

The choice of US studies results in the exclusion of several Commonwealth and

European studies both of which have longer and better-funding bullying research projects

including longitudinal studies that have followed groups of children into young
 195

adulthood. Some effects such as the physical and post-traumatic stress symptoms are

researched in greater depths outside the US.

Another drawback is publication bias which manifests as both a tendency to

accept data only from certain sanctioned sources, ignoring ―grey‖ data, data which does

not have official approval, and a tendency to under-represent non-significant results

(Rothstein, Sutton, & Borenstein, 2005). In the case of the bullying literature, there does

not appear to be a large body of grey data in that most published material on the web or

in non-scholarly books are oriented toward supporting victims rather than quantifying the

bullying phenomenon and its effects. The potential bias against non-significant results

may be mitigated by that fact that many studies present data on effects that are not

necessarily the main focus of the research. The results of this systematic review will need

to be tested with a real population before the results can be shown to be generalizable.

Another difficulty is the vulnerability that meta-analyses have to research bias

(Slavin, 1995). Slavin (1995) makes the point that virtually all study criteria and effect

size measures are incomplete and that the only way to handle bias is to be as overt as

possible about the choices made. In this case, bias is avoided by including all studies that

meet the inclusion criteria.

In the following sections the systematic review will be described and the results

analyzed. The concepts inherent in the studies will be organized and categorized using

multiple raters to see the fit with theory.

 196

Part VI Systematic Review Results

Thirty-eight research articles comprising 39 separate studies met the criteria for

inclusion in this review. Table 5 shows the authors, the variables measured (represented

as dimension, which will be more fully elaborated below), the number of subjects, and

the age group of the samples.

The total number of subjects in the review is 92,179 and the ages of the sample

span the stages of childhood but, due to the variation in ages, were coded into groups: L –

lower age, U – upper age, and A – all ages. Middle and junior high school children were

part of the younger end of studies of high school students and at the older end of studies

of elementary students; therefore, most of the studies designated with an L or U actually

included a considerable middle school cohort. A few studies limited their sample to that

middle group and were included in the older category.

Table 5: Studies Selected for Review

# Study Authors Outcomes Measured (organized by dimension) N Age
Group

1 Adelmann aggression, physical, social, substance use, suicide 50148 L

(2005)
2 Berthold & academic, aggression, anxiety, cognitive, mood, spatial 591 L
Hoover (2000)

3 Boulton play, social 89 L

(1999)

4 Brunstein mood, suicide 2342 U

Klomak
(2008)

5 Buhs & Ladd academic, behavior, social, spatial 399 L

(2001)

6 Buhs, Ladd & academic, aggression, social 380 L

Herald (2006) (table 5 continued on next page)
 197

7 Cleary (2000) suicide 1569 U

8 Cunnigham aggression, bonding, bullying behaviors, school norms 517 U

(2007)

9 Egan & Perry cognitive, externalization, internalization, physical, social 189 L

(1998) (competence)
(Table 5 continued on the next page)
10 Finkelhor, aggression, anxiety, mood 2030 All
Ormrod &
Turner (2007a)

11 Finkelhor, symptoms of traumatic stress * All

Ormrod &
Turner
(2007b)

12 Finnegan, aggression, anxiety, behavior, cognitive 184 U

Hodges &
Perry (1998)
13 Glew et al. academic, aggression, mood , social, spatial 5391 U
(2008)
14 Graham & anxiety, cognitive, social 418 U
Juvonen
(1998)

15 Grills & anxiety, cognitive 279 U

Ollendick
(2002)

16 Hanish & aggression, social 1199 L

Guerra (2000)

17 Hodges & externalization, internalization, physical, prosocial, social 173 L

Perry (1999)

18 Hodges, externalization, internalization, physical 229 L

Malone, Perry
(1997)

19 Juvonen, academic, cognitive, mood, social 243 U

Nishina &
Graham
(2000)

21 Kochenderfer- externalization, internalization, social 388 L

Ladd, &
Skinner (2002)

22 Ladd & aggression, bonding, social, spatial 396 L

Burgess (table 5 continued on next page)
(2001)
 198

23 Ladd, academic, bonding, social, spatial 200 L

Kochenderfer
& Coleman
(1997)

24 Lopez & academic, behavioral, cognitive, mood, social 508 U

DuBois (2005) (Table 5 continued on the next page)

25 Morrow, aggression, mood, social 533 L

Hubbard,
Rubin &
McAuliffe
(2008)

26 Mouttapa et al. Social 1368 U

(2004)

27 Nansel et al. Aggression 15686 A

(2003)
28 Nishina, academic, anxiety, cognitive, mood, physical, social 1526 U
Juvonen &
Witkow
(2005)

29 Nylund, mood, spatial 2307 U

Nishina, (Table 5 continued on the next page)
Bellmore &
Graham
(2007)

30 Paquette & cognitive, social 76 U

Underwood
(1999)

31 Pellegrini, aggression, dominance, mood, physical, social 154 U

Bartini &
Brooks (1999)

32 Peskin et al. Internalization 1303 U

(2007)
33 Schwartz et al. academic, cognitive, mood, social (and social competence) 199 L
(2005)

34 Schwartz et al Study 1: academic, aggression, social * L

35 (2008) Study 2: academic, aggression, social
n/a
310

36 Seals & Young cognitive, mood 454 U

(2003)

37 Srabstein et al. anxiety, mood, physical *n/a A

(2006)

38 Storch & PTSD 201 U

Esposito (table 5 continued on next page)
 199

(2003)
Totals 92,179 *

* The number of subjects is not included in this list for the following reasons: Finkelhor
2007 and 2007 b report data on the same sample; Study 1 of Schwartz 2008 appears to
use the same sample as Schwartz, 2005 as the number, mean age and gender proportion
are identical; Srabstein et al., 2006 reports data from the same group as Nansel, 2003.

Prevalence

In this systematic review the reported prevalence rates of victimization vary

between 9% and 35%. Only 15 studies presented prevalence data in the form of a

percentage; the rest of the studies presented the prevalence data in terms of the scores on

victimization scales from which the percentage could not be inferred31. Victimization

was measured using several types of techniques: the most common technique was peer

and self-nominations in which the children themselves identified those among them who

had been victimized; standard instruments were also common (B. J. Kochenderfer &

Ladd, 1997; Neary & Joseph, 1994; Olweus, 1996) as were original instruments whose

items were substantially the same as those from standardized scales. All victimization

questionnaires and the textual preambles that frequently accompany them include such

the major features of the bullying phenomenon such as a perceived imbalance of power,

direct and indirect forms, and attacks on social status.

One of the major sources of shared method variance, inconsistency in method, in

the measures of prevalence is the variation in the temporal aspects of the individual study

design. Some studies asked about the occurrence of bullying as far back as three years;

other studies used a contemporaneous criterion such as acts of bullying observed and

31
Requests were sent to several authors for the prevalence rates or the data from which those rates could be
extrapolated.
 200

videotaped. Studies that showed prevalence rates as high as 35% are likely to include a

significant number of children who successfully negotiated their encounters with bullies

and may be correspondingly less symptomatic: that cohort of former victims would likely

show lower rates of anxiety and depression. Despite the fact that repetition, the other

temporal aspect of bullying, is now a universal part of the definition of bullying, the

authors of the selected studies neglected to provide information on how and whether

repetition was measured—in other words, none of the authors specified whether and to

what degree the bullying repeated over time.

In an effort to minimize statistical error, methods for determining the status of

victim within the studies reviewed were categorized as high and low acuity. Low acuity

measures were most frequently used in large scale studies such as those conducted by

Nishina (2005) and Adelmann (2005) and were usually comprised of four or less items to

determine victimization. High acuity measures require significant researcher time and

were more common in smaller scale studies; they used elaborate surveys, in-depth

interviews, or detailed observations as well as peer, and self-nomination techniques in

which interviewers asked children to sort photographs of their peers to identify those

among them whom they perceived as being victimized.

The wide variation in prevalence of bullying among the studies does not appear to

be uniquely the result of differences in the measures employed. Studies that used similar

methods and instruments still yielded different prevalence rates among the different

samples. As has been noted by earlier researchers (Boulton, 1991; Sharp et al., 2000),

local factors appear to account for a large part of the variance in prevalence rates. Other

factors may be at play for which there is no data; for example, some school subcultures
 201

may make it less or more likely that a child would be willing to nominate a peer or

themselves as a victim. It is possible that in schools in which administrators or teachers

abuse their own power by hitting or humiliating children (Omigbodun et al., 2008) there

may be conspiracies of silence, cultural prohibitions against discussing adult abusive

behaviors, that extend to children; in such environments the prevalence rates may appear

lower.

Theories and Claims

This writer’s assumption that there was little or no theory underpinning the

studies was borne out by the literature review. However, the assumption that the claims

supporting the studies would be largely based on the victims’ risk for perpetrating future

violence was not borne out. While that impression was the result of reading dozens of

articles published in the US, many of which were indirectly or directly advocating for

resources and policy changes; researchers did not tend to make claims that went far

beyond the parameters of their own projects: the claims on which the studies were

predicated largely reflected the variables being examined and were generally concerned

with the well-being of the victimized children. Researchers did not base their claims on

the fear that the victims of bullies would become violent young adults themselves.

Only four studies provided any theoretical framework for their research; yet, none

provided a theory that explained the mechanism of effect of bullying per se with the

exception of Cunningham (2007) which used Catalano’s social development theory. This

theory resembles Habermas’s theory of communicative action in its emphasis on the

importance of social skills and social connectedness and the idea that diminishing a

child’s social world will result in problems with development.

 202

None of the authors predicated their research on the strengths perspective but

rather looked for negative sequelae of the victim experience. The embedded assumption

common to all the selected studies was that bullying was a problem requiring eradication.

None of the authors attempted to discover any positive aspects of the phenomenon nor

did they seek to elicit the opinions of adults or children who support or justify bullying.

The results of the studies, however, show little evidence for a positive effect of

victimization, although it can be assumed that there are cohorts of children among the

samples studied who were victimized but did not experience ill effects, and possibly a

cohort of children for whom escaping bullying was associated with positive effects (these

will be discussed below as possible sources of statistical error).

As will be discussed below, Habermas’s theory of communicative action predicts

that some of the worst effects of bullying would come from its ability to separate the

target child from sources of comfort, support, and autonomous growth.

The Results of the Studies

All the results were turned into the covariation metric – Cohen’s d, aside from

those presented as Odds Ratios, and all but one of the significant results were in the

expected direction. As has been discussed, the methodological variation among studies

makes it difficult to compare the effects sizes since the research design and operational

definitions are so different between studies. These results will be described below and

are shown in a set of tables.

Measurements of Instrument Acuity

The idea that lower acuity victimization instruments would result in lower levels

of significance in outcome measurements was only marginally borne out by the data.
 203

58% of results from studies using lower acuity instruments bore significant results versus

68% for studies using higher acuity instruments. This 10% difference was not sufficiently

large to justify adjusting the significance levels within this review.

The Organization of Variables

The 39 studies included in this review measured the correlations between

victimization by peers and a total of 236 separate outcomes. These measurements ranged

from single items that indicated the presence of a single discrete effect, such as the

endorsement of the item ―I feel sad most days‖, considered to be an indicator of

depression (Glew et al. 2005), to the individual score of internalization on a complex

scale such as the Achenbach Child Behavior Checklist, a multifaceted, multi-item

instrument that includes indicators of symptoms of both depression and anxiety (Hodges

& Perry, 1999). These 236 outcome variables have been divided into eight categories

based on the general topic investigated. In this section these measures and the logic of

their categorization into typological dimensions will be discussed.

To make sure that this writer was not imposing an organization onto the variables

that was influenced by his theoretical readings, independent raters were recruited. Four

independent raters were given the list of variables in the form of the actual concepts used

by the study authors. Raters were asked to put these variables into categories that

intuitively or logically fit together. While none of the raters considered temporal or

spatial features in their categories, there was almost complete agreement among the raters

about the nature of the variables32. These categories are summarized in Table 6.

32
Areas of disagreement were explained by difference of interpretation. For example, one rater put almost
all social variables into cognitive and affective categories. When asked, that rater explained that this was
because of the emotional and cognitive effects of social dimensions such as loneliness and isolation.
 204

Table 6: Typology of Variables

Latent Variable Processes Effects related to Bullying

Spatial Contemporaneous Limitation of Limitation of space, anxiety including

Autonomy, decrease in safety incipient agora- and xenophobia,
limitation (thinness) of experience

Physical/Temporal Contemporaneous Effects, Precursors Anxiety and physical symptoms

to longer term effects including sleep and eating disorders,
lack of exercise, biological
requirement for social interaction,
Play Limitation of Play as a Stand-in Limitation of communicative
Communicative Action processes including support and
unstructured thinking.

Affective-Temporal Continuing and repetitive mood Development of depression/phobias,

states development of distorted emotional
and cognitive processes, possible self-
medication

Cognitive-Temporal Repetition of negative self talk Development of distorted cognitive

processes.
A worldview colored by negative
experiences,

Social Narrowing of skills and role Narrowing of social world.

acquisition Overemphasis of instrumental.

Behavioral Distortion or production of Physical manifestations such as self-

problematic behaviors and –other harming

*Suicide Distortion of behavioral as above Self-harming

*Suicide was included as a separate category despite the fact that it fits the typology of a
behavioral effect. This was done because of two reasons: the behavior is so dangerous
that it merits its own category, and the Odds Ratio metric is unique and difficult to
compare to other metrics.

In the following sections these typological categories more fully explored.

Typologies Derived from the Review

Table 7 shows these dimensions ranked according to the variables by which they

were measure.. For example, measurements of social consequences of bullying, the most

frequently measured category, appeared as 63 separate measures in 21 of the 38 studies;

only one study directly measured play although information about play can be

extrapolated from other measures of activity and isolation.

 205

Table 7: Variables Measured within Typological Dimensions

Typological Dimensions

Social
Loneliness
Social Competence
Social Connectedness
Bonding

Play and Activity (related to the

Social but a separate dimension)

Affective and Cognitive Temporal

Depression
Anhedonia
Anxiety
Internalization

Cognitive
Self-Esteem
Self-Blame

Behavioral
Aggression
Aggressive Friends
Conduct
Dominance
Externalization
Substance Abuse
Academic
Attendance
Test Scores and GPA
(Table continues..)
Classroom Participation
School Liking
School is Hard
Academic Problems

Physical Symptoms (various)

Suicide
Spatial
Feeling Unsafe in School
School Avoidance

PTSD

The various methods used to determine the effects of bullying correspond to

various levels of visibility. For example, some of the data on play and social connections
 206

were done through observation, as were some of the measures of bullying and other

behavior measurements. Subjective states were measured largely using perturbative

techniques, the measurement of visible reactions to latent or theorized phenomena

through interviews or instruments such as the items on symptoms of depression and

anxiety. Bullying and other complex phenomena tend to be hidden from adult eyes and

are only made visible when they are labeled and become subject to general attention.

Each of the dimension in Table 7 will be discussed separately below beginning

with the spatial effects.

Spatial

Five studies measured the degree to which victimized children find their school

environment aversive. While these items measured fear and insecurity and were placed

with other measures of anxiety by all the independent raters, the decision was made to

disaggregate them from anxiety because they related specifically to the school

environment.

Table 8: Results from Spatial Dimension

Variable % Significant Range of Scores and Mean Means for Upper and Lower
Grades
Feel Unsafe at 100% 0.3504 to 0.4706 n/a no upper grade data
School 0.4347
School 100% 0.2417 to 0.4945 n/a ―
Avoidance One Odds Ratio: 2.0 (1.6 -
2.6)
0.3744

Anxiety in the face of a genuine threat is a rational reaction and a separate phenomenon

from generalized anxiety which would be present even in the absence of aversive

stimulus. The studies measured the children’s perception of safety in their school as well

as the desire to avoid the school (Berthold & Hoover, 2000; Buhs & Ladd, 2001; Buhs et
 207

al., 2006; Glew et al., 2008; B. J. Kochenderfer-Ladd & Skinner, 2002). Feeling unsafe

at school was measured using four separate but similar scales with only one measurement

using the same scale; all but one scale were measures of school adjustment and one scale

measured social anxiety and included a subscale for social anxiety. School avoidance was

measured by two separate school adjustment scales.

The results show a high degree of consistency between studies in that all studies

showed significant results in the same direction. All the studies were done at the lower

grades so no data were available on the safety of older children. As would be consistent

with the age of the children studies the feeling of lack of safety was higher than their

level of school avoidance: smaller children are not as free as older children to avoid

school. In the next section, however, it will be seen that these children also experience a

number of aversive physical symptoms that may be directly related to their fear of going

to school.

While none of the authors theorized about the spatial characteristics of the

bullying phenomenon, the seven studies provided measurements that demonstrate that

victims do in fact experience a narrowing of safe physical space at significant rates. It is

conceivable that being afraid of the school environment will turn out to be an effective

indicator of bullying, particularly for younger children who may not yet have formed a

concept of bullying and are not as able as older children to miss school. In other words,

by simply surveying children on their level of fear in school as well as perhaps, their

loneliness, educators and social workers can identify children who are being bullied or

are at risk.
 208

Physical Dimension

The two selected studies that measured physical effects looked at looked at a large

variety of effects, although with very little duplication among studies. These included:

stomach ache, dizziness, backache, headache, and obesity; as well as general morbidity

(scores on standardized scales of physical symptoms). Three instruments were used to

measure the physical effects of bullying: the HBSC, Health Behavior in School-aged

Children (WHO, 2009), the Minnesota Student Survey (Minnesota Department of

Children Families & Learning, 2001) and one original instrument; one instrument was

used to measure physical strength, a sub-scale of the Peer Nomination Inventory

(Wiggins & Wender, 1961). The HBSC includes an aggregate score of the degree of

morbidity depending on the number of different symptoms as well as the perceived

severity of the symptom.

Table 9: Results from the Physical Dimension

Variable % Significant Range of Scores and Mean Means for Upper and Lower
Grades
Physical 100% -0.4084 to 1.1858 n/a no upper grade data
Strength -0.7843
Backache single result 1.58 (1.10-2.54) ―
significant
Dizziness ― 1.46 (1.07-1.99) ―
Headache ― 1.70 (1.20 – 2.40) ―
Obesity ― 2.4047
Severe ― 3.56 (2.46-5.15) ―
Morbidity
Stomach Ache ― 1.76 (1.22-2.54) ―
Nutritional 100% *0.1001 two upper grade measures
Behavior
* This result comes from one study using a large sample which explains the significance
of this relatively mild correlation

The severe morbidity score requires special attention as it indicates that

victimized children are experiencing more than one physical symptom and perceive

themselves to be experiencing these symptoms at a more severe level than do their non-

bullied peers. The morbidity score is an aggregate score that includes the individual
 209

scores and allowed for a rating of severity. While this single score comes from one study

the sample was over 15,000 and the scale used was the World Health Organization

Health Behavior in School-children (HBSC) Check List (WHO, 2009) whose reliability

and validity have been amply established through very large cross-national studies

(Currie et al., 2008).

Obesity was based on a body mass index of the HBSC and the result of over 2.0 is

the highest result in this review. The result does not indicate the degree to which obesity

was present prior to the bullying. While most of the results of this study were published

as Odds Ratios this result was published as means, standard deviations and number of

subjects for bullied only and non-bullied comparisons group. As Hopkins (2001)

suggests the standard deviations of the two group were averaged when converting to a

Cohen’s d metric, but this did not substantially lower the covariation.

The high score for obesity compared to non-bullied and bullying groups seems to

fly in the face of research that shows obese children at equal risk for both bully and

victim roles (Janssen et al., 2004). This result clearly indicates that victimized children

have higher body mass but this result is inconsistent with the quite low association within

the same large study between poor nutritional choices and victimization: it is hard to

understand how the body mass could be so high if the nutritional choices are not terrible

unless the increase in body mass were the result of the markedly less activity that

victimized children show. This decreased activity will be shown in the play section

below.

The fact that all measures of physical strength correlated negatively with

victimization is consistent with the definition of bullying itself as an imbalance of power:

 210

a positive result in physical strength would almost certainly obviate any effects result as

the score would not likely be for a child who was genuinely victimized according to the

traditional definition.

PTSD Trauma Symptoms

Two studies looked at symptoms of trauma in younger children victimized by

bullies.

Table 10: PTSD

Variable % Significant Range of Scores and Mean Means for Upper and
Lower Grades
PTSD 100% .588 to .795 insufficient for
0.6724 comparison (only one
result was from an older
cohort)
Spatial 100% 0.2417 to 0.4945 insufficient for
0.4046 (one result was a comparison (only one
significant Odds Ratio not result was from an older
included in the mean) cohort)

Both studies measuring PTSD used the same measure the Trauma Symptom

Checklist for Children, TSCC (Briere, 1996). While all four measures show significant

results the range of scores from 0.2417 to .4945 is mild to moderate in significance.

The 100% significant rate of PTSD symptoms, all studies found significant results

for these variables, among the exclusively lower grades tested came as a surprise as a UK

study of PTSD in older victims of bullying did not show anything like that rate (Mynard,

Joseph, et al., 2000). It may be that the trauma of victimization at the hands of peers is

significantly lower in cultures that have taken universal stances against bullying as

opposed to the US culture which still largely ignores the problem; another explanation of

this difference may be that older children have been exposed to bullying both in school

and in the media thereby removing the sudden shock dimension of trauma.
 211

The fact that all measures showed a significant result does not indicate that all

children who were victimized experienced trauma. It would be important to test an older

group and to regress scores to determine co-occurring symptoms among traumatized

children over time. It would seem that a cohort of victimized children exhibit symptoms

typical of post-traumatic stress including intrusive thoughts about the traumatic event and

other symptoms such as irritability and sleeping difficulties. It is clear that many young

children experience bullying as a trauma, which is characterized by a sudden change;

another group of victimized children may experience damage due to cumulative stress

over time and may experience as somewhat different set of symptoms.

Play and Activity

Activity and play behavior were joined into one dimension since much voluntary

physical activity involves some form of play, ranging from informal or organized sports

to fantasy play with imaginative and creative interaction. Four of the measurements were

from observation and two from sub-scale drawn from the Teacher Checklist (Dodge &

Coie, 1987) and the Teacher Rating Scale of School Adjustment (Birch & Ladd, 1997).

One study, (Boulton, 1999), directly measured play behavior including the size of the

playgroup, conversation during play, and solitary play. The results show ample evidence

that victimized children play less than their peers. The size of the play network is strongly

negatively correlated with victimization; sedentary activity and playing alone is

positively correlated.
 212

Table 11: Play and Activity results

Variable % Significant Range of Scores and Means for Upper and
Mean Lower
Grades
Solitary play 2 scores 0.1250 & 0.1500 n/a no lower grade data
100%
Social Play Group 100% -1.1240 to -0.4080 ―
mean 0.647
Play conversation 2 scores 50% -0.1001 & 0.8990 n/a
Activity single non-
significant
result

The one significant and positive result that is in an unanticipated direction is play

conversation: in this observational study victimized children were found to be

significantly more involved in conversation as compared to a non-victimized group. This

finding is the single significant anomaly in this review and was recognized as an anomaly

by the author (Boulton, 1999). The item was for boys and indicated that they had been

observed in conversation at significantly high numbers which is inconsistent with the

other very high correlations within that one study for solitary play and very low social

play. Boulton suggests that further study is required to determine whether this result

indicates a phenomenon that can be generalized or was simple a local phenomenon.

Social

The social dimension is the most frequently investigated outcome of bullying and

is comprised of four sub-dimensions: social connectedness, social competence,

loneliness, and bonding. Dimensions were organized based on shared features which

were clear at face value.

Social connectedness, with 40 outcome measurements, is the most frequently

measured social sub-dimension. The Child Behavior Scale (Ladd & Profilet, 1996) was

the most frequent measure used in six discrete measurements, five measures were
 213

original to the study, two measurements were made with the Metropolitan Area Child

Study survey (Metropolitan Area Child Study Group, 2002); and two were drawn from

direct observation of peer interaction. All the variables relate in different ways to the

quantity and quality of the victimized child’s social life, albeit from different

perspectives. Measurements of social connection look at popularity through indicators

such as the number of friends, peers’ nominations of popularity, and perceptions of social

acceptance or rejection.

Measures of social distance were kept in this category but may need to be

disaggregated as a separate but related variable in future research: it is possible that some

children who are distant and withdrawn may still be considered popular; however, it

seems more likely that withdrawal and unpopularity are closely related in most cases.

Romantic appeal among a teenage sample was maintained as a separate sub-category

with the idea that unpopular children may not be perceived as romantically appealing for

reasons of their social status apart from their physical features.

Table 12: Social Results

Variable % Significant Range of Scores and Mean Means for Upper and
Lower
Grades
Loneliness 100% 0.366 to 1.757 lower grades*
mean with outliers removed .5196
.5752 upper grades*
.6031
Social Exclusion 87% 0.0801 to 2.075 lower 0.5984
mean 0.6125 (one significant upper 0.6499
OR)
Social Bonding 60% -1.0361 to -0.1202 n/a insufficient data
mean -0.4336 from upper grades
Social 80% -0.2828 to -0.7717 ―
Competence mean -0.3582
Number of 70% -0.8471 to -0.1403 (one OR) lower -0.3596
Friends mean -0.3825 upper -0.4397
Romantic Appeal two results 100% -0.5608 & -0.6992 n/a upper only
*outliers removed
 214

Loneliness is highly correlated with victimization. There were no non-significant

results in any of the several studies that measured loneliness. This is likely due to the low

method variance as many authors used the same scale (Cassidy & Asher, 1992) or items

that are very similar. Loneliness, based on the different means for the two age groups,

also shows a moderate increase over time.

Loneliness is the subjective perception of discomfort with one’s social situation

and differs from other outcome measures in the social dimension in that loneliness is a

measurement of an intrapsychic state; this requires a self-report measure of the degree of

distress felt about social isolation. Loneliness, because of its subjective nature may not

always act as an indicator of actual social isolation: some children may experience

loneliness despite a high number of friends and positive social interactions; conversely,

some withdrawn and rejected children may not experience social isolation as aversive,

preferring solitude or the company of one chosen friend to more expansive social lives.

Jobe and colleagues, however, in a study of college students (Jobe & Williams White,

2007) found that adults with autism report high levels of loneliness despite a

corresponding aversion to a high stimulus social environment.

Social exclusion also, not surprisingly, comes out high both on the percentage of

studies showing significant results and the strength of the correlations for both lower and

upper grades. These measures were changed to be uniform in direction as they measured

both peer preference and peer rejection. While there appears to be an increase between

lower and upper grades the difference is not significant: what can be said is that

victimized children are equally excluded throughout their school experience.

 215

Social bonding appears to be similarly compromised by bullying experiences.

Four studies measured some form of attachment and commitment which were

categorized as bonding, defined as a force or feeling that unites people. These included

measures of attachment, commitment, closeness, conflict and school liking drawn from

the Teacher Rating of School Adjustment instrument (Birch & Ladd, 1997) and a similar

original instrument (Cunningham, 2007) based on Catalano’s Social Development theory

(2004). The research results show that children who are victimized are also less bonded

to others. While only 60% of bonding measures were significant, the size of those effects

was quite high. This difference is very likely due to method variance and will require

further study. Teacher-student conflict was kept in the social bonding category as it was

used as a corollary of teacher-student closeness within Ladd and Burgess’s study and the

authors proposed it as an indicator of bonding (Ladd & Burgess, 2001): the authors take a

nuanced view of conflict and propose that the conflict can be interactional in that it can

be the result of social distance and not always or exclusively the cause of social

distance33. Further study would be required to determine whether and to what degree

conflict is closely correlated with bonding or an unrelated phenomenon.

Five studies looked at social competence or its negative – social ineffectiveness

and interpersonal problems. Three scales: the Social Behavior Rating Scale (Schwartz,

Farver, Chang, & Lee-Shim, 2002), the Self-Perception Profile for Children (Harter,

1999) and the Self-Report Coping Scale (Causey & Dubow, 1992) were used in all 14

measurements giving the results a fair degree of method consistency.

33
Ladd and Burgess (2001) suggest that conflict arise in the vacuum that is created when there is social
distance.
 216

Measuring social competence in children of different ages is difficult as social

competence is not a static phenomenon, changing over developmental stages. Social

competence, like intelligence, develops throughout childhood: a child entering

Kindergarten with low social skills is at risk for victimization (B.J. Kochenderfer &

Ladd, 1996) but a child with adequate social skills who is targeted by bullies may have

fewer opportunities to learn the more advanced social skills in a non-coercive

environment and may quickly fall behind in her social development.

Several measures looked at the numbers of friendships or the size of the social

group of victimized children, the vast majority using similar peer nomination methods in

which children identify peers and rate or describe their friendships. The aggregated

results show that victimized children tend to be more socially isolated at all ages. The

studies measured the size of friendship groups as well as the number of best friends.

While there was a 30% rate of non-significant result there was no result in the contrary

direction: no study showed that victimized children had the same or higher numbers of

friends compared to non-victimized children. The mean for the upper grades was

somewhat higher indicating that social exclusion may worsen over time.

The self-perception of romantic appeal among older grades was also significant

with a high negative correlation with victimization. This is taken from a sub-scale of the

Social Experience Questionnaire (Crick & Grotpeter, 1996). This negative perception

process indicates hopelessness for future social relationships and should be checked to

see if it correlates with depression and suicide in a larger sample. While the victimized

sample is shown to be physically less strong than its non-victimized cohort many of these

children may possess physically attractive features. It is interesting that the two measures,
 217

which were separated by gender, are substantially the same because it would be expected

that the lack of physical strength would be an asset at best and neutral at least for girls

and a negative for boys.

The picture that the data provides shows a cohort of lonely and isolated children.

While social competence and the number of friends can be related, a child may be

isolated even if she were to possess measurably high social competence. In fact, some

popular children are selected as the target of relational aggression precisely because they

have a rich social life. Relational aggression, including cyberbullying, often very

effectively damages the reputation and social status of a child with a rich and varied

social network.

The next section looks at the affective results for these isolated children.

Cognitive Temporal

The cognitive sub-dimension includes all measures of any form of self-appraisal:

self-esteem, self-worth, and self-efficacy. While these are very separate concepts, they all

share the quality of cognitive self-appraisal that is variable in negative and positive

directions. The directions of measures were made uniform and the dimension was divided

into two main categories self-worth and self-blame. A single measurement of self-

efficacy for aggression did not fit either category and was maintained in the cognitive

dimension, rather than moved to the behavioral dimension, as it is a subjective

perception.

Table 13: Results in the Cognitive/Temporal Dimension

Variable % Sig- Range of Scores and Mean Means for Upper and Lower
nificant Grades
Self-worth 81% -1.8581 to 0.3450 Lower 0.125
Upper - 0.6141
Self-blame 57% - 0.2828 to 1.065 n/a
0.2902 (upper grades only)
 218

Variable % Sig- Range of Scores and Mean Means for Upper and Lower
nificant Grades
Self-efficacy one non- one measure for the lower grades
for aggression significant

Self-esteem was measured using three separate instruments although Harter’s

Self-Perception Profile for Children (1996) was used in 11 discrete measurements with a

sub-scale of the Child Depression Inventory being the next most frequently used with

three. The items in all three scales are similar giving this dimension a very low level of

method variance. Low self-worth is particularly significant for older children as there is

over six-fold decrease in self-worth between upper and lower grades. This indicates

clearly that children who remain in the victim role will have their self-esteem

systematically undermined through their bullying experiences, and, perhaps, that

teenagers who experience victimization for the first time in the older grades will

experience a fast undermining of their self-esteem. Oddly, self-blame does not appear to

be correlated to negative self-esteem.

Self-blame and self-derogation, which was folded into the self-blame sub-

dimension, were both seen as essentially the same phenomenon and were measured by

two separate instruments: an original scale accounts for 8 of the measurements and the

Social Experience Questionnaire (Crick & Grotpeter, 1996) was used once. Self-blame

was only measured in the upper grades and did not show as strongly significant as other

measures. This indicates that many victimized children are not ascribing blame to

themselves for their bullying even though the bullying is undermining their self-esteem.

Unfortunately, the meta-analytic method does not allow scores to be regressed to

determine whether the cohort of self-blaming children suffer the same or, as seems likely,

more severe symptoms than their non-self-blaming peers. It would be important to

 219

measure self-blame in the earlier grades to see if that corresponds to the relatively low

rates in older children.

Self-efficacy for aggression was included in the cognitive category as it is a

cognitive assessment indicating that the child does not rate herself capable of acting out

aggressively. The one result was for a study of a lower grade cohort and did not show a

significant result in either direction. That study cohort, however, did not consider itself

capable of acting aggressively which is not surprising given the powerlessness inherent in

the bullying experience.

Affective Temporal

The affective temporal category includes all measures of emotional processes.

These measures used the children themselves, their parents, or peers as informants and

asked about the victimized child’s subjective emotional state. These include items on

specific moods, anxiety, and more global measures of anxiety and depression.

Internalization is a general term for emotionally aversive states that include and conflate

anger, depression and anxiety, the inwardly directed symptoms. Table 14 shows the

ranges, means and proportion of significant results.

 220

Table 14: Results in the Affective/Temporal Dimension

Variable % Significant Range of Scores and Mean Means for Upper and Lower
Grades
Depression 100% 0.3871 to 1.1890 Lower 0.4682
0.6988 Upper 0.7744
Anhedonia* 100% 0.2622 to 0.4296 n/a (only 2 items from upper level
0.3602 samples, both were Odds Rations)
Anxiety 70% 0.1001 to 0.7473 Lower 0.3505
0.3788 (one OR significant) Upper 0.3029
Emotional 1 significant 1.0078 n/a
Problems result
Emotional 1 significant 0.4296 n/a
Lability result
Internal- 87% 0.345 to 1.8051 n/a scores for lower grades only
ization 1.2501 (two OR results)
Irritability 1 result in- n/a
significant
OR
* Anhedonia refers to all items that indicate a low mood or lack of enjoyment

Depression and anxiety appear to be significant problems for victimized children.

Depression is measured primarily using established scales: The Children’s Depression

(Kovacs, 1992) inventory was used 14 times with the Beck scale (Beck & Steer, 1993),

the Child Behavior Check List (Achenbach, 1991), the Trauma Symptoms in Children

Checklist (Briere, 1996) and the Health Behavior in Schools Checklist (WHO, 2009) and

original items used in one to three of the measurements. One hundred percent of the

measures of depression on scales as well as individual items under the rubric of

anhedonia - low mood and unhappiness - were significant. A comparison of the means

for depression scores in younger and older children show a 60% increase over time: by

far the largest increase between age cohorts in any measurement within this review.

While some part of this variance may be due to the difference in measures and capacity

of children to accurately report their mood, it would be very unlikely that this explains a

60% increase. Coupled with a slight decrease in anxiety (.3505 to .3029) this seems to

bear out Parker and Ashgari’s (1997) prediction that anxiety in young children turns into

early onset depression, which they show is co-morbid with other life difficulties.
 221

One single item on emotional problems in general was separated into its own sub-

category and showed a high degree of significance. Emotional lability was also unique

and also showed a significant positive result. A single item on irritability did not show a

significant correlation to bullying, but this was in one study.

Anxiety was quite common for victimized children although the means remain

relatively steady over time. Anxiety was also measured using similar scales as for

depression (HBSC, TSCC and the MASC Multidimensional Anxiety Scale – Children

(March, Parker, Sullivan, Stallings, & Conners, 1997). Given the rise in depression over

time, it seems likely that one group of victimized children develops generalized anxiety

symptoms while a larger group becomes increasingly depressed over time. Some, of

course, will develop both anxiety and depression.

Internalization is the aggregate score on complex scales such as the Achenbach,

the Peer Nomination Inventory (Wiggins & Wender, 1961), and the Strengths and

Difficulties Questionnaire (Goodman, 2001). Internalization conflates depression and

anxiety as well as cognitive problems and was highly significant with an average - of

1.25. Measures of internalization were given only to lower grade children and showed an

87% rate of significant result.

Academic

The academic dimension includes such typical academic measures as grades and

test scores, and teacher, self, and peer ratings of academic achievement; measures of

attendance and tardiness are also included in this dimension as are perceptions of school.

Schwartz (2008) reported on both math and reading and found that victimized children

scored significantly lower on math but did score in a positive direction, although not
 222

significantly, on reading. Two separate measures of participation are represented.

Autonomous participation refers to a sub-scale that measures the independent and self-

directed activity: this measure, while in a negative direction, was not at a significant

level. Class participation, however, was strongly significant in a negative direction

indicating that victimized children tend toward less class participation

Table 15: Results from the Academic Dimension

Variable % Significant Range of Scores and Means for Upper and Lower
Mean Grades
Marks and scores 85% -0.1201 to – 0.6521 Lower -0.28105
-0.2990 Upper - 0.4734
Learning Significant 0.3242 n/a
problem
Attendance 33% -0.02 to -.3242 n/a no upper grade data
-0.1120
School is hard Significant 0.3914 single n/a
Participation 67% -0.3660 to -0.1001 n/a no lower grades data
-0.2021

Cooperative part -0.1403 single n/a

School liking Significant -0.1807 single n/a

Marks and test scores for victimized children were significant in 85% of the

results at moderate levels for younger grades and significantly higher levels for older

grades indicating the tendency for victimized children’s academics to slide over time.

Learning problems was kept as a separate measure since it is based on subjective

perception: this single result is also significant and close to other scores within the range.

Attendance was significantly lower but this measure is not that sensitive in the lower

grades as children don’t have as much freedom to skip classes or play truant. Most of

these absences can be explained by the physical symptoms victimized children

experience (discussed above).

 223

Victimized children are seen, at older grades, to like school less, to find school

hard and to participate less in class. Cooperative participation was not folded into general

participation because the authors made the distinction and because this item retains

features of social competence: it is possible for a child to participate academically while

not cooperating socially.

The results of studies indicate a pervasive dampening of academic achievement

and a decrease in comfort and enjoyment of the academic setting. Finding school hard is

the highest co-variation in this dimension which is intuitively logical in that this would

also capture the victimized children who are academically successful. Victimized

children’s academic performance either continues a downward spiral in older grades or is

a more sensitive measure for the upper grades. This may be due to the increasing

workload and complexity of material in the older grades. Measures of perception indicate

that victimized children find school less rewarding, less pleasant, and experience more

problems learning. Two

The positive direction for reading scores among victimized children is only one of

two anomalies found in this review. While not significant it nevertheless was in an

unexpected direction. It would be interesting to further study the reading of victimized

children as well as socially isolated children in general. It can be hypothesized that some

low play children spend more time reading, perhaps as a way to retain a semblance of

social contact.

Behavioral

The behavioral dimension is comprised of a number of separate indicators all of

which directly involve a measureable outward behavior. These include measures of

 224

several types of aggression as well as measures of externalization, a global measure of

behavior problems. Included are measures of delinquent behavior, anger as rated by peers

and teachers, and violent and delinquent behaviors including carrying weapons and

fighting.

Table 16: Behavioral Results

Variable % Significant Range of Scores and Mean Means for Upper and Lower
Grades
Aggression 62% -0.1001 to 1.0361 n/a only one result from upper
mean 0.4493 grades
Aggressive 60% 0.02 to 1.093 Lower 0.1928
friends 0.3728 Upper n/a
Conduct 50% 0.3115 to 1.0078 n/a
0.4971
Externalization 100% 0.5608 to 0.9256 No upper grade data
0.8437
Substance Use 100% 0.1202 to .899 No lower grade data
mean 0.5130
Dominance one non-
significant

Results within this review dimension are less homogeneous than other

dimensions. Aggression itself was measured using 7 separate scales and one peer

nomination protocol and scores range from -0.1 to over +1.0 which is over one standard

deviation difference. All but one result was from the upper grades. This wide range is

likely explained by the fact that some studies separate out bully/victims while others do

not: this would result in higher aggression levels in those that include bully/victims, a

separate subset of aggressive victims34. Another likely explanation is that some habitual

victims tend to be targeted precisely because of their emotional lability. One cohort of

victimized children, therefore, may be quite compliant and non-aggressive while another

may be aggressive.

34
It is this writer’s contention that self-report of victimization among bullies can be an exculpatory strategy
or an example of hostile attribution, the attribution of hostile motive to neutral or benign actions.
 225

Endorsement of aggression and violent behavior – approval of such behavior –

was included in this category, but may function separately if it were found in future

research that some children endorse aggressive acts without actually performing them

themselves.

While the data shows that victimized children are not as physically strong as their

cohort, after the first years in each educational institution --elementary, middle (and

junior high), and high school– they are likely to be more powerful than the younger

cohort and therefore do have access to a cohort that they could themselves bully.

Externalization, the global scalar measure for all externally visible problematic

behaviors was measured with the Child Behavior Checklist (Achenbach, 1991) and the

Peer Nomination Inventory (Wiggins & Wender, 1961) is the highest average score in

behavioral dimension at 0.8437. Conduct problems, while only significant in half the

results is the second highest co-variation in this dimension at 0.4971. Only one study

(Adelmann, 2005) measured the use of substances.

Adelmann’s study, with over 50,000 middle school children, it is the largest

included in this review. The victimization measure used in the Adelmann study is

considered to be low acuity and the prevalence rate of 35% is the highest in this review.

Substance use was very high among victimized teens with measures of smoking, alcohol

and drug use folded into one sub-dimension and having a 100% significance and an

average co-variation of 0.5130.

The data in this dimension were not collected uniformly in upper and lower

grades and with high shared method variance to the extent that it would be difficult to
 226

draw conclusions: further research should be done on a wide sample using the same

method and controlling for bully/victims.

Despite the inconsistencies, the results of the behavioral dimension appear to

refute the commonly held perception that victims tend to be compliant and non-violent.

Rather, it would appear that they lack the physical resources to defend themselves, and,

perhaps, the belief that their efforts at self-defense could be effective.

Suicide

The studies of suicidal behavior are not sufficiently homogeneous to merit

comparison in the way that the other studies. The methods differ and the statistical

technique of Odds Ratio is significantly different than the continuous measures discussed

above. Nevertheless, this section indicates that the risk of suicide is a significant effect of

victimization and, therefore, a significant public health concern.

Table 17: Suicide Study Results

Variable % Individual Scores and Means for Upper and
Significant Mean Lower
Grades
Suicidal Behaviors 100% boys 1.69 (1.20-2.31) Upper Only
girls 2.54 (1.20-2.31)
correlation 0.3034
Suicide Attempt (2 results) 100% 18.23 (5.11 – 65.10) boys ―
19.40 (6.52- 57.59)
Suicidal Ideation (2 results) 100% 19.65 (6.85 – 56.34) boys ―
27.38 (8.98- 83.53) girls
―Others would be better off without me‖ not correlations 0.294 ―
single result significant
Violent and Suicidal Behavior 100% 2.61 (1.78 – 3.69) boys ―
2.61 (1.66-3.90) girls

The shared method variance within this dimension and between the other

dimensions in this review poses difficulties in comparing and contrasting results. As had

been pointed out above, all but one suicide study was done using Odds Ratio effect sizes

and there is an apparent lack of homogeneity in scores for similar phenomena within the

suicide dimension: scores range from a low of 1.69/1 for suicidal behaviors to 27.38 for
 227

suicidal ideation. The results for suicide attempt and ideation are ten times more

significant than the result for suicidal behaviors which is too high a shared method

variance to conflate. It is not immediately evident why the results would be so different.

Suicide attempt was measured using three separate scales the Youth Risk

Behavior Survey (Centers for Disease Control and Prevention, 1998), the Minnesota

Student Survery (Minnesota Department of Children Families & Learning, 2001) and the

Diagnostic Interview Schedule for Children (Shaffer, Fisher, Lucas, Dulcan, & Schwab-

Stone, 2000). Suicidal ideation was measured using one instrument, the Suicidal Ideation

Questionnaire (Reynolds, 1988). Since none of the suicide studies attempted to determine

other forms of trauma and victimization it is unclear how much victimization alone

contributed to the suicidal thoughts. This can be said of all the studies including the two

Finkelhor studies since they put sibling and peer abuse in the same cohort (2007a,

2007b).

Despite the method variance it is clear that victimized children are at high risk for

suicide and, perhaps, suicidal and homicidal combinations. All of the studies reported

significant associations between experiences of victimization and suicide attempt, suicide

behavior, and suicidal ideation. The odds of a victimized child having suicidal ideation

and suicide attempts are particularly high but have very large confidence intervals for all

four results which are typical of studies with a relatively small sample. Violent and

suicidal behavior was maintained as a separate category as the two phenomena may not

be related. Some children may be suicidal without having any intention of harming others

while some combine both. This again gives credence to the notion that victimized

children can pose risk for violence including school shootings (Vossekuil et al., 2002).
 228

These teens may be particularly dangerous for such behaviors as vehicular homicide

which may frequently be deliberate suicides that go uncounted if there is no note or other

clear indication of suicidal intention (Freeman, Rossignol, & Hand, 2009).

While all the studies included for review used adolescent samples, Holt and

colleagues (2007), in a study that couldn’t be included due to the lack of convertible data,

also found statistically significant rates of suicidal ideation among all ages, including

young children.35

As this project is being completed, the suicides of two teenagers have been getting

significant media attention. One, a boy, was bullied with homophobic epithets and

another, a girl, was hounded on the Internet.

35
The author was contacted but was not able to furnish the standard deviations required to convert the
means and Ns to a co-variation metric.
 229

PART VII SYNTHESIS OF FINDINGS

In this section the findings will be synthesized. The findings will be analyzed for

their fit with Habermas’s theory, a model of these effects will be depicted, and the two

discrete anomalies found in the review will be discussed. The final section will include a

discussion of the possible positive or neutral effects of bullying experiences and will give

suggestions for future research and interventions. The section ends with a brief discussion

of the place this dissertation may play in the eventual creation of a theory that can be

used to guide social work practice.

The results of this review suggest a logical order to the temporal and spatial

effects of bullying that start with the initial spatial and physical reactions which lead to

changes in the quality of the social environment. This social and spatial experience leads

to repeated affective and cognitive processes that lead to an academic slide, depressive

symptoms and generalized anxiety, and destructive behaviors up to and including suicide.

There is an important caveat that needs to be considered in analyzing this review:

no study isolated peer victimization from other forms of trauma and victimization. The

two related studies by Finkelhor and colleagues (2007a, 2007b) do isolate bullying from

most other forms of victimization but they include sibling victimization with bullying -

the two phenomena may in many cases be unrelated; nevertheless, some of the processes

that compromise or change development may be similar with other forms of

victimization. For example, sibling abuse would make the home as unsafe as the school

perhaps resulting in an additive effect.

This review does not imply causation. The results of this review show a set of

associations that appear to coalesce into a single model: no definitive conclusion about
 230

causation could be made at this stage of scholarship. This review also presents a, perhaps,

idiosyncratic interpretation of Habermas’s concepts: Professor Habermas is retired from

his academic positions and this writer was not able to find a way to correspond with him

on the interpretation of his ideas.

 231

Fit with the Theory of Communicative Action

The systematic review finds overwhelming evidence that Habermas’s theory of

communicative action can explain a large part of the effects of bullying. Most of the

theoretical concepts relevant to the topic of bullying correspond closely to the dimensions

presented in the previous section. In this section the core concepts of the theory will be

compared to the results of the systematic review. In terms of Dubin’s schema it appears

that the theory and empirical results are very close if not identical and is sufficient to

explain all the results.

Such Habermasian concepts as cognitive distortion - in the form of problems with

self-esteem and self-blame - and social competence were measured directly, and the both

binding force and play deprivation were measured indirectly in that both these concepts

can be extrapolated from the data that show victimized children to be socially isolated,

play less, have fewer social attachments and are less physically active.

While the effects of bullying on identity formation was not directly or explicitly

measured, despite the availability of instruments that measure identity in older children,

there is plenty of passive evidence for distortions of identity within this review. A direct

measure of identity would include such things as preferences and interactional styles such

as introspection or sense of humor.

No study measured either role formation or validity claims – attempts to influence

others through communicative and consensual means – of children who are bullied.

While two studies measured bonding this was not taken to be synonymous with

Habermas’s binding energy since that concept hasn’t been directly investigated at this

point: it would be hard to make the case that bonding and attachment were the same as
 232

binding energy when that concept has not been empirically verified. This lack of

confirmation of Habermas’s direct concepts was to be expected as the original research

was not based on Habermas’s theory of communicative action.

Table 18 relates the mechanisms of effect and the predicted results described

above. The first two columns are the predicted mechanisms through which the bullying

experience can damage a child, the third column answers whether or not there is evidence

in the studies reviewed. Evidence that refutes Habermas’s theory would have been

placed in that column as well but no such contradictory evidence was found.

Table 18: Empirical Evidence for Predicted Results

Mechanism of Effect Predicted Results Empirical Evidence
Passive Effects  lower levels of play and activity  yes
 lower levels of autonomy
 decline in social skills  no
 distorted identity  yes
 yes

Active Effects  direct confirmation of theorized  no

processes such as observing validity
claims among children, direct
measurement of illocutionary
bonding, and direct observation of
role rehearsal
 distorted identity based on the
habitual role of victim  partial
 fewer effective instrumental skills
 tendency for the victim role to  no
become active across geographic
contexts  no
 illocutionary bonding
 cognitive and affective distortion
 not in this review
 yes

NB: active and passive results may repeat in the

subsequent categories
ambient stress/shrinking  lack of enjoyment of school  yes
of safe space  lack of safety at school  yes
 sleep disturbance  yes
 concentration disturbance  not directly
 lower academic performance  yes
(Table 18 continued on next page)
 233

Mechanism of Effect Predicted Results Empirical Evidence

deprivation of play  less (more circumscribed) rehearsal of  no
(Pepler) adult roles
 less enjoyment of social world  yes
 less physical activity through play  yes
binding force  attachment and connection  yes (indirectly)
identity problems  distortion or lack of identity formation  yes
(Table continues…)
diminished opportunity  bullying as a preparation for an adult  no data directly measuring
for role preparation role (perhaps with the implicit or role formation
explicit support of adults or older
children) as suggested by Brunkhorst.
 role formation  no
 social skills formation  yes
diminished opportunity  less opportunity to acquire the skills  yes
to improve social skills needed for taking different social
roles in different contexts
 loneliness
 deep imprinting of instrumental social  yes
skills at the expense of the skills  no
required for consensus building

Habermas’s theory of communicative action appears to subsume and explain

several theories and links intrapsychic motivations with socio-cultural forces acting, as

Ritzer (2001) suggests, as a meta-theory. While it is true that Habermas distills many

existing theories and explains the generation of theory as part of the process of

communication, this writer prefers to characterize the theory of communicative action as

extremely broad in its boundaries rather than subsuming other theories. The theory of

communicative action is simply a good social theory that is able to explain the

phenomenon of bullying more clearly than any other.

Habermas derives many of his ideas by synthesizing the work of foundational

thinkers in many disciplines. Considering that all the social scientists contributing to this

review were operating at least in part within the intellectual traditions Habermas

elucidated, it may be inevitable that the results adhere to these essential concepts.
 234

The theory of communicative action, then, can make a complex phenomenon

visible and cohesive. Theory gives focus to analysis and helps to adapt or create language

that is helpful in grasping the social situations and contexts of bullying. Habermas’s

terminology is precise and hard to replace but may lack the type of poetry that it takes to

motivate people to make change.

Anomalies

There were two anomalies within the review, far fewer than had been anticipated.

Of the two anomalies found only one was significant; they were, however, related. The

significant finding was that victimized children were observed talking during play time

which appears to contradict the data that shows that they tend to play less and be more

socially isolated; the other anomalous finding was that they scored, as a group, in a

positive direction for reading scores in comparison to universally depressed scores for all

other academic areas among victimized children. It may be that victimized children are

using decoding skills in order to protect themselves in the social environment; they may

also derive some secondary social benefit from reading; and they may be conversing with

peers in a non-play way, perhaps commiserating with other children who are also

somehow isolated, victimized or rejected.

None of the data appeared to contradict propositions based on Habermas’s theory.

The Visibility of Effects Processes

The theory of communicative action makes visible the nature of instrumental and

communicative states. As had been shown in Table 4 above, various research methods

look at human processes that have varying degrees of visibility. For this systematic

review those obscured processes include internal emotional states of depression and
 235

anxiety and cognitive processes and the visible processes include the quantitative

measures of a child’s social life and behavior. As was discussed above such latent

variables as bonding and cognitive distortion were actually well measured. The link

between play and communicative action is still tenuous but there does appear to be a

negative correlation between victimization experiences and free play.

The following sections elaborate on some of these theorized processes. What

follows is a model created from this systematic review.

Final Model: Developmental Effects

Bullying affects children adversely on several levels. What might seem to be a

global effect of victimization on children’s functioning is, in fact, a specific process

through which children’s short-term well-being and long-term development is affected.

The systematic review data have confirmed enough of the effects predicted by

Habermas’s theory of communicative action to begin the process of creating a model of

the effects of bullying.

The purpose of the model is to understand the patterns and temporal processes

involved when a child is victimized. Theoretical concepts and relationships can be

applied to show that the data work together. The data in this review confirms the passive

effects of bullying: social, spatial, physical, affective- and cognitive-temporal processes,

and play deprivation. Active effects of bullying would need to be confirmed by new

research specifically based on Habermas’s theory.

The data suggest a temporal sequence in the effects of bullying. The initial,

contemporaneous, effect of bullying, together with the set of physical effects, is a

narrowing of social space. The theory of communicative action suggests that social
 236

effects would mediate the other effects of bullying. Those children who can maintain

consensual relationship and retain a space for free play would be predicted to remain

relatively less affected by their experience of victimization; this would need to be

confirmed through a research protocol that tests the direct effects of the theorized

variables.

Bullying can start at any time, although, since it is predicated on power

imbalance, it is likely to start earlier for children entering the system constitutionally

smaller and meeker. Some children have a continuous or episodic, long-term experience

of bullying, while others have a sudden bullying experience after years of successful

social interactions.

Sudden, long-term, or episodic experiences of bullying affect the social world in

different ways. Children with successful social lives can suffer a sudden and cataclysmic

change in their social life; long-term targets will experience a delay in their social skills

development; children with episodic experiences will have periods of remission, in which

their symptoms diminish or disappear. A large part of the latter two groups begin to

experience cognitive and affective distortions over time followed by or consequent to

physical problems, attentional and academic problems, and affective problems up to and

including suicidal ideation and suicide attempts. The processes by which children are

damaged may in most cases follow a rational set of steps.

The ability to rise above conflict, whether petty or crucial, is for Habermas the

highest form of social competence, but suggesting that we rise above conflict does not

mean he is asking us to remain passive in the face of injustice. He is, however, asking us

to stop it from affecting our life globally. As is evident from the relatively low scores for
 237

the victimized cohort some children who are bullied, even repeatedly, may not be

affected globally. They are finding a way to endure the experience without having to

internalize it.

The children who are badly affected by these continuous and repeating

experiences appear to have specific kinds of cognitive reactions to repeated humiliations.

These may be understood as a rational response to real experience rather than arising

from an inherent weakness within the child. A large part of the victimized cohort

experience diminished self-worth and some experience a process in which they blame

themselves for the bullying. It can be surmised that self-blame will be closely associated

with the more severe effects and levels of effects: this should be the subject of future

research.

The concept map in Figure 8 shows a preliminary model for the effects of

bullying almost all of which is directly or indirectly supported by the research. The model

shows how the initial bullying results in a present effect of lessening pleasure and

increasing anxiety in a child. Continuing victimization leads to a set of social effects

including isolation from peers which in turn leads to the deprivation of play and

communicative states and results in weaker social bonds and fewer social skills. Without

other strong supportive environments the child may then experience repeating negative

cognitive and affective process unless protected by internal resources that prevent the

internalization of victimization. These repeated internal processes result in distortions of

identity and, potentially, a set of serious consequences such as depression, conduct

disorders, violence and suicide.

 238

This review begins to address the mechanisms by which children’s development

is affected by peer victimization. Finkelhor (2007b) in suggesting the idea of

developmental victimology would like to find those effects that are common to the other

major forms of victimization of children. A further research project reviewing the effects

of emotional, physical and sexual abuse could identify any common features as well as

syndromes or symptoms that are specific to particular forms of abuse.

Figure 9: Concept Map of Mechanism of Effect

Children Who Are Less Affected/Sources of Statistical Error

In this section sources of error will be explored. In the statistical sense, the term

―source of error‖ refers to the measure of the estimated difference between the observed

effect and the theorized effect. In this case, the largest likely source of statistical error

would be the group of children who remain unaffected by bullying experiences. In the

social sciences error can be understood as all latent, theorized, variables that can explain

the difference in scores. This would include random error which arises from simple

mistakes in orthography or comprehension. In this case the ―error‖ would be the reasons

that children are unaffected by bullying.

 239

Some children who have been bullied appear to be unaffected by the experience

because of their fundamental resilience; others because of problems such as autism

spectrum disorders that make children resistant to social interaction. These children might

have been subject to bullying and would score high on measures of victimization but may

score low on the effects measurements if they do not mind being isolated. Research into

the characteristics, attributes, and strategies of these children will be helpful to design

interventions and to control for statistical error.

Another source of error relates to children who have escaped the victim role. At

any point bullying may end. While bullying tends to be stable, there are always children

whose victim status ends for a variety of reasons: an intervention may help a child learn

to be less affected by bullying, a child may experience a growth spurt that has him

suddenly towering over his former tormentor, or a child may enter a higher status social

group that is less vulnerable to bullying. Further, a bullying child may learn a more

effective set of social skills and thus stop the bullying themselves. It would be expected

that some symptoms of victims would stop after the bullying ceases.

The degree of error involving escaped victims is dependent on the retrospective

span that the researcher uses in her survey (e.g., a question like, Have you been bullied

within the past year?). An escaped victim would endorse herself as a victim if she had

been bullied within that time span but, depending on her resilience, may no longer show

any symptoms related to that bullying event. The longer the retrospective span used, the

higher the source of error. This error could be controlled by asking children who escaped

the victim role to self identify and using regression techniques to determine the

characteristics and effects that differ from children still in that role.
 240

As was stated above, some children may be positively affected by their bullying

experiences. Smith and Talamelli (2004) documented a cohort of children who ascribe

positive effects such as an improvement in self-esteem and self-efficacy after

successfully negotiating bullying experiences. This cohort, depending on the

retrospective span, could endorse themselves as having been bullied but may show higher

self-esteem, self-efficacy, and improved academic performance. Another group of

children not distressed by being bullied may be children raised in highly oppressive and

abusive environments; these children may have little expectation of communicative

states.

Perhaps the most significant source of error in the social dimension is the variable

entry and exit from victimization roles. In order to control for this error longitudinal

methods should be used to track both the well-being of children as well as their initial and

ongoing victimization experiences.

None of the selected studies specifically looked for positive effects of bullying;

however, the evidence that bullying has a global negative impact is overwhelming. It is

likely, as Smith and Talamelli (2004) found, that a proportion of victimized children

having escaped the victim role derived benefit from the bullying experience. It is also

possible that some multiply victimized children derive a benefit from being victimized in

as yet unforeseen ways: for example, a submissive stance may be adaptive for a child

who is being victimized at home and who would be put at risk if she became more

powerful. There is also some possibility that adult masochism may have roots in

childhood experiences and that some children enjoy the bullying as a form of negative

attention.
 241

Such positive effects would be seen in a meta-analysis as a source of error, in that

some proportion of victimized children would score higher on positive traits and lower on

negative effects. Sources of error are discussed in the next section.

 242

Suggestions for Future Research

The above review leads to many suggestions for future research. These will be

briefly outlined below.

Creating Standards

It would be ideal for the community of bullying researchers to create standards

that will make it easier to build knowledge in concert. Studies using comparable

measures and definitions could be easily compared using various meta-analytic

techniques. It may be time for bullying researchers to come together in person and over

the Internet to discuss issues of standardization and to set an agenda for future directions

in bullying research.

Support and Repair: Strengths

Habermas’s theory of communicative action may be able to subsume the strengths

perspective. The bullying research in this review looks at the absence of or alienation

from experiences of support, repair, and solidarity in victimized children’s lives; direct

research into children’s interactions needs to be done to identify acts of solidarity. This

writer conducted a three-year research study, Searching for Safety, into violence and

bullying in an inner city school and identified many minute36 interactions of support and

repair that fit into the schema of communicative action (Kogan & Chandan, unpublished

data).

36
The term minute is used because the acts that were observed tended to be subtle within the context of
violence and bullying: rather than directly confronting the bully, children showed solidarity with victims
through sharing of toys and food, touch and symbolic play, such as talking about superheroes saving the
day.
 243

Future research could observe and document repair processes and compare the

qualities of the social lives of victimized and non-victimized children. It would be

important also to explore the social world of children who escape the victim role or those

who are unaffected by bullies in order to develop novel interventions that prevent or

mitigate the effects of bullying.

Social Skills and Social Connections

The systematic review shows that there is no further need to study the social

connectedness of victimized children. It is clear from multiple studies that they have

smaller friendship networks and are significantly lonelier than their non-victimized peers.

Loneliness may be present in most or all victimized children and factor analysis research

should be done to determine whether measures of loneliness can be used as an indicator

of victimization, or perhaps, an indicator that victimization is having a clearly negative

impact. If loneliness were a good indicator that bullying was occurring, surveying

children to determine their loneliness would be an efficient way for educators and social

workers to identify the often hidden population of victims.

It would be hard to underestimate the importance of social skills development

within childhood for the quality of life of both the child in the present day as well as the

adult that will emerge. A rich social world can be compared to a rich constellation

shimmering stars on a moonless night. A single warm interaction can open an individual

life to infinite possibilities. A chance encounter can spark a relationship that can expose a

young adult to new ways of thinking, music, arts, job opportunities and marriage

prospects. By contrast a narrow social world is a darker world, albeit comfortable for
 244

those with social anxiety or an aversion to unpredictable stimulation, with less

serendipity.

Essentialist Thinking of Victims

It has already been shown that essentialist thinking contributes to the acts of

bullying (S.A. Gelman, 2003; Giles, 2003). The theory of communicative action would

suggest that a self-reification or essentialist thinking process may be fundamental to the

cognitive distortions that victims suffer. It would be important to test children affected

with anxiety, depression, and low self-esteem to determine whether they, as a group, have

a tendency toward essentialist thinking. This would entail the development of an

instrument as the studies done by Giles and Gelman and Heyman have been qualitative in

nature. If it turns out that essentialist thinking is a precursor to any or all ill effects of

bullying, then interventions to change that would be in order.

If reification contributes to the act of bullying and is also a mechanism of

distortion for victims the best intervention may be curricular. Children who can stop the

essentialist thinking will have an easier time developing productive social skills and

being protected from the ill effects of trauma and stress. Schools may be able to teach

flexible thinking and could modify their recess and physical education time to allow

autonomous activities while ensuring that they are relatively free of bullying.

Identity Formation and Bullying

None of the studies conducted in the US measured the effects of victimization on

the formation of identity per se; however, the aggregated data can be seen as synonymous

with identity using the definition of identity as stable characteristics over time and space

(A. van Hoof & Raajimakers, 2003). The study cannot draw a conclusion about any
 245

individual child’s identity but can certainly be used to confirm that victimization distorts

identity in negative directions affecting the cognition, behavior, mood and social life of

the child.

Low play and cognitive distortions such as low self-esteem together with greater

social isolation and diminished social skills can be expected to compromise the formation

of identity, those set of verbal and physical actions and reactions that repeat over time in

such a way that these characteristics can identify the individual. Future research would be

helpful in particular looking at how different forms of bullying can affect identity. It is

possible that researchers will find a pattern to the identity of some victims such that they

are more likely to be selected as victims across social contexts.

 246

Developmental Victimology

As was suggested above, future research could compare the constellation of

effects of other forms of abuse experienced in childhood in order to determine whether

there are common effects between and among different forms of victimization. Mynard,

Joseph and Alexander’s (2000)

research shows different types of victimization have different sequelae, so it is more

likely that there will be common effects to the qualities of the different forms of

victimization. For example, forms of family violence that rely on isolation will likely

result in similar effects as social bullying; similarly, it is likely that emotional abuse by

adults will result in diminishing self-worth in similar ways that verbal victimization does

in children. Determining these commonalities and differences will require a concerted

research effort using both secondary data as well as original research.

Suggestions for Bullying Interventions

The value to the social work profession for testing Habermas’s theory is to guide and

improve practice in order to improve the well-being of children. If a theory is able to

explain the nature of a social phenomenon including its effects it follows logically that it

would then be able to form the basis for a novel and effective intervention. For bullying,

a comprehensive theory should prevent bullying from occurring, stop it when it does

occur and mitigate its ill effects.

Such a project to really end deleterious bullying would require intervention on several

levels. In the US legislation at the State level requires district-level intervention; but an

effective intervention based on the theory of communicative action should involve a

 247

change in the national education policy as well as a change in the culture in favor of

genuinely egalitarian and inclusive practices. Such a vision, at this time, may seem

chimeric; however, Habermas’s theory provides a roadmap for future social work policy

advocacy.

The above review suggests several interventions that can be formed from the theory

of communicative action. Some of these interventions would not differ from existing

anti-bullying interventions, whereas others would look radically different. What follows

is a series of ideas for interventions that are suggested by the theory.

Types of Speech/Act and Bullying Interventions

A communicative action intervention would consider the different types of

speech/act: instrumental and communicative. Whereas several existing anti-bullying

programs emphasize the acquisition of positive social skills, which would be considered

communicative speech/acts, a good intervention based on the theory should also

emphasize promote the acquisition of instrumental skills. In line with Habermas’s neutral

definition of instrumental action such a program would teach children how to initiate

instrumental action effectively and respectfully. As had been pointed out above, children

have many naturally-occurring opportunities to employ instrumental speech/acts: children

may take informal and formal instrumental roles such as team captain, school prefect or

hall monitor, and older relatives and friends often keep an eye out for the safety of

smaller children safe from cars or unknown adults; some children may humanely help

other children learn about and adhere to school norms.

An intervention program based on communicative action would include modules of

teaching the skills of instrumental action in order to divert the bullying impulse into
 248

positive directions. Teaching and promoting instrumental skills would involve teaching

children how to use social power while taking into considerations the ability and desires

of others, rather than riding roughshod over the autonomy of others. Such an approach

would have the added advantage of changing the framework from a bully/victim

dichotomy to one in which the instrumental aspirations at the heart of a bully’s actions

can be accepted and respected. Such aspirations, however, require considerable resources.

Rich interest-based curriculum and extra-curricular activities are essential elements of

a theory of communicative action-based program. A child who is motivated to bully by a

desire to entertain can be provided with creative opportunities to build on that desire

through theater or music programming, and a child whose bullying is motivated by a

desire to enforce norms could become involved in student government. The latter

suggestion would involve considerable adult support to integrate children with

―problems‖ into activities frequently exclusive to higher functioning and compliant

children. Such suggestions would require educators to endorse the ideas of

communicative action and integrate these into pedagogical theory.

Social Skills and Social Bonding

Almost every credible bullying intervention program stresses the importance of

raising the level of social skill, particularly for the bullying child. A program based on the

theory of communicative action would also seek to improve social skills but would do so

equally for both victims and bullies. This is because the theory and the research in this

review show that victimization delays and impedes the acquisition of social skills.

Social skills have primacy within the theory because they lead to deeper and more

effective social bonding which, in turn, opens up an infinitely variable richness of the
 249

social world. These social bonds may be deep and intimate but others may be more

superficial and temporary. Social bonds lead to healthier social action than does the

instrumental action based on coercion and simple coordination. Commitment, convention

and covenant describe the bonds underlying very strongly motivated action; these

contrast with instrumentally-derived action, which represent compliance with benign or

malign authority.

Interpreting Habermas’s schema, such healthy bonding necessarily requires children

to consider the standpoints of others and requires that the child has had a rich and diverse

experience of other children in social contexts with sufficient autonomy for self-

exploration. These contexts are most frequently found in free or freer play activities.

The systematic review above suggests that the deprivation of play is one of the main

detrimental effects of bullying. An effective intervention would be to establish or restore

free play contexts. After bullying has become part of a school culture, however, it would

likely continue to instrumentalize a large segment of a free play environment. In other

words, without an intervention bullies would continue to instrumentalize most

autonomous contexts.

The promotion of free play presents a paradox for social workers and educators

attempting to prevent, ameliorate, and arrest bullying. Autonomy is a complex issue

when applied to the problem of bullying. The theory would suggest that victimized

children still need an autonomous environment yet helping and teaching professionals

tend to increase structure and adult supervision for children experiencing any kind of

problem. Extra-curricular activities that provide more autonomous environments are

often comprised of the least needy and most academically accomplished children, and
 250

many such activities are available only to middle and upper class children. This writer

was unable to find the quantitative data that would support the former statement; but it

has certainly been true in his practice experience spanning three countries and two

continents.

Communicative action occurring in autonomous social contexts is essential to the

development of a healthy identity; yet bullying occurs unchecked where adult supervision

is sparse. The social skills curriculum, Second Step **, includes regular discussion about

events that occur that day on the playground and how the children applied social skills

they’ve learned in a real-world context. An intervention program should include such a

curriculum component as well as opportunities for children to interact in unstructured

contexts. Adult supervision, however, is essential so long as the adults limit their

intervention to teaching appropriate skills on the playground.

On the level of macro-level, national policy the institutional arrangements of the

school system, as Noguera (1995)suggests, need to be reconstituted to allow for greater

autonomy. An effective policy would allow more opportunities for marginalized and

distressed children to experience more, not less, autonomy. It would involve training

teachers to implement more unstructured activities into classroom activities as well as

seeing arts and special interests as a universal right for all children rather than a finite

commodity for a privileged few. Such a policy would also need to pay attention to non-

academic aspects of the school system in general including how transitions and

transportation occur and how playgrounds are supervised.

 251

Toward a Lifeworld Theory of Social Work Practice

The systematic review in this dissertation is on the topic of school bullying, yet,

this writer’s application of Habermas’s theory of communicative action to the problem of

school bullying has led him to the conclusion that Habermas’s social theory has the

potential to provide the basis for an effective and comprehensive social work theory

making a significant contribution to social work’s major methods. Habermas’s theory not

only adds considerably to our understanding of the social world, it is also able to subsume

and integrate a number of existing theories across disparate disciplines. In this section,

this writer will explore the assertion that Habermas’s theory of communicative action

shows promise as a general social work practice theory.

As has been reiterated in this paper, the articles selected for the school bullying

review, with few exceptions, offered no coherent theory about how victimized children

are affected by bullying. Through this project, this writer found that Habermas’s theory

of communicative action is able to subsume the theories that explain how bullying arises

and also shows how bullying affects children. The theories that Habermas’s theory

subsumes in relation to bullying are: social information-processing theory which

emphasizes an internal representation of social processes; social learning theory which

emphasizes the ability of experiences to result in children’s distorted behaviors; and

attachment theory, which suggests that the ability of a child to form secure bonds is an

essential component to future well-being. In addition, Habermas’s theory makes visible

the actions of support which constitute the anodyne power of supportive acts similar to

the factors described in social work’s strengths perspective literature, as is evidenced by

the deleterious effects of their absence.

 252

The data from this review of bullying, interpreted from a Habermasian theoretical

frame, show what happens when children experience a surfeit of instrumental social

states. The victimized children frequently react through their bodies with anxiety and

school avoidance. Those children who continue to be adversely affected are significantly

lonelier, and their social skills development is arrested. It is clear from the data that these

affected children play less and are generally less physically active. With time, this social

isolation has a global effect on mood and academic functioning for children that

internalize negative self-assessments and sentiments; and, some percentage of victimized

children will act out with behaviors that injure themselves and others; those most

symptomatic may develop chronic anxiety and depression.

The Theory of Communicative Action’s Unique Contribution

Habermas’s theory of communicative action, through the universal pragmatics’

method, is rigorously based on life processes. Universal pragmatics functions for social

science in a way that is similar to the null hypothesis in hard science, eliminating all ideas

whose opposite cannot be proven to be true in the social world. While universal

pragmatics is most often described as the search for the conditions for mutual

understanding, this description misses the larger scope of Habermas’s work with its

ability to explain both meaning making and action—a far wider sphere of theorizing that

includes most or all of the social world, including the connection to the subjective and

intrapsychic experience. Universal pragmatics also looks for ideas that are capable of

crossing disciplines: this is similar to the way the social work profession finds ideas from

all appropriate disciplines with which to intervene in the real world.

 253

Habermas describes the nature of human speech and the coordination of social

actions. His division of speech/acts into instrumental and communicative categories

allows the social work profession to have a deeper understanding of the nature of social

environments. An instrumental social context, by its nature, limits the autonomy of the

people involved by lowering the number of choices of appropriate responsive

speech/acts; by contrast, a truly communicative social context allows for a theoretically

infinite number of choices of speech/acts. Bullying, as a form of oppression, is an

instrumental social context that can damage communicative social contexts in the form of

play and autonomous social interaction.

This writer’s project was limited to school bullying, yet adult experiences of

abuse and oppression may affect people through similar mechanisms, although without

the element of rapid developmental change that characterizes childhood. Domestic

violence perpetrators, for example, tend to isolate their partners from social supports and

systematically undermine their partners’ self-esteem in ways that are similar to child

bullies’ speech/acts against their victims. Longitudinal research being conducted in

Europe will presently have data about the quality of the marriages of the bullies and

victims studied in childhood; it will be interesting to see whether the phenomenon of

school bullying transfers into adulthood in the form of domestic abuse or community

violence.

By explaining the conditions for communication and the mechanisms of social

action, Habermas’s theory of communicative action could be used to explain not only

bullying but other problems within the social work field. Habermas’s theory is broad

enough to explain the etiology of large social forces and their connection to minute
 254

internal processes. By linking intrapsychic processes with larger social forces, the theory

covers the same territory as such social work models as the ecological and bio-

psychosocial models.

Habermas’s theory of communicative action not only subsumes the theories

mentioned above in relation to bullying, but also materialist theories, such as

behaviorism, evolutionary biology, and economic theories that are predicated on rewards

that accrue to a person through particular behaviors. Habermas, indeed, extends social

theory into an intersubjective dimension that acknowledges a set of behaviors that people

engage in even in the absence of reward: these behaviors include meaning making,

altruism, and acts of solidarity and support that can’t always be reciprocated, and may, at

times, be risky, as in the case of defending a victim. By subsuming materialist theory,

Habermas allows that much human activity is verifiably motivated by rewards, but under

some conditions, human behavior occurs without reasonable expectation of reward.

Habermas’s theory of communicative action has considerable breadth and can be

applied to a wide range of social phenomena. The ability of Habermas’s theory to

subsume various social work theories confirms Ritzer’s notion that the theory is a ―meta-

theory.‖ This writer, however, prefers to see Habermas’s theory of communicative action

from Dubin’s schema of theory building in which the theory can be seen to have a very

large boundary: it is able to explain and predict intrapsychic effects such as cognitive and

behavioral distortion, as well as organization and maintenance of complex societies.

Habermas’s theory explains all non-genetic sources of social action, in other words, all

action that is socially, cognitively, affectively, as well as biologically determined. His

notion of the lifeworld contains the biological needs and constraints that human
 255

communities must furnish to flourish. Systems that are in synch with these human needs

and constraints provide services and organization to complex groupings of people.

Pathology occurs when the systems diverge from the lifeworld.

A Lifeworld Theory of Social Work

This writer proposes that a new theory based on Habermas’s theory of

communicative action be coined for social work. This writer will call this the ―Lifeworld

Theory.‖ The Lifeworld Theory would explain the connection between larger social

phenomena and the cognitive, affective, and even physical elements of the individual

human being. The Lifeworld Theory would absorb social work ideas about non-genetic

sources of distortion, such as family influences, social forces that limit organic drives,

and interest-based motivations.

In the lifeworld, the boundaries between communicative and instrumental states

are not simple: within the same social group, sub-groups may have very different

experiences. Communicative action within communicative social states should not be

seen as an ideal utopian context that would exist if not for nefarious faceless conspirators,

although there is room in Habermas’s theory for that degree of social manipulation;

rather, his theory should be understood as a way of analyzing the social world. Within the

same geographical area, the nature of the social community could vary greatly as positive

sub-groups coalesce around individuals and communities with good social skills.

Strengths Perspective and Social Work Methods

A Lifeworld Theory could be used for all major social work methods:

casework, group work, community organization, administration, and policy. It also has
 256

the potential to explain the nature of human strength and resilience as part of a body of

theory rather than, as is currently practiced, a model or perspective.

The power of communicative action and the binding force of consensually-

derived action explain the effectiveness of group work as a treatment and mutual support

modality. Community organization could benefit from the nuanced analysis of

communicative and instrumental states, and social work administrators practicing from a

Lifeworld Theory would have very specific skills and tools to analyze the informal and

formal communication within agencies and make sure the systems stay aligned with the

lifeworld. A Lifeworld Theory-based casework approach would certainly result in

interventions that could uncover and mobilize strengths and social skills.

Considerable practice and research would be required to determine whether and to

what degree a Lifeworld Theory could help social workers intervene effectively. The

apparent applicability to all social work methods, however, indicates that this theory

would be a good candidate for a general social work theory that could be taught to social

workers and effectively implemented in practice settings.

 257

APPENDIX I Criteria for Study Selection: Databases and Key Words

Bibliographic Databases:
BiblioBranchee (France and Quebec)
Cochrane Collaboration
Campbell Collaboration C2-SPECTR
Database of reviews of effectiveness (DARE online),
Dissertation Abstracts International (DAI)
ERIC
MEDLINE
NCAAN Information (Child Abuse and Neglect)
Psychological Abstracts (PsycINFO, PsycLIT)
SCOPUS
Social Care Institute for Excellence (SCIE) (UK)
Social Work Abstracts
Social Science Citation Index
Social Sciences Abstracts
Social Service Abstracts
Sociological Abstracts (Sociofile)
Urban Studies Abstracts

Search Engines:
Biblioline
De.licio.us (conceptual ontologies website)
Francite
Google/Google Scholar
Lexus Nexus
Lycos (France)

Keywords and subject headings were searched systematically (searches were modified
according to the specific database):
Bullying
Child
Child Abuse
Cyberbullying
Harassment/Harcellement
Identite/Identity
Intervention
Intimidation
Peer Victimization
Peer Harassment
Power/Pouvoir
Prevention
School
 258

REFERENCES

Abramovitz, M. (1981). Business and health reform: Worker's compensation and health
insurance in the progressive era. Columbia University, New York.

Achenbach, T. M. (1991). Manual for Child Behavior Checklist/ 4-18 and 1991 Profile.
Burlington, VT: University of Vermont, Dept. of Psychiatry.

Achenbach, T. M., & Rescoria, L. (2001). Manual for the ASEBA school-age forms and
profiles. Burlington, VT: University of Vermont, Research Center for Children,
Youth and Families.

Adelman, H. S., & Taylor, L. (1998). Reframing mental health in schools and expanding
school reform. Educational Psychologist, 33, 134-152.

Adelmann, P. K. (2005). Social environmental factors and preteen health-related

behaviors. Journal of Adolescent Health, 36, 36-47.

Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Nevitt Sanford, R. (Eds.).
(1982). The Authoritarian Personality. New York: Norton & Company.

Ahmed, E., & Braithwaite, V. (2004). Bullying and victimization: Cause for concern for
both families and schools. Social Psychology of Education, 7, 35-54.

Allen-Meares, P., Washington, R. O., & Welsh, B. L. (2000). Social work services in
schools. Boston: Allyn & Bacon.

Allen, A. R. (2007). Systemically distorted subjectivity? Philosophy and Social

Criticism, 33(5), 641-650.

Alsaker, F. D. (1993). Isolement et maltraitance par les pairs dans les jardins d'enfants:
Comment mesurer ces phenoomenes et quelles sont leurs consequences? Enfance,
47(3), 241-260.

Altemeyer, B. (1981). Right-wing authoritarianism. Winnipeg, Canada: University of

Manitoba Press.
 259

Ambert, A. (1994). A Qualitative Study of Peer Abuse and Its Effects: Theoretical and
empirical implications. Journal of Marriage and the Family, 56, 119-130.

Andrews, G. J., & Chen, S. (2006). The production of tyrannical space. Children’s
Geographies, 42(2), 239–250.

Antonovsky, A. (1996). The salutogenic model as a theory to guide health promotion.

Health Promotion International, 11(1), 11-18.

Arendt, H. (2006). On Revolution. New York: Viking Press.

Arora, C. (1994). Measuring bullying with the life in school checklist. Pastoral Care in
Education, 12(3), 11-16.

Arora, C. (1996). Defining bullying: Toward a clearer general understanding. School

Psychology International, 17, 317-329.

Arora, C., & Thompson, D. (1987). Defining bullying for a secondary school.
Educational and Child Psychology, 4, 110-120.

Asher, S. R., & Wheeler, V. A. (1985). Children’s loneliness: A comparison of rejected

and neglected peer status. Journal of Consulting and Clinical Psychology, 53(4),
500-505.

Aspinwall, P. S., Greene, B. R., & Morrison, D. R. (1994). Calabi-Yau moduli space,
mirror manifolds and spacetime topology change in String Theory. Nuclear
Physics, B416, 414-480.

Astor, R. A. (1998). Moral reasoning about school violence: Informational assumptions

about harm within school subcontexts. Educational Psychologist, 33(4), 207-221.

Astor, R. A., & Meyer, H. A. (1999). Where girls and women won't go: Female students',
teachers', and social workers views of school safety. Social Work in Education,
21(4), 201-219.

Astor, R. A., Meyer, H. A., & Behre, W. J. (1999). Unowned places and times: Maps and
interviews about violence in schools. American Educational Research Journal,
36(1), 3-42.
 260

Astor, R. A., Meyer, H. A., & Pitner, R. O. (2001). Elementary and middle school
students' perceptions of violence-prone school subcontexts. Elementary School
Journal, 101(5), 511-528.

Bachrach, P., & Barantz, M. (1962). The two faces of power. American Political Science
Review, 56, 947-952.

Bachrach, P., & Botwinick., A. (1992). Power and empowerment: A radical theory of
participatory democracy. Philadelphia: Temple University Press.

Baldry, A. (2004). The impact of direct and indirect bullying on the mental and physical
health of Italian youngsters. Aggressive Behavior, 30, 343-355.

Ballard, M., Argus, T., & Remley, T. P., Jr. . (1999). Bullying and School Violence: A
Proposed Prevention Program. NASSP Bulletin, 38-47.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.

Psychological Review, 84, 191-215.

Bandura, A. (1979). Social learning theory. Englewood Cliffs, NJ: Prentice Hall.

Bangert-Drowns, R. L. (1986). Review of developments in meta-analytic methods.

Psychological Bulletin, 99, 388-399.

Baumrind, D. (1991). The influence of parenting style on adolescent competence and

substance use Journal of Early Adolescence, 11(1), 56-95.

Beck, A. T., & Steer, R. A. (1993). Manual for the Beck Depression Inventory. San
Antonio, TX: Psychological Corporation.

Berthold, K. A., & Hoover, J. H. (2000). Correlates of bullying and victimization among
intermediate students in the Midwestern US. School Psychology International,
21(1), 65-78.

Bertschinger, E., Olbrich, E., Ay, N., & Jost, J. (2006). Autonomy: an information-
theoretic perspective. Paper presented at the ALife X conference. Retrieved from
http://www.santafe.edu/research/publications/workingpapers/06-10-035.pdf
 261

Birch, S. H., & Ladd, G. W. (1997). The teacher-child relationship and childrens' early
school adjustment. Journal of School Psychology, 35, 1-17.

Bjorkqvist, K., Lagerspetz, K. M. J., & Kaukiainen, A. (1992). Do girls manipulate and
boys fight? Development trends in regards to direct and indirect aggression.
Aggressive Behavior, 18, 117-127.

Black, S. A., & Jackson, E. (2007). Using bullying incident density to evaluate the
Olweus bullying programme. School Psychology International, 28(5), 623-638.

Blackburn, J. A., Dulmus, C. N., Theriot, M. T., & Sowers, K. M. (2004). Kids-parents-
school personnel: Few are talking and even less are listening Journal of Evidence-
Based Social Work, 1(2/3), 45-57.

Blagasklonny, M. V., & Pardee, A. B. (2002). Conceptual Biology: Unearthing the gems.
Nature, 416, 373.

Blakely, M. L. (2004). Unmasking bullies and victims: A guide to their physiological,

psychological, and emotional patterns. Coral Springs, Fl.: Lumina Press.

Blass. (2000). Obedience to authority: Current perspectives on the Milgram paradigm.

Hillsdale, NJ: Lawrence Erlbaum.

Borradori, G. (2003). Philosophy in a time of terror: Dialogues with Jurgen Habermas

and Jacques Derrida. Chicago: University of Chicago Press.

Borsboom, D., Mellenbergh, G. J., & van Heerden, J. (2003). The theoretical status of
latent variables. Psychological Review, 110(2), 203-219.

Bosworth, K., Espelage, D. L., & Simon, T. (1999). Factors associated with bullying
behavior in middle school students. Journal of Early Adolescence, 19(3), 341-
362.

Boulton, M. J. (1999). Concurrent and longitudinal relations between children's

playground behavior and social preference, victimization and bullying. Child
Development, 70(4), 944-954.
 262

Bowen, N. K., Bowen, G. L., & Ware, W. B. (2002). Neighborhood social

disorganization, families, and the educational behavior of adolescents. Journal of
Adolescent Research, 17(5), 468-490.

Bowers, L., P.K., S., & Binney, V. (1994). Perceived Family Relationships of Bullies,
Victims and Bully/Victims in Middle Childhood. Journal of Social and Personal
Relationships, 11(2), 215-232.

Bowlby, J. (1999). Attachment and Loss (2 ed. Vol. 1). New York: Basic Books.

Brante, T., 31(2), . (2001). Consequences of realism for sociological theory-building.

Journal for the Theory of Social Behaviour, 31(2), 167 - 195.

Briere, J. (1996). Trauma symptom checklist for children, professional manual. : .

Odessa, FL: Psychological Assessment Resources.

Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature

and design. Cambridge, MA: Harvard University Press.

Brown, R. H., & Goodman, D. (2001). Jurgen Habermas’ theory of communicative

action: An incomplete project. In G. S. Ritzer, B. (Ed.), Handbook of social
theory. Thousand Oaks, CA: Sage Publications.

Brown, S., & Taylor, K. (2008). Bullying, education and earnings: Evidence from the
National Child Development Study. Economics of Education Review, 27, 387-
401.

Brunkhorst, H. (2004). Critical Theory and Empirical Research In D. M. Rasmussen & J.

Swindal (Eds.), Critical Theory.

Brunstein Klomek, A., Marrocco, F., Kleinman, M., Schonfeld, I. S., & Gould, M. S.
(2008). Peer victimization, depression, and suicidality in adolescents. Suicide and
life-threatening behavior, 38(2), 166-180.

Buhs, E. S., & Ladd, G. W. (2001). Peer rejection as an antecedent of young children's
school adjustment: An examination of mediating processes. Developmental
Psychology, 37(4), 550-560.
 263

Buhs, E. S., Ladd, G. W., & Herald, S. L. (2006). Peer exclusion and victimization:
Processes that mediate the relation between peer group rejection and children's
classroom engagement and achievement? Journal of Educational Psychology,
98(1), 1-13.

Bukowski, W. M., & Adams, R. (2005). Peer relationships and psychopathology:

Markers, moderators, mechanisms and meanings. Journal of Clinical Child and
Adolescent Psychology, 34(1), 3-10.

Burger, P., & Luckman, T. (1966). The social construction of reality. Garden City, NY:
Doubleday.

Burt, K. B., Obradovic, J., Long, J. D., & Masten, A. S. (2008). The interplay of social
competence and psychopathology over 20 years: Testing transactional and
cascade models. Child Development, 79(2), 359-374.

Bus, L. W. (1987). The evolution of individuality. Princeton, NJ: Princeton University

Press.

Bybee, J. S. (2002). Memorandum for John Rizzo Acting General Counsel of the Central
Intelligence Agency.

Carney, J. (2000). Bullied to death: Perceptions of peer abuse and suicidal behavior
during adolescence. School Psychology International, 25(2), 213–223.

Caspi, A., Moffitt, T., Entner Wright, B., & Silva, P. (1998). Early failure in the labor
market: Childhood and adolescent predictors of unemployment in the transition
to adulthood. American Sociological Review, 63, 424-451.

Cassidy, J., & Asher, S. (1992). Loneliness and Social Dissatisfaction Questionnaire. In
G. Kogan (Ed.). Croton-on-Hudson, NY.

Catalano, R. F., Haggerty, K. P., Oesterle, S., Fleming, C. B., & Hawkins, J. D. (2004).
The importance of bonding to school for healthy development: Findings from the
Social Development Research Group. Journal of School Health, 74(7), 252-261.

Causey, D. L., & Dubow, E. F. (1992). Development of a self-report coping measure for
elementary school children. Journal of Clinical Child Psychology, 21(1), 47-59.
 264

Centers for Disease Control and Prevention. (1998). Youth risk behavior surveillance-US
1997. Morbidity and Mortality Weekly Report, 47(SS-3).

Chen, X., Chang, L., Liu, H., & He, Y. (2008). Effects of the peer group on the
development of social functioning and academic achievement: A longitudinal
study in Chinese children. Child Development, 79(5), 239-251.

Cheung, M. W.-L., & Au, K. (2005). Applications of multilevel structural equation

modeling to cross-cultural research. Structural Equation Modeling, 12(4), 598-
619.

Cheyne, J. A. (1998). Mimesis and semiotics of bullying: A group- and self-defining

ritual. Retrieved from
http://watarts.uwaterloo.ca/~acheyne/Misc/MimeticBullying.html

Chinn, S. (2000). A simple method for converting an odds ratio to effect size for use in
meta-analysis. Statistics in Medicine, 19, 3127-3131.

Christie, J. F., & Johnsen, P. (1987). Reconceptualizing constructive play: A review of

the empirical literature. Merrill-Palmer Quarterly, 33(4), 439-452.

Clarke, E. A., & Kiselica, M. S. (1997). A systemic counseling approach to the problem
of bullying. Elementary School Guidance and Counseling, 31, 310-324.

Cleary, S. D. (2000). Adolescent victimization and associated suicidal and violent

behaviors. Adolescence, 35(140), 671-682.

Coie, J., Dodge, K., Terry, R., & Wright, V. (1991). The role of aggression in peer
relations: An analysis of aggression episodes in boys’ play groups. Child
Development, 62, 812-826.

Coie, J., Lochman, J., Terry, R., & Hyman, C. (1992). Predicting early adolescent
disorder from childhood aggression and peer rejection. Journal of Consulting and
Clinical Psychology, 60(5), 783-792.

Cole-Hamilton, I., Harrop, A., & Street, C. (2005). Making the Case for Play: Gathering
the Evidence. Children, Youth and Environments, 15(2). Retrieved from
http://thunder1.cudenver.edu/cye_journal/review.pl?n=36
 265

Cook, T. D. (2002). Advanced statistics: Up with odds ratios! A case for odds ratios
when outcomes are common. Academy of Emergency Medicine, 9(12), 1430-
1434.

Cowie, H., Naylor, P., Rivers, I., Smith, P. K., & Pereira, B. (2002). Measuring
workplace bullying. Aggression and Violent Behavior, 7(1), 33-51.

Craig, W. M. (1998). The relationship among bullying, victimization, anxiety and

aggression in elementary school children. Personality and Individual Difference,
24(1), 123-130.

Craig, W. M., & Pepler, D. J. (2003). Identifying and targeting risk for involvement in
bullying and victimization. Canadian Journal of Psychiatry, 48(9), 577-582.

Craig, W. M., Pepler, D. J., & Blais, J. (2007). Responding to bullying: What works?
School Psychology International, 28(4), 465-477.

Crain, M. M., Finch, C. L., & Foster, S. L. (2005). The relevance of the social
information processing model for understanding relational aggression in girls.
Merrill-Palmer Quarterly, 51(2), 213-249.

Crick, N. R. (1996). The role of overt aggression, relational aggression, and prosocial
behavior in the prediction of children’s future social adjustment. Child
Development, 67, 2317–2327.

Crick, N. R., Casas, J. F., & Mosher, M. (1997). Relational and overt aggression in
preschool. Developmental Psychology, 33(4), 579-588.

Crick, N. R., & Dodge, K. (1994). A review and reformulation of social information
processing mechanisms in children’s development. Psychological Bulletin, 114,
74-110.

Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender and social-
psychological adjustment. Child Development, 66, 710-722.

Crick, N. R., & Grotpeter, J. K. (1996). Children's treatment by peers: Victims of

relational and overt aggression. Development and Psychopathology, 8(2), 367-
380.
 266

Crick, N. R., Grotpeter, J. K., & Bigbee, M. A. (2002). Relationally and physically
aggressive children's intent attributions and feelings of distress for relational and
instrumentla peer provocations. Child Development, 73(4), 1134-1142.

Crossley, N. (1996). Intersubjectivity: The Fabric of social becoming. London: Sage

Publications.

Crothers, L. M., Field, J. E., & Kolbert, J. B. (2005). Navigating power, control and being
nice: Aggression in adolescent girls' friendships. Journal of Counseling &
Development, 83, 349-354.

Crothers, L. M., & Levinson, E. M. (2004). Assessment of bullying: A review of methods

and instruments. Journal of Counseling & Development, 82(4), 496-503.

Cunningham, N. J. (2007). Levels of bonding to school and perceptions of the school

environment by bullies, victims and bully victims. Journal of Early Adolescence,
27(4), 457-478.

Currie, C., Gabhainn, S. N., Godeau, E., Roberts, C., Smith, R., Currie, D., et al. (2008).
Inequalities in young people's health: HBSC International report from the
2005/2006 survey. Edindburgh, Scotland: The University of Edinburgh.

Dahl, R. (1957). The concept of power. Behavioral Science, 2, 201-205.

Dahrendorf, R. (1958). Out of utopia: Toward a reorientation of sociological analysis.

American Journal of Sociology64, 64, 115-127.

Davies, B. (1982). Life in the classroom and playground: The accounts of primary school
primary school children. Boston: Routledge & Kegan Paul.

Derrida, J. (1974). Of grammatology, publ. s., 1974. Baltimore, Md: John Hopkins
University Press.

Derrida, J. (1988). Letter to a Japanese friend (A. Benjamin & D. Wood, Trans.). In D.
Wood & R. Bernasconi (Eds.), Derrida and differance. Evanston, Il: Northwest
University Press.
 267

DeVoe, J. F., Peter, K., Kaufman, P., Miller, A., Noonan, M., Snyder, T. D., et al. (2004).
Indicators of school crime and safety: 2004.

Dijkstra, J. K., Lindenberg, S., & Veenstra, R. (2007). Same-gender and cross-gender
peer acceptance and peer rejection and their relation to bullying and helping
among preadolescents: Comparing predictions from gender-homophily and goal-
framing approaches Developmental Psychology, 43(6), 1377-1389.

Dixon, R., Smith, P. K., & Jenks, C. (2004). Managing difference and managing
bullying: A case study of one secondary school. Journal of School Violence, 3(4),
41-58.

Dodge, K. A. (2004). Returning to roots: On social information processing and moral

development. Child Development, 75(4).

Dodge, K. A., & Coie, J. D. (1987). Teacher rating of depression. Journal of Personality
and Social Psychology, 53, 1146-1156.

Dodge, K. A., & Frame, C. (1982). Social cognitive biases and deficits in aggressive
boys. Child Development, 53, 620-635.

Dubin, R. (1978). Theory building. New York: Free Press.

Ducharme, J. M., Folino, A., & DeRosie, J. (2008). Errorless acquiescence training: A
potential "keystone" approach to building peer interaction skills in children with
severe problem behavior. Behavior Modification, 32(1), 39-60.

Due, P., Holstein, B. E., Lynch, J., Diderichsen, F., Gabhain, S. N., Scheidt, P., et al.
(2005). Bullying and symptoms among school-aged children: International
comparative cross sectional study in 28 countries. European Journal of Public
Health, 15(2), 128-132.

Dupper, D. R., & Meyer-Adams, N. (2002). Low-level violence: A neglected aspect of

school culture. Urban Education, 37(3), 350-364.

Egan, S., & Perry, D. (1998). Does low self-regard invite victimization? Development
Psychology, 14(2), 299-399.
 268

Ellis, W. E., & Zarbatany, L. (2007). Explaining friendship formation and friendship
stability. Merrill-Palmer Quarterly, 53(1), 79-104.

Engels, F. (1954). Dialectics of nature (C. Dutt, Trans.). Moscow: Foreign Languages
Publishing House.

Erikson, E. H. (1974). Dimensions of a new identity. New York: Norton.

Espelage, D. L., Bosworth, K., & Simon, T. R. (2001). Short-term stability and
prospective correlates of bullying in middle-school students: An examination of
potential demographic, psychosocial and environmental influences. Violence and
Victims, 16(4), 411-426.

Espelage, D. L., Holt, M. K., & Henkel, R. R. (2003). Examination of peer-group

contextual effects on aggression during early adolescence. Child Development,
74(1), 205-220.

Estell, D. B., Farmer, T. W., & Cairns, B. D. (2007). Bullies and victims in rural African
American youth: Behavioral characteristics and social network placement.
Aggressive Behavior, 33, 145-159.

Farrington, D. (1993). Understanding and preventing bullying. Crime and Justice: A

review of research (pp. 381-458). Chicago: University of Chicago Press.

Faust, J., & Forehand, R. (1994). Adolescents' physical complaints as a function of

anxiety due to familial and peer stress: A causal model. Journal of Anxiety
DIsorders, 8(2), 139-153.

Ferry, J. M. (1991). Les puissance de l'experience: essais sur l'identite contemporaine

(Vol. 1, Le sujet et le verbe). Paris: Les editions du Cerf.

Finkelhor, D. (1995). The victimization of children in a developmental perspective.

American Journal of Orthopsychiatry, 65(2), 177-193.

Finkelhor, D. (2007a). Developmental victimology: The comprehensive study of child

victimizations. In R. C. Davis, A. J. Lurigio & S. Herman (Eds.), Victims of crime
(3 ed., pp. 9-34). Thousand Oaks, CA: Sage.
 269

Finkelhor, D. (2007b). Developmental victimology: The comprehensive study of

childhood victimization. In R. C. Davis, A. J. Luirigio & S. Herman (Eds.),
Victims of crime (pp. 9-34). Thousand Oaks: CA: Sage.

Finkelhor, D., Ormrod, R., & Turner, H. A. (2007a). Poly-victimization: A neglected

component in child victimization. Child Abuse and Neglect, 31, 7-26.

Finkelhor, D., Ormrod, R., & Turner, H. A. (2007b). Polyvictimization and trauma in a
national longitudinal cohort. Development and Psychopathology, 19(1), 149-166.

Finkelhor, D., Ormrod, R., Turner, H. A., & Hamby, S. L. (2005a). Child Maltreatment,
10(1), 5-25.

Finkelhor, D., Ormrod, R., Turner, H. A., & Hamby, S. L. (2005b). Measuring poly-
victimization using the Juvenile Victimization Scale. Child Abuse and Neglect,
29, 1297-1312.

Finnegan, R. A., Hodges, E. V. E., & Perry, D. G. (1998). Victimization by peers:

Associations with children's reports of mother-child interaction. Journal of
Personality and Social Psychology, 75(4), 1076-1086.

Fook, J. (1993). Radical casework: A theory of practice. St. Leonards, Australia: Allen
and Unwin.

Fook, J. (2002). Social Work: Critical theory and practice. London: Sage.

Forero, R., McLellan, L., Rissel, C., & Bauman, A. (1999). Bullying behaviour and
psychosocial health among school children in New South Wales, Australia: Cross
sectional survey. British Medical Journal, 319, 344-348.

Foucault, M. (1973). The birth of the clinic: An archaeology of medical perception (1

ed.). New York: Pantheon Books.

Foucault, M. (1976). Histoire de la folie a l'age classique. Paris: Gallimard.

Foucault, M. (1977a). Discipline and punish : the birth of the prison. New York:
Pantheon Books.
 270

Foucault, M. (1977b). Power and Knowledge. New York: Random House.

Foucault, M. (1978). The History of Sexuality. New York: Pantheon Books.

Foucault, M. (Ed.). (1997). Power. Essential Works of Foucault 1954 – 1984. London:
Penguin.

Freedberg, S. (1989). Self-Determination: Historical perspectives and effects on current

practice. Social Work, 34, 33-38.

Freeman, M. D., Rossignol, A. M., & Hand, M. L. (2009). Applied forensic

epidemiology: The Bayesian evaluation of forensic evidence in vehicular
homicide investigation. Journal of Forensic and Legal Medicine, 16(2), 83-92.

Frey, K. S., Hirschstein, M. K., Snell, J. L., Van Schoiack Edstrom, L., MacKenzie, E. P.,
& Broderick, C. J. (2005). Reducing playground bullying and supporting beliefs:
An experimental trial of the Steps to Respect program. Developmental
Psychology, 41(3), 479-491.

Frost, J., & Jacobs, P. (1995). Play deprivation: A factor in juvenile violence. Dimensions
of Early Childhood, 23, 14-17.

Gamson, W. (1968). Power and discontent. Homewood,IL: The Dorsey Press.

Gamson, W. (1982). Encounters with unjust authority. Homewood, Illinois: The Dorsey
Press.

Garbarino, J., & Bedard, C. (2002). Parents under siege: Why you are the solution, not
the problem in your child’s life. New York Simon and Shuster.

Gaventa, J. (1980). Power and powerlessness: Quiessence and rebellion in an

Appalachian valley : . Chicago, Il: University of Illinois Press.

Gelman, S. A. (2003). The essentialist child: Origins of essentialism in everyday thought.

New York: Oxford University Press.
 271

Gelman, S. A., & Heyman, G. D. (1999). Carrot-eaters and creature-believers: The

effects of lexicalization on children's inferences about social categories.
Psychological Science, 10(6), 489-493.

George, J. M., & Gareth, J. M. (2000). The role of time in theory and theory building
Journal of Management, 26(4), 657-684.

Giddens, A. (1995). Politics, sociology, and social theory: Encounters with classical and
contemporary social thought. Stanford, CA: Stanford University Press.

Gifford-Smith, M., & Rabiner, D. L. (2004). Social information processing and children’s
social competence: A review of the literature. In J. B. Kupersmidt & K. A. Dodge
(Eds.), Children’s peer relations: From development to intervention to policy.
Washington, DC: American Psychological Association.

Gilbert, C. (1972). Community power structure: Propositional inventory, tests, and

theory. Gainesville, FL: University of Florida.

Giles, G. W. (2003). Children's essentialist beliefs about aggression. Developmental

Review, 23, 413-443.

Glass, G. V., McGaw, B., & Smith, M. L. (1981). Meta-analysis in social research.
Thousand Oaks: CA: Sage.

Glew, G. M., Fan, M.-Y., Katon, W., & Rivara, F. P. (2008). Bullying and school safety.
Journal of Pediatrics, 152, 123-128.

Glew, G. M., Fan, M., Katon, W., Rivara, F. P., & Kernic, M. A. (2005). Bullying,
psychosocial adjustment, and academic performance in elementary school.
Archives of Pediatrics and Adolescent Medicine, 159, 1026–1031.

Goldhagen, D. (1996). Hitler’s Willing Executioners: Ordinary Germans and the

Holocaust. New York: Knopf.

Goodman, R. (2001). Psychometric properties of the Strengths and Difficulties

Questionnaire. Journal of the American Academy of Child and Adolescent
Psychiatry, 40(11), 1337-1345.
 272

Gottgtroy, P., Kasabov, N., & MacDonell, S. (2003). Building evolving ontology maps
for data mining and knowledge discovery in biomedical informatics. Knowledge
Engineering and Discovery Institute, Auckland University of Technology.
Retrieved from
http://www.aut.ac.nz/resources/research/research_institutes/kedri/downloads/pdf/
gottgtroybsmtp.pdf

Gottman, J. M. (1994). What predicts divorce?: the relationship between marital

processes and marital outcomes. Hillsdale, NJ: Lawrence Erlbaum Associates.

Gourevitch, P. N. Y. (1999). We wish to inform you that tomorrow we will be killed with
our families: Stories from Rwanda. New York: Farar, Straus & Giroux.

Graham, S., & Juvonen, J. (1998). Self-blame and peer victimization in middle school:
An attributional analysis. Developmental Psychology, 34(3), 587-599.

Granoveter, M. S. (1973). The strength of weak ties. American Journal of Sociology,

78(6), 1360-1380.

Graves, C. (2004). Personal communication.

Grills, A. E., & Ollendick, T. H. (2002). Peer victimization, global self-worth, and
anxiety in middle school children. Journal of Clinical Child and Adolescent
Psychology, 31(1), 59-68.

Gruber, T. (1993). A translation approach to portable ontology specifications. Knowledge

Acquisition, 3(5), 199-220.

Guittierrez, L. (1990). Working with women of color: An empowerment perspective.

Social Work, 35(2), 149-153.

Guittierrez, L., & Ortega, A. (1991). Developing methods to empower Latinos: The
importance of groups. Social Work with Groups, 14(2), 23-43.

Gumpel, T. M., H. (2000). Children’s perceptions of school-based violence. British

Journal of Educational Psychology, 70, 391-404.
 273

Habermas, J. (1971). Knowledge and human interests (J. Shapiro, Trans.). Boston:
Beacon Press.

Habermas, J. (1979). Communication and the Evolution of Society (T. McCarthy, Trans.).
London: Heinemann.

Habermas, J. (1984). The theory of communicative action. Reason and the rationalization
of society (T. McCarthy, Trans. Vol. 1). Boston: Beacon.

Habermas, J. (1987). The theory of communicative action. Lifeworld and system: A

critique of functionalist reason (T. McCarthy, Trans. Vol. 2). Boston: Beacon.

Habermas, J. (1992). Political experience and the renewal of Marxist theory. In P. Dews
(Ed.), Autonomy and solidarity: Interviews with Jurgen Habermas. New York:
Verso.

Habermas, J. (1996). Between facts and norms: Contributions to a discourse theory of

law and democracy (W. Rehg, Trans.). Cambridge, MA: MIT Press.

Habermas, J. (2000). On the pragmatics of communication. Cambridge, MA: MIT Press.

Habermas, J. (2001a). On the pragmatics of social integration: Preliminary studies in the

theory of communicative action (B. Fulner, Trans.). Cambridge, MA: MIT.

Habermas, J. (2001b). Reflections on Communicative Pathology (1974) (B. Fulner,

Trans.). On the pragmatics of social integration: Preliminary studies in the
theory of communicative action. Cambridge, MA: MIT Press.

Habermas, J. (2001c). Universal pragmatics: reflections of a theory of communicative

competence (B. Fultner, Trans.). On the pragmatics of social interaction:
Preliminary studies in the theory of communicative action. Cambridge, MA: MIT
Press.

Hanish, L. D., & Guerra, N. G. (2004). Aggressive victims, passive victims, and bullies:
Developmental continuity or developmental change? Merrill-Palmer Quarterly,
50(1), 17-38.
 274

Hanish, L. D., & Guerra., N. G. (2000). Predictors of peer victimization among urban
youth. Social Development, 9(4), 521-543.

Harter, S. (1999). The construction of the self: A developmental perspective. New York
Guilford Press.

Hassard, J. (1991). Multiple paradigms and organizational analysis: A case study.

Organization Studies, 12(2), 275-299.

Hawker, D. S. J., & Boulton, M. J. (2000). Twenty years' research on peer victimization
and psychosocial maladjustment: A meta-analytic review of cross-sectional
studies. Journal of Child Psychology and Psychiatry, 41(4), 441-455.

Hawkins, M. (1997). Social Darwinism in European and American thought, 1860–1945:

Nature as model and nature as threat. Cambridge: Cambridge University Press.

Haynes, S. N., & Hayes O'Brien, W. (2000). Principles and practice of behavioral
assessment. New York: Kluwer/Plennum.

Haynie, D. L., Nansel, T., Eitel, P., Crump, A. D., Saylor, K., Yu, K., et al. (2001).
Bullies, victims and bully/victims: Distinct groups of at-risk youth. Journal of
Early Adolescence, 21(1), 29-49.

Hazler, R. J., Hoover, J. H., & Oliver, R. (1992). What kids say about bullying. The
Executive Educator, 20-22.

Healy, K. (2000). Critical Social Work: What is it? Critical Social Work, 2(1).

Heath, J. (2001). Communicative action and rational choice. Cambridge: MIT Press.

Hegar, R. L., & Yungman, J. J. (1989). Toward a causal typology of child neglect.
children and Youth Services Review, 11, 203-220.

Heinemann, P.-P. (1972.). Mobbning. Grupp'dld bland burn ocltyrrxna (cited in Pikas,
1975). Stockholm: Natur och kultur.

Hinde, R. A. (1976). Interactions, relationships and social structure Man, 2(1), 1-17.
 275

Hodges, E. V. E., Malone, M., & Perry, D. (1997). Individual risk and social risk as
interacting determinants of victimization in the peer group. Developmental
Psychology, 33, 94-101.

Hodges, E. V. E., & Perry, D. G. (1999). Personal and interpersonal antecedents and
consequences of victimization by peers. Journal of Personality and Social
Psychology, 76(4), 677-685.

Hoel, H., Rayner, C., & Cooper, C. L. (1999). Workplace bullying. International Review
of Industrial and Organisational Psychology, 14, 195-230.

Holt, M. K., Finkelhor, D., & Kaufman Kantor, G. (2007). Multiple victimization
experiences of urban elementary school students: Associatoins with psychosocial
functioning and academic performance. Child Abuse and Neglect, 31, 503-515.

Honneth, A. (1992). Integrity and Disrespect: Principles of a conception of morality

based on the theory of recognition. Political Theory, 20(2), 187-201.

Hopkins, W. G. (2001). A new view of statistics. Retrieved from

http://www.sportsci.org/resource/stats/index.html

Horkheimer, M. (1982). Critical theory. New York Seabury Press.

Horkheimer, M. (2003). The Rationalism Debate in Contemporary Philosophy. In C.

Theory (Ed.), Critical Theory. NY: Sage.

How, A. (2003). Critical Theory. New York: Palgrave: MacMillan.

Hunter, S. C., Boyle, J. M. E., & Warden, D. (2007). Perceptions and correlates of peer
victimization and bullying. British Journal of Educational Research, 77, 797-810.

Ialongo, N., Vaden-Kiernan, N., & Kellam, S. (1998). Early peer rejection and
aggression: Longitudinal relations with adolescent behavior. Journal of
Development and Physical Disabilities, 10(2).

Ife, J. (1997). Rethinking social work: Towards critical practice. Melbourne: Longman.
 276

Janssen, I., Craig, W. M., Boyce, W. F., & Pickett, W. (2004). Associations between
overweight and obesity with bullying behaviors in school-aged children.
Pediatrics, 113(5), 1187-1194.

Jay, M. (1973). The dialectical imagination: A history of the Frankfurt School and the
Institute of Social Research 1923-1950. Boston: Little, Brown & Company.

Jobe, L. E., & Williams White, S. (2007). Loneliness, social relationships, and a broader
autism phenotype in college students. Personality and Individual Difference,
42(8), 1479-1489.

Juvonen, J., Nishina, A., & Graham, S. (2000). Peer harassment, psychological
adjustment, and school functioning in early adolescence. Journal of Educational
Psychology, 92(2), 349-359.

Kaltiala-Heino, R., Rimpela, M., Marttunen, M., Rimpela, A., & Rantanen, P. (1999).
Bullying, depression, and suicidal ideation in Finnish adolescents. British Medical
Journal, 319, 348-351.

Kamenetsky, C. (1996). Children’s literature in Hitler’s Germany: The cultural policy of

National Socialism. Athens, Ohio: Ohio University Press.

Kaplan, A. (1964). The conduct of inquiry; methodology for behavioral science. San
Francisco: Chandler Pub. Co.

Kearins, K. (1998). Power in organisational analysis: Delineating and contrasting a

Foucauldian perspective. Electronic Journal of Radical Organizational Theory,
2(2),

Keifer, C. (1984). Citizen empowerment: A developmental perspective. Prevention in

Human Services, 3, 9-36.

Keltner, D., Capps, L., Kring, A. M., Young, R. C., & Heerey, E. A. (2001). Just teasing:
A conceptual analysis and empirical review. Psychological Bulletin, 127, 229-
248.

Keshvala, S. H. (2008). A test of a social capital model with economically disadvantaged

adolescents. University of Wisconsin.
 277

Keyes, C. L. M. (2004). Risk and resilience in human development. Research in Human

Development, 1(4), 223-227.

Kim, S. Y., & Leventhal, B. (2008). Bullying and suicide: A review. International
Journal of Adolescent Mental Health, 20(2), 133-154.

Kochenderfer-Ladd, B. J., & Skinner, K. (2002). Children's coping strategies: Moderators

of effects of peer victimization? Developmental Psychology, 38(2), 267-278.

Kochenderfer-Ladd, B. J., & Wardrop, J. L. (2001). Chronicity and instability of

children's peer victimization experiences as predictors of loneliness and social
satisfaction trajectories. 72, 1, 134-151.

Kochenderfer, B. J., & Ladd, G. W. (1996). Peer victimization: Cause or consequence of

school maladjustment? Child Development, 67, 1305-1317.

Kochenderfer, B. J., & Ladd, G. W. (1997). Victimized children’s responses to peers

aggression: Behaviors associated with reduced versus continued victimization.
Development and Psychopathology, 9, 59-73.

Kogan, G., & Chandan, U. (2004).

Kokko, K., & Pulkkinen, L. (2000). Aggression in childhood and long-term

unemployment in adulthood: A cycle of maladaption and some protective factors.
Developmental Psychology, 36, 463-472.

Kondrat, M. (1995). Concept, act, and interest in professional practice: Implication of an

empowerment perspective. Social Science Review, 69(3), 405-428.

Kovacs, M. (1992). Children’s Depression Inventory. North Tonawanda, NY: Multi-

Health Systems.

Kowalski, R. M., & Limber, S. P. (2007). Electronic bullying among middle school
students. Journal of Adolescent Health 41, S22-S30.

Kumpulainen, K., & Rasanen, E. (2000). Children involved in bullying at elementary

school age: their psychiatric symptoms and deviance in adolescence. An
epidemiological sample. Child Abuse and Neglect, 24(12), 1567-1577.
 278

Kumpulainen, K., Räsänen, E., Henttonen, I., Almqvist, F., Kresanov, K., Linna, S., et al.
(1998). Bullying and psychiatric symptoms among elementary school-age
children. Child Abuse and Neglect, 22, 705-717.

Kumpulainen, K., Rasanen, E., & Puura, K. (2001). Psychiatric disorders and the use of
mental health services among children involved in bullying. Aggressive Behavior,
27, 102-110.

Ladd, G. W., & Burgess, K. (2001). Do Relational Risks and Protective Factors Moderate
the Linkages Between Childhood Aggression and Early Psychological and School
Adjustment
Child Development 72(5), 1579-1601.

Ladd, G. W., Kochenderfer, B. J., & Coleman, C. C. (1997). Classroom peer acceptance,
friendship, and victimization: Distinct relational systems that contribute uniquely
to children's school adjustment? Child Development, 68(6), 1181-1197.

Ladd, G. W., & Profilet, S. M. (1996). The Child Behavior Scale: A teacher-report
measure of young children's aggression, withdrawn and pro-social behaviors.
Developmental Psychology, 32, 1008-1024.

Ladd, G. W., & Troop Gordon, W. (2003). The role of chronic peer difficulties in the
development of children’s psychological adjustment problems
Child Development, 74(5), 1344-1367.

Ladd, G. W., & Troop Gordon, W. (2005). Trajectories of peer victimization and
perceptions of the self and schoolmates: Precursors to internalizing and
externalizing problems. Child Development, 76(5), 1072 - 1091.

Lagerspetz, K. M. J., Bjorkqvist, K., & Peltonen, T. (1988). Is Indirect Aggression

Typical of Females?: Gender differences in aggressiveness in 11 – 12 year old
children. Aggressive Behavior, 14(403-414).

Lazarsfeld, P. F. (1937). Some remarks on the typological procedures in the social

research. . Studies in philosopy and social science, 6, 119-139.

Lazarsfeld, P. F., & Henry, N. W. (1968). Latent Structure Analysis. . Boston: Houghton
Mifflin.
 279

Lazarsfeld, P. F., & Rosenberg, M. (Eds.). (1955). The Language of Social Research: A
reader in the methodology of social research. New York: The Free Press.

Leary, T. (1956). A manual for the use of the interpersonal system of personality.
Berkeley, CA: Psychological Consultation Service.

Legkauskas, V., & Jakimaviciute, A. (2007). Relationship between involvement of

schoolchildren in bullying at school and experience of physical and emotional
abuse in the family. Social Sciences/Socialiniai Mokslai, 3(57).

Lehnert, M. (2007). Typologies in social inquiry. In T. Gschwend & F. Schimmelfennig

(Eds.), Research design in political science: How to practice what they preach.
New York: Palgrave MacMillan.

Lever, J. (1978). Sex differences in the complexity of children's play and games.
American Sociological Review, 43, 471-483.

Levy, B. (1990). Rethinking empowerment in v. n1 Journal of Progressive Human

Services 1.

Lewis, M. W., & Grimes, A. J. (1999). Metatriangulation: Building theory from multiple
paradigms. Metatriangulation: Building theory from multiple paradigms, 24(4),
672-690.

Li, Q. (2006). Cyberbullying in schools: A research of gender differences. School

Psychology International, 27(10).

Likert, R. (1932). A technique for the measurement of attitudes. Archives of Psychology,

140, 1-55.

Limber, S. P., & Small, M. A. (2003). State Laws and Policies to Address Bullying in
Schools. Child Development, 322.

Limper, R. (2000). Cooperation between parents, teachers and school boards to prevent
bullying in education: An overview of work done in the Netherlands. Aggressive
Behavior, 26, 124-134.
 280

Loeber, R., & Dishion, T. J. (1984). Boys who fight at home and school: Family
conditions influencing cross-setting consistency. journal of Consulting and
Clinical Psychology, 52, 759-768.

Long, J. D., & Pellegrini, A. D. (2003). Studying change in dominance and bullying with
linear mixed models. School Psychology International, 32(3), 401-417.

Lopez, C., & DuBois, D. L. (2005). Peer victimization and rejection: Investigation of an
integrative model of effects on emotional, behavioral, and academic adjustment in
early adolescence. Journal of Clinical Child and Adolescent Psychology, 34(1),
25-36.

Lowenstein, L. F. (1978). Who is the bully? Bulletin of the British Psychological Society,
31, 147-149.

Luhman, N. (1990). Political theory in the welfare state. New York: DeGreyter.

Lukacs, G. (1971). History and Class Consciousness: Studies in Marxist dialectics.

Cambridge MA: MIT Press.

Lukes, S. (1974). Power: A radical view London: MacMillan.

Lumsden, L. (2002). Preventing Bullying Available from

http://ericcass.uncg.edu/virtuallibrary/bullying/1068.html

Luthar, C., & Cicchetti, D. (2000). The construct of resilience: Implications for
interventions and social policy. Development and Psychopathology, 72, 857-885.

Lyons, L. C., & Morris, W. A. (2010). The meta-analysis calculator.

Macklem, G. L. (2003). Bullying and power: Social teasing in children's group. New
York: Kluwer/Plennum.

Macrae, C. N., & Bodenhausen, G. V. (2000). Social cognition: Thinking categorically

about others. Annual Review of Psychology, 51, 93-120.

March, J. S., Parker, J. D. A., Sullivan, K., Stallings, P., & Conners, C. K. (1997). The
Multidimensional Anxiety Scales for Children (MASC): Factor structure,
 281

reliability, and validity. Journal of American Academy of Child and Adolescent

Psychiatry, 36, 554–565.

Marcuse, H. (1974). Marxism and Feminism. Womem's Studies, 2.

Margolin, G., & Gordis, E. B. (2000). Effects of family and community violence on
children. Annual Review of Psychology, 51, 445-479.

Marx, K., & Engel, F. (1955). The communist manifesto with selections from the
Eighteenth Brumaire of Louis Bonaparte and Capital. Arlington Heights, Illinois:
Harlan Davidson.

McCarthy, T. (1984). Translators' note. The theory of communicative action. Reason and
the rationalization of society (Vol. 1). Boston: Beacon.

McDermott, F. E. (1975). Self-determination in social work. Boston: Routledge and

Kegan Paul.

McGrath, M. J. (2007). School bullying: Tools for avoiding harm and liability. Thousand
Oaks, CA: Sage

McGuire, W. J. (1997). Creative hypothesis generation in psychology: Some useful

heuristics. Annual Review of Psychology, 48, 1-20.

McKinney, J. C. (1966). Constructive Typology and Social Theory. New York: Appleton-
Century-Croft.

McKinney, J. C., & Kerckhoff, A. C. (1962). Toward a codificatoin of typological

procedure. Sociological Inquiry, 32(128-135).

Meehan, J. (1994). Autonomy, recognition and respect: Habermas, Benjamin and

Honneth. Constellations, 1, 270-285.

Metropolitan Area Child Study Group. (2002). A cognitive-ecological approach to

preventing aggression in urban settings: Initial outcomes for high-risk children.
Journal of Consulting and Clinical Psychology, 70(1), 179-194. Journal of
Consulting and Clinical Psychology, 70(179-194).
 282

Milgram, S. (1974). Obedience to Authority: An experimental view. New York: Harper

and Row.

Miller, J., Martin, I., & Schamess, G. (2003). School violence and children in crisis:
Community and school interventions for social workers and counselors. Denver:
Love Publising.

Miller, J., & Vaillancourt, T. (2007). The relation between childhood peer victimization
and adult perfectionism: Are victims of indirect aggression more perfectionistic?
Aggressive Behavior, 33, 1-12.

Minnesota Department of Children Families & Learning. (2001). Minnesota Student

Survey: Key trends through 2001.

Mishna, F. (2003). Peer victimization: The case for social work intervention. Families in
Society, 84(4), 512-520.

Mishna, F. (2007). Personal communication.

Mohr, A. (2006). Family variables associated with peer victimization: Does family
violence enhance the probability of being victimized by peers? Swiss Journal of
Psychology, 65(2), 107-116.

Moran, S., Smith, P. K., Thompson, D., & Whitney, I. (1993). Ethnic differences in
experience of bullying: Asian and white children. British Journal of Educational
Psychology, 63, 431-440.

Moreau, M. (1990). Empowerment through advocacy and consciousness-raising:

Implications of a structural approach to social work. Journal of Sociology and
Social Welfare, 17(2), 41-52.

Morrison, B. (2006). School bullying and restorative justice: Toward a theoretical

understanding of the role of respect, pride, and shame. Journal of Social Issues,
62(2), 371-392.

Morrow, M. T., Hubbard, J. A., Rubin, R. M., & McAuliffe, M. D. (2008). The relation
between childhood aggression and depressive symptoms. Merrill-Palmer
Quarterly, 54(3), 316-340.
 283

Mouttapa, M., Valente, T., Gallaher, P., Rohrbach, L. A., & Unger, J. B. (2004). Social
network predictors of bullying and victimization. Adolescence, 39(154), 315-335.

Mullaly, B. (1997). Structural social work: Ideology, theory, and practice. Don Mills,
Ontario: Oxford University Press.

Mynard, H., Joseph, S., & Alexander, J. (2000). Peer-victimisation and posttraumatic
stress in adolescents. Personality and Individual Differences, 29, 815-821.

Mynard, H., Lawrence, H., & Joseph, S. (2000). Development of the multidimensional
peer-victimization scale. Aggressive Behavior
26, 169-178.

Mytton, J., DiGuiseppi, C., Gough, D., Taylor, R., & Logan, S. (2006). School-based
secondary prevention programmes for preventing violence. Cochrane Database of
Systematic Reviews, 3,

Nansel, T. R., Overpeck, M., Pilla, R. S., Ruan, W. J., Simons-Morton, B., & Scheidt, P.
(2001). Bullying behaviors among U.S. youth: Prevalence and association with
psychosocial adjustment. Journal of the American Medical Association, 285(16),
2094-2100.

Nansel, T. R., Overpeck, M. D., Haynie, D. L., Ruan, W. J., & Scheidt, P. C. (2003).
Relationships between bullying and violence among U.S. youth. Archives of
Pediatric Adolescent Medicine, 157, 348-353.

NASW. (2011). School Social Worker Crisis.

Neary, A., & Joseph, S. (1994). Peer victimization and its relationship to self-concept and
depression among Australian schoolgirls. Personality and Individual Difference,
16, 183-186.

Newman, K., Fox, C., Harding, D., Mehta, J., & Roth, W. (2004). Rampage: The social
roots of school shootings. New York: Basic Books.

Nishina, A., Juvonen, J., & Witkow, M. R. (2005). Sticks and stones may break my
bones, but names will make me feel sick: The psychosocial, somatic, and
scholastic consequences of peer harassment. Journal of Clinical Child and
Adolescent Psychology, 34(1), 37-48.
 284

Noguera, P. A. (1995). Preventing and producing violence: A critical analysis of

responses to school violence. Harvard Educational Review, 65(2), 189-212.

Nylund, K., Nishina, A., Bellmore, A., & Graham, S. (2007). Subtypes, severity, and
structural stability of peer victimization: What does latent class analysis say?
Child Development, 78(6), 1706-1722.

O’Connell, P., Pepler, D. J., & Craig, W. M. (1999). Peer involvement in bullying:
Insights and challenges for intervention. Journal of Adolescence, 22, 437-452.

O’Moore, M., & Kirkham, C. (2001). Self-esteem and its relationship to bullying
behavior. Aggressive Behavior, 27, 269–283.

Office of United States Chief of Counsel. (1946). Nazi conspiracy and aggression /Office
of United States Chief of Counsel for Prosecution of Axis Criminality.

Olweus, D. (1973). Personality and aggression. In J. K. Coie & D. D. Jensen (Eds.),

Nebraska symposium on motivation (pp. 261-321). Lincoln, NE: University of
Nebraska Press.

Olweus, D. (1978). Aggression in the schools: Bullies and whipping boys. New York:
Wiley.

Olweus, D. (1980). Familial and temperamental determinants of aggressive behavior in

adolescent boys: A causal analysis. Developmental Psychology, 16(6), 644-660.

Olweus, D. (1993). Victimization by peers: Antecedents and long-term outcomes. In K.

H. Rubin & J. B. Asendorf (Eds.), Social withdrawal, inhibition, and shyness.
Hillsdale, NJ: Erlbaum.

Olweus, D. (1996). The Revised Olweus Bully/Victim Questionnaire. Bergen, Norway:

Research Center for Health
Promotion (HEMIL Center).

Omigbodun, O., Dogra, N., Esan, O., & Adedokun, B. (2008). Prevalence and correlates
of suicidal behaviour among adolescents in southwest Nigeria. International
Journal of Social Psychiatry, 54(1), 34-46.
 285

Ottman, H. (1982). Cognitive interests and self-reflection. In J. B. Thompson & D. Held

(Eds.), Habermas: Critical debates (pp. 79-97). London: Macmillan.

Owens, L., Shute, R., & Slee, P. (2000). "Guess what I just heard!": Indirect aggression
among teenage girls in Australia. Aggressive Behavior, 26, 67-83.

Owens, L., Slee, P., & Shute, R. (2000). "It hurts a hell of a lot...": The effects of indirect
aggression on teenage girls. School Psychology International, 214(359-376.).

Painter, J. E., Borba, C. P. C., Hynes, M., Mays, D., & Glanz, K. (2008). The use of
theory in health behavior research from 2000 to 2005: A systematic review.
Annals of Behavioral Medicine, 35, 358-362.

Paquette, J. A., & Underwood, M. K. (1999). Gender differences in young adolescents'

experiences of peer victimization: Social and physical aggression. Merrill-Palmer
Quarterly, 45(2), 242-266.

Parker, G., Wilhelm, K., & Asghari, A. (1997). Early onset depression: the relevance of
anxiety Journal Social Psychiatry and Psychiatric Epidemiology, 32(1), 30-37.

Parsons, T. (1969). Politics and social structure. Toronto: MacMillan.

Peets, K., Hodges, E. V. E., Kikas, E., & Salmivalli, C. (2007). Hostile attributions and
behavioral strategies in children: Does relationship type matter? Developmental
Psychology, 43, 889-900.

Pellegrini, A. D. (1996). Observing children in their natural worlds: A methodological

primer. Mahwah, New Jersey: Lawrence Erlbaum.

Pellegrini, A. D., & Bartini, M. (2000). A longitudinal study of bullying, victimizatioin,

and peer affiliation during the transistion from primary school to middle school.
American Educational Research Journal, 37(3), 699-725.

Pellegrini, A. D., Bartini, M., & Brooks, F. (1999). School bullies, victims, and
aggressive victims. Factors relating to group affiliation and victimization in early
adolescence. Journal of Educational Psychology, 91(2), 216–224.
 286

Pellegrini, A. D., Blatchford, P., Kato, K., & Baines, E. (2004). A short-term longitudinal
study of children's playground games in primary school: Implications for
adjustment to school and social adjustment in the USA and the UK. Social
Development, 13(1), 107-123.

Pellegrini, A. D., & Smith, P. K. (1998). Physical activity play: The nature and funciton
of a neglected aspect of play. Child Development, 69(3), 577-598.

Pepler, D. J., & Craig, W. M. (1995). A peek behind the fence: naturalistic observations
of aggressive children with remote audiovisual recording. Developmental
Psychology, 314, 548-553.

Pepler, D. J., Craig, W. M., Ziegler, S., & Charach, A. (1994). An evaluation of an anti-
bullying intervention in Toronto schools. Canadian Journal of Community Mental
Health, 13(2), 95-110.

Percy-Smith, B., & Matthews, H. (2001). Tyrannical Spaces: young people, bullying and
urban neighbourhoods. Local Environment: The International Journal of Justice
and Sustainability, 6(1), 49 - 63.

Perren, S., & Hornung, R. (2005). Bullying and delinquency in adolescence: Victims’ and
perpetrators’ family and peer relations. Swiss Journal of Psychology 64(1).

Perry, D., Williard, J., & Perry, L. (1990). Peers’ perceptions of the consequences that
victimized children provide peer aggressors. Child Development, 61, 1310-1325.

Peskin, M. F. (2007). Bullying and victimization and internalizing symptoms among low-
income Black and Hispanic students. Journal of Adolescent Health, 40, 372-372.

Petersen, L., & Rigby, K. (1999). Countering bullying at an Australian secondary school.
Journal of Adolescence, 22(4), 481-492.

Petticrew, M., & Roberts, H. (2006). Systematic reviews in the social sciences: A
practical guide. Oxford: Blackwell Publishing.

Pfeffer, J. (1997). New directions for organizationaltTheory: Problems and prospects.

New York: Oxford University Press.
 287

Pikas, A. (1975). A treatment of mobbing in schools: Principles for and the results of the
work of an anti-mobbing group. Scandinavian Journal of Educational Research,
19, 1-12.

Poteat, D. L., & Espelage, D. L. (2007). Predicting psychosocial consequences of

homophobic victimization in Middle School students. Journal of Early
Adolescence, 27(2), 175-191.

Price, R. (2007). The way things look. Oxford: Oxford University Press.

Rappaport, J. (1987). Terms of empowerment: exemplars of prevention: Toward a theory

for community psychology. American Journal of Community Psychology, 15(2),
121-145.

Rasmussen, D. M., & Swindal, J. (Eds.). (2002). Jurgen Habermas (Vol. 2). Thousand
Oaks, CA: Sage.

Rasmussen, D. M., & Swindal, J. (Eds.). (2002). Jurgen Habermas (Vol. 1). Thousand
Oaks, CA: Sage.

Reynolds, W. (1988). Suicidal Ideation Questionnaire. Odessa, FL: Psychological

Assessment Resources.

Rigby, K. (1998). Suicidal ideation and bullying among Australian secondary school
children. Australian Educational and Developmental Psychologist, 15(1).

Rigby, K. (1999). Peer victimization at school and the health of secondary school
students. British Journal of Educational Psychology, 69, 95-104.

Rigby, K. (2000). Effects of peer victimization in schools and perceived social support on
adolescent well-being. Journal of Adolescence, 23, 57-68.

Rigby, K. (2002a). A meta-evaluation of methods and approaches to reducing bullying in

pre-schools andin early primary school in Australia. Canberra, Australia: Crime
Prevention Branch, Attorney-General's Department.

Rigby, K. (2002b). New perspectives on bullying. London: Jessica Kingsley Publishers.

 288

Rigby, K. (2003). Addressing bullying in schools: Theory and practice. Trends and Issues
in Criminal Justice, 259, 1-6.

Rigby, K., & Slee, P. (2001). Dimensions of interpersonal relation among Australian
children and implications for psychological well-being. Journal of Social
Psychology, 133(1), 33-42.

Ritzer, G., & Smart, B. (Eds.). (2001). Handbook of social theory. . Thousand Oaks, CA:
Sage Publications.

Rivers, I., Duncan, N., & Besag, V. E. (2007). Bullying: A handbook for educators and
parents. Westport, CN: Greenwood/Prager.

Rodkin, P., & Hodges, E. V. E. (2003). Bullies and victims in the peer ecology: Four
questions for psychologists and school professionals. School Psychology Review,
32(2), 384-400.

Rodkin, P. C., Farmer, T. W., Pearl, R., & Van Acker, R. (2000). Heterogeneity of
popular boys: Antisocial and prosocial configurations. Developmental
Psychology, 36(1), 14-24.

Roland, E. (2000). Personal Correspondence (eMail correspondance ed.).

Roland, E. (2002). Aggression, depression, and bullying others. Aggressive Behavior,

28(3), 198-206.

Roland, E. (2002). Bullying, depressive symptoms and suicidal thoughts. Educational

Research, 44(1), 55-67.

Rothstein, H., Sutton, A., & Borenstein, M. (Eds.). (2005). Publication bias in meta-
analysis: Prevention, assessment and adjustments. Hoboken, NJ: John Wiley &
Sons.

Rubin, K. H., Fein, G., & Vandenberg, B. (1983). Play. In E. M. Hetheringon (Ed.),
Handbook of child psychology: Socialization, personality and social development
(Vol. 4). New York: Wiley.
 289

Rubin, K. H., Maioni, T. L., & Hornung, M. (1976). Free play behaviors in middle and
lower-class preschoolers: Parten and Piaget revisited. Child Development, 47,
414-419.

Ruff, C. (2003). Growth in bone strength, body size, and muscle size in a juvenile
longitudinal sample. Bone, 33, 317-329.

Saalman, H. (1971). Haussmann: Paris transformed. New York: George Braziller.

Safilios-Rothschild, C. (1970). The study of family power structure: A review 1960-

1969. Journal of Marriage and the Family, 32, 539-552.

Saleebey, D. (1996). The strengths perspective in social work: Extensional cautions.

Social Work, 41(3), 295-305.

Salmivalli, C. (1999). Participant role approach to school bullying: Implications for

interventions. Journal of Adolescence, 22, 453-459.

Salmivalli, C., & Helteenvuori, T. (2007). Reactive, but not proactive aggression predicts
victimization among boys. Aggressive Behavior, 33(3), 198-206.

Salmivalli, C., & Lagerspetz, K. M. J. (1998). Stability and change of behavior in

connection with bullying in schools. Aggressive Behavior, 24, 205-218.

Salmivalli, C., Lagerspetz, K. M. J., Björkqvist, K., Österman, K., & Kaukiainen, A.
(1996). Bullying as a group process: Participant roles and their relations to social
status within the group. Aggressive Behavior, 22, 1-15.

Salmon, G. M., James, A., & Smith, D. M. (1998). Bullying in schools: Self-reported
anxiety, depression and self-esteem in secondary school children. British Medical
Journal, 317, 924-925.

SAMSHA. (2003). President's New Freedom Commission on Mental Health: Achieving

the promise: Transforming mental health care in America Retrieved from
http://www.mentalhealthcommission.gov/reports/FinalReport/toc.html+President
%E2%80%99s+New+Freedom+Commission+on+Mental+Health+(2003).+Achie
ving+the+promise:+Transforming+mental+health+care+in+America&cd=1&hl=e
n&ct=clnk&gl=us.
 290

Scaglione, J., & Scaglione, A. R. (2006). Bully-proofing children: A hands-on guide to

stop bullying. Lanham, MD: Rowman and Littlefield.

Schmidt, M. E., & Bagwell, C. L. (2007). The Protective Role of Friendships in Overtly
and Relationally Victimized Boys and Girls. Merrill-Palmer Quarterly, 53(3),
439-460.

Schwartz, D., Dodge, K., H., & Coie, J. (1993). The emergence of chronic peer
victimization in boys’ play groups. Child Development, 64(6), 1755-1772.

Schwartz, D., Farver, J. M., Chang, L., & Lee-Shim, Y. (2002). Victimization in South
Korean children's peer groups. Journal of Abnormal Child Psychology, 30, 113-
125.

Schwartz, D., Gorman, A., Duong, M., & Nakamoto, J. (2008). Peer relationships and
academic achievement as interacting predictors of depressive symptoms during
middle childhood. Journal of Abnormal Psychology, 117(2), 289-299.

Schwartz, D., Gorman, A. H., Dodge, K., A., Petit, G. S., & Bates, J. E. (2008).
Friendships with peers who are low or high in aggression as moderators of the
link between peer victimization and declines in academic functioning. Journal of
Abnormal Child Psychology, 36(5), 719-730.

Schwartz, D., Hopmeyer Gorman, A., Nakamoto, J., & Toblin, R. L. (2005). Victimizatin
in the peer group and children's academic functioning. Journal of Educational
Psychology, 97(3), 425-435.

Schwartz, D., & Proctor, L. (2000). Community violence exposure and children's social
adjustment in the school peer group: The mediating roles of emotion regulation
and social cognition. Journal of Consulting and Clinical Psychology, 68, 670-683.

Schwartz, D., Proctor, L., & Chien, D. (2001). The aggressive victim of bullying:
Emotional and behavioral dysregulation as a pathway to victimization by peers In
J. Juvonen & S. Graham (Eds.), School-based peer harassment: The plight of the
vulnerable and victimized. Boston, MA: Guilford Press.

Seals, D., & Young, J. (2003). Bullying and victimization: Prevalence and relationship to
gender, grade level, ethnicity, self-esteem and depression. Adolescence, 38(152),
735-747.
 291

Search Institute. (2007). What kids need: Developmental assets. Retrieved July, 2004,
from http://www.search-institute.org/developmental-assets

Segrin, C. (2000). Social skills deficits associated with depression. Clinical Psychology
Review, 20(3), 379-403.

Selman, R., & Demorest, A. (1984). Observing troubled children’s interpersonal

negotiation strategies: Implications of and for a developmental model. Child
Development, 55, 288-304.

Shaffer, D., Fisher, P., Lucas, C. P., Dulcan, M. K., & Schwab-Stone, M. E. (2000).
NIMH Diagnostic Interview Schedule for Children Version IV: Description,
differences from previous versions, and reliability of some common diagnoses.
Journal of the American Academy of Child and Adolescent Psychiatry, 39, 28-38.

Shapiro, D. (1981). Autonomy and rigid character. New York: Beacon Books.

Shapiro, D. (2000). Dynamics of character: Self-regulation in psychopathology. New

York: Basic Books.

Sharp, S., Thompson, D., Arora, C., & Cho, S. (2000). How long before it hurts? An
investigation into long-term bullying. School Psychology International, 21(1), 37-
46.

Shellard, E., & Turner, J. R. (2004). Safe and Secure Schools. ERS Focus On . Arlington,
VA: Educational Research Service.

Simmons, R. (2000). Odd girl out: The hidden culture of aggression in girls. New York:
Harcourt.

Skinner, B. F. (1969). Contingencies of reinforcement: a theoretical analysis. New York:

. New York: Appleton-Century-Crofts.

Slavin, R. E. (1995). Best evidence synthesis: An intelligent alternative to meta-analysis.

Journal of Clinical Epidemiology, 48(1), 9-18.

Slee, P. (1994). Situational and interpersonal correlates of anxiety associated with peer
victimisation. Child Psychiatry and Human Development, 25(2), 97–107.
 292

Smetana, J. G. (1995). Parenting styles and conceptions of parental authority during

adolescence. Child Development, 66, 299-316.

Smith, J. D., Schneider, B. H., Smith, P. K., & Ananiadou, K. (2004). The effectiveness
of whole-school anti-bullying programs: A synthesis of evaluation research.
School Psychology International, 33(4), 547-560.

Smith, J. D., Stewart, R., & Cousins, J. B. (2003). Anti-bullying interventions: A survey
of Ontario schools. The CAP Journal, 12(1), 9-11.

Smith, P. K. (1997). Bullying in Life-Span Perspective: What can studies of school

bullying and workplace bullying learn from each other? Journal of Community
and Applied Social Psychology, 7, 249-255.

Smith, P. K., & Boulton, M. J. (1990). Rough and tumble play, aggression and
dominance: Perception and behavior in children’s encounters. Human
Development, 33, 271-282.

Smith, P. K., & Brain, P. (2000). Bullying in schools: Lessons from two decades of
research. Aggressive Behavior, 26, 1-9.

Smith, P. K., Cowie, H., Olafsson, R. F., & Liefooghe, A. P. D. (2002). Definitions of
bullying: A comparison of terms used, and age and gender differences, in a
fourteen-country international comparison. Child Development, 73(4), 1119–
1133.

Smith, P. K., & Levan, S. (1995). Perceptions and experiences of bullying in younger
pupils. British Journal of Educational Psychology, 65(4), 489-500.

Smith, P. K., Madsen, K., & Moody, J. (1999). What causes the age decline in reports of
being bullied at school? Towards a developmental analysis of risks of being
bullied. Educational Research, 41, 267-285.

Smith, P. K., Morita, Y., Junger-Tas, J., Olweus, D., Catalano, R., & Slee, P. (1999). The
nature of school bullying: A cross national perspective. London: Routledge.

Smith, P. K., Pepler, D. J., & Rigby, K. (Eds.). (2004). Bullying in schools: How
successful can interventions be? Cambridge: Cambridge University Press.
 293

Smith, P. K., & Sharp, S. (1994). School bullying: Insights and perspectives. London:
Routledge.

Smith, P. K., Talamelli, L., Cowie, H., Naylor, P., & Chauhan, P. (2004). Profiles of non-
victims, escaped victims and new victims of school bullying. British Journal of
Educational Psychology, 74, 565-581.

Sourander, A., Jensen, P., Ronning, J. A., Niemala, S., Helenius, H., Sillanmaki, L., et al.
(2007). What Is the early adulthood outcome of boys who bully or are bullied in
childhood? The Finnish "From a Boy to a Man" study. Pediatrics Vol. 120 No. 2
August 2007, pp. 397-404, 120(2), 397-404.

Spinrad, T. L., Eisenberg, N., Harris, E., Hanish, L. D., Fabes, R. A., Kupanoff, K., et al.
(2004). The relation of childrens everyday nonsocial peer play behavior to their
emotionality, regulation and social functioning. Developmental Psychology,
40(1), 67-80.

Spriggs, A. L., Iannotti, R. J., Nansel, T. R., & Haynie, D. L. (2007). Adolescent bullying
involvement and perceived family, peer and school relations: Commonalities and
differences across race/ethnicity. Journal of Adolescent Health, 41(3), 283-293.

Srabstein, J. C., McCarter, R. J., Cheng, S., & Huang, Z. J. (2006). Morbidities associated
with bullying behaviors in adolescents: School-based study of American
adolescents. International Journal of Adolescent Mental Health, 18(4), 587-596.

Stack, C. B. (1975). All our kin: Strategies for survival in a Black community. New York:
Harper & Row.

Stanley, L., & Arora, C. (1998). Social exclusion amongst girls: Their self-esteem and
coping strategies. Educational Psychologist, 14(2), 94-100.

Stassen Berger, K. (2007). Update on bullying at school: Science forgotten?

Developmental Review, 27, 90-126.

Steinberg, L. D., Lamborn, S. D., Darling, N., Mounts, N. S., & Dornbusch, S. M. (1994).
Over-time changes in adjustment and competence among adolescents from
authoritative, authoritarian, indulgent, and neglectful families. Child
Development, 65, 754-770.
 294

Stevens, V., De Bourdeaudhuij, I., & Van Oost, P. (2000). Bullying in Flemish schools:
An evaluation of anti-bullying intervention in primary and secondary schools.
British Journal of Educational Psychology, 2000(70), 195-210.

Storch, E. A., & Esposito, L. E. (2003). Peer victimization and posttaumatic stres among
children. Child Study Journal 33(2), 91-96.

Storch, E. A., Roth, D. A., Coles, M. E., Heimberg, R. G., Bravata, E. A., & Moser, J.
(2003). The measurement and impact of childhood teasing in a sample of young
adults. Anxiety Disorders, 18, 681-694.

Straus, M., & Gelles, R. (1979). Measuring intrafamily conflict and violence: The
Conflict Tactic Scale. Journal of Marriage and the Family, 41, 75-85.

Straus, M., & Gelles, R. (Eds.). (1990). Measuring intrafamily conflict and violence: The
Conflict Tactic Scale. New York: Transaction Publishers.

Strom, P. S., & Strom, R. D. (2005). When teens turn cyberbullies. Education Digest,
71(4), 35-41.

Sunde Peterson, J. (2006). Bullying and the gifted: Victims, perpetrators, prevalence, and
effects. Gifted Child Quarterly, 50(2), 148-168.

Sutton, A., Smith, P. K., & Swettenham, J. (1999). Bullying and ―theory of mind‖: A
critique of the ―social skills deficit‖ view of anti-social behaviour. Social
Development, 8(1), 117-134.

Sutton, J., & Smith, P. K. (1999). Bullying as a group process: An adaptation of the
Participant Role approach. Aggressive Behavior, 25, 97-111.

Swearer, S. M., & Espelage, D. L. (2004). Introduction: A social-ecological framework

of bullying among youth. In D. L. Espelage & S. M. Swearer (Eds.), Bullying in
American schools: A socia-ecological perspective on prevention and intervention.
Mahwah, NJ: Lawrence Erlbaum.

Trochim, W. M. K. (1989). An introduction to concept mapping for planning and

evaluation. Evaluation and Program Planning, 12(1), 1-16.
 295

Twemlow, S. W. (2004). Preventing violence in schools. Psychiatric Times, 21(4).

Twemlow, S. W., Fonagy, P., Sacco, F. C., & Barbour, J. R. J. (2006). Teachers who
bully students: A hidden trauma. International Journal of Social Psychiatry,
52(3), 187-198.

Twemlow, S. W., Fonagy, P., Sacco, F. C., Gies, M. L., Evans, R., & Ewbank, R. (2001).
Creating a peaceful school learning environment: A controlled study of an
elementary school intervention to reduce violence. American Journal of
Psychiatry, 158(5), 808–810.

Van de Ven, A. H. (1989). Nothing Is Quite So Practical as a Good Theory. Academy of

Management Review, 14(486-489).

van Hoof, A., Raaijmakers, Q. A. W., van Beel, Y., Hale, W. W., & Aleva, L. (2008). A
multi-mediation model on the relations of bullying, victimization, identity, and
family with adolescent depressive symptoms. Journal of Youth and Adolescence,
37, 772-782.

van Hoof, A., & Raajimakers, Q. A. W. (2003). The search for the structure of identity
formation. Identity: An international journal of theory and research, 3(3), 271-
289.

Vanerkolk, B. (1994). The body keeps the score: Memory and the evolving
psychobiology of posttraumatic stress. Harvard Review of Psychiatry, 1, 253-265.

Viswevaran, C., & Ones, D. S. (1995). Theory testing: Combining meta-analysis and
structural equations modeling. Personnel Psychology, 28(4), 865-885.

von Friedeburg, L. (2007). History of the Institute for Social Research. Retrieved from
http://www.ifs.uni-frankfurt.de/english/history.htm

Vossekuil, B., Fein, R. A., Reddy, M., Borum, R., & Modzeleski, W. (2002). The final
report and findings of the Safe School Initiative: Implications for the prevention
of school attacks in the United States.

Vreeman, R. C., & Carroll, A. E. (2007). A systematic review of school-based

interventions to prevent bullying. Archives of Pediatrics and Adolescent
Medicine, 16(1), 78-88.
 296

Warwick, I., Chase, E., Aggleton, P., & Sanders, S. (2004). Homophobia, sexual
orientation and schools: A review and implications for action.

Weber, M. (Ed.). (1947). The theory of social and economic organization. New York:
Oxford University Press.

Weick, K. E. (1989). Theory construction as disciplined imagination. Academy of

Management Review, 14(4), 516-531.

Wentzel, K., & Asher., S. (1995). The academic lives of neglected, rejected, popular and
controversial children. Child Development, 66, 754-763.

White, S. (1995). The Cambridge companion to Habermas. Cambridge: Cambridge

University Press.

WHO. (2009). Health behavior in school-aged children Retrieved from

http://www.hbsc.org/

Wiggins, J. S., & Wender, C. L. (1961). The Peer Nomination Inventory: An empirically-
derived sociometric measure of adjustment in preadolescent boy. Child
Development, 64, 643-677.

Williams, K. R., & Guerra, N. G. (2007). Prevalence and predictors of internet bullying.
Atlanta, GA: Centers for Disease Control and Prevention.

Wolke, D., Woods, S., Bloomfield, L., & Karstadt, L. (2000). The association between
direct and relational bullying and behavior problems among primary school
children. Journal of Child Psychology and Psychiatry, 41(8), 989-1002.

Woods, S., & Wolke, D. (2004). Direct and relational bullying among primary school
children and academic achievement. Journal of School Psychology, 42(2), 135–
155.

Yang, B. (2002). Meta-analysis research and theory building. Advances in Developing

Human Resources, 4(3), 296-316.

Yitzhak, A. (1987). Belzec, Sobibor, Treblinka: The Operation Reinhard Death Camps.
Bloomington and Indianapolis, IN: Indiana University Press.
 297

Young, R., & Sweeting, H. (2004). Adolescent bullying, relationships, psychological

well-being, and gender-atypical behavior: A gender diagnosticity approach. Sex
Roles, 50(7/8), 525-537.

Zimbardo, P. G. (1972). Stanford prison experiment: a simulation study of the psychology

of imprisonment. Stanford, CA: Philip G. Zimbardo, Inc.